archive/blender-archive
Archived
1
1
Fork 0
Code Pull Requests Packages Activity
This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
a262ea8c47f8106f66e16935c556f383ef306861
blender-archive/source/blender/blenkernel/intern/pointcache.c

4103 lines
115 KiB
C
Raw Normal View History

doxygen: prevent GPL license block from being parsed as doxygen comment.
2011-02-23 10:52:22 +00:00
/*
Patch to change license to GPL only, from GSR.
2008-01-07 19:13:47 +00:00
* ***** BEGIN GPL LICENSE BLOCK *****
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
Patch to change license to GPL only, from GSR.
2008-01-07 19:13:47 +00:00
* of the License, or (at your option) any later version.
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
correct fsf address
2010-02-12 13:34:04 +00:00
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
*
* The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
* All rights reserved.
*
Clean up headers a bit more.
2011-02-23 18:03:40 +00:00
* Contributor(s): Campbell Barton <ideasman42@gmail.com>
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
*
Patch to change license to GPL only, from GSR.
2008-01-07 19:13:47 +00:00
* ***** END GPL LICENSE BLOCK *****
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
*/
doxygen: blender/blenkernel tagged.
2011-02-27 20:40:57 +00:00
/** \file blender/blenkernel/intern/pointcache.c
* \ingroup bke
*/
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <sys/stat.h>
#include <sys/types.h>
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "MEM_guardedalloc.h"
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "DNA_ID.h"
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
#include "DNA_dynamicpaint_types.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "DNA_modifier_types.h"
#include "DNA_object_types.h"
Cleanup: add _types.h suffix to DNA headers
2018-02-07 11:14:08 +11:00
#include "DNA_object_force_types.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "DNA_particle_types.h"
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
#include "DNA_rigidbody_types.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "DNA_scene_types.h"
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
#include "DNA_smoke_types.h"
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
#include "BLI_blenlib.h"
Smoke: Patch #20955 / fix by Damien Plisson (damien78) * Smoke got it's own thread loop now and can be esc'ed now * Apple OpenMP bug workaround until they fix the GCC (couldn't test this one) * removing some "static" keyword to prevent crashing on frame display * should also fix problems with wavelet.noise crashing (Bug #20710) Fluid: * sleep delay has been reduced to be more responsive * Fluid progress is displayed in percent using the mouse cursor (like smoke)
2010-02-07 14:25:09 +00:00
#include "BLI_threads.h"
header re-shuffle, some headers contained unneeded refereces to other headers, better include inline with the C files that need them
2010-08-10 05:41:51 +00:00
#include "BLI_math.h"
Add number and memory size formatting throughout the UI This commit adds number formatting (thousands separator) to the baking panel. It also adds a new function to format memory sizes (KB/GB/etc) and applies it to the baking panel and scene stats. The new function is unit tested. Reviewers: Severin Tags: #user_interface Differential Revision: https://developer.blender.org/D1248
2018-05-25 22:17:15 +02:00
#include "BLI_string.h"
split BKE_utildefines.h, now it only has blender specific defines like GS() MAKE_ID, FILE_MAXDIR, moved the generic defines to BLI_utildefines.h. no functional changes.
2011-01-07 18:36:47 +00:00
#include "BLI_utildefines.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
Refactor translation code out of blenfont - Add blentranslation `BLT_*` module. - moved & split `BLF_translation.h` into (`BLT_translation.h`, `BLT_lang.h`). - moved `BLF_*_unifont` functions from `blf_translation.c` to new source file `blf_font_i18n.c`.
2015-08-16 17:32:01 +10:00
#include "BLT_translation.h"
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
Smoke: Patch #20955 / fix by Damien Plisson (damien78) * Smoke got it's own thread loop now and can be esc'ed now * Apple OpenMP bug workaround until they fix the GCC (couldn't test this one) * removing some "static" keyword to prevent crashing on frame display * should also fix problems with wavelet.noise crashing (Bug #20710) Fluid: * sleep delay has been reduced to be more responsive * Fluid progress is displayed in percent using the mouse cursor (like smoke)
2010-02-07 14:25:09 +00:00
#include "PIL_time.h"
Refactor: BLI_path_util (split out app directory access) This module is intended for path manipulation functions but had utility functions added to access various directories.
2014-11-23 14:37:13 +01:00
#include "BKE_appdir.h"
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
#include "BKE_anim.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "BKE_cloth.h"
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
#include "BKE_dynamicpaint.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "BKE_global.h"
#include "BKE_main.h"
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
#include "BKE_modifier.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "BKE_object.h"
#include "BKE_particle.h"
#include "BKE_pointcache.h"
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
#include "BKE_scene.h"
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
#include "BKE_smoke.h"
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
#include "BKE_softbody.h"
remove references to BKE_utildefines where its not needed. - move GS() define into DNA_ID.h - add BLI_utildefines as an automatic include with makesrna generated files.
2011-01-07 19:18:31 +00:00
Merge of itasc branch. Project files, scons and cmake should be working. Makefile updated but not tested. Comes with Eigen2 2.0.6 C++ matrix library.
2009-09-24 21:22:24 +00:00
#include "BIK_api.h"
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
make bullet optional again
2013-01-23 07:26:39 +00:00
#ifdef WITH_BULLET
# include "RBI_api.h"
#endif
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
/* both in intern */
fix error in last commit
2012-03-27 01:52:49 +00:00
#ifdef WITH_SMOKE
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
#include "smoke_API.h"
use cmake WITH_MOD_ * to exclude include paths which are not used.
2012-03-27 00:17:57 +00:00
#endif
options WITH_LZO and WITH_LZMA for cmake and scons (default to true) pointcache.c also needed to have checks for these defines.
2009-09-25 02:40:10 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
#ifdef WITH_OPENVDB
#include "openvdb_capi.h"
#endif
options WITH_LZO and WITH_LZMA for cmake and scons (default to true) pointcache.c also needed to have checks for these defines.
2009-09-25 02:40:10 +00:00
#ifdef WITH_LZO
CMake: unbundle LZO library Patch T41989 by @hasufell
2015-03-13 22:33:31 +11:00
# ifdef WITH_SYSTEM_LZO
# include <lzo/lzo1x.h>
# else
# include "minilzo.h"
# endif
# define LZO_HEAP_ALLOC(var,size) \
Fix T41974: blender ships vulnerable minilzo version Update lzo library due to security issue Update lzo to latest version. This fixes a security issue (reported in T41974) This also fixes some (false-positive) valgrind warnings. Reviewers: sergey Differential Revision: https://developer.blender.org/D803
2014-09-28 01:05:36 +06:00
lzo_align_t __LZO_MMODEL var [ ((size) + (sizeof(lzo_align_t) - 1)) / sizeof(lzo_align_t) ]
options WITH_LZO and WITH_LZMA for cmake and scons (default to true) pointcache.c also needed to have checks for these defines.
2009-09-25 02:40:10 +00:00
#endif
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Fix T41974: blender ships vulnerable minilzo version Update lzo library due to security issue Update lzo to latest version. This fixes a security issue (reported in T41974) This also fixes some (false-positive) valgrind warnings. Reviewers: sergey Differential Revision: https://developer.blender.org/D803
2014-09-28 01:05:36 +06:00
#define LZO_OUT_LEN(size) ((size) + (size) / 16 + 64 + 3)
options WITH_LZO and WITH_LZMA for cmake and scons (default to true) pointcache.c also needed to have checks for these defines.
2009-09-25 02:40:10 +00:00
#ifdef WITH_LZMA
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
#include "LzmaLib.h"
options WITH_LZO and WITH_LZMA for cmake and scons (default to true) pointcache.c also needed to have checks for these defines.
2009-09-25 02:40:10 +00:00
#endif
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
/* needed for directory lookup */
#ifndef WIN32
style/name cleanup: have EDBM_* functions match our style guide and also match BM_ function naming conventions
2012-03-27 04:46:52 +00:00
# include <dirent.h>
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
#else
style/name cleanup: have EDBM_* functions match our style guide and also match BM_ function naming conventions
2012-03-27 04:46:52 +00:00
# include "BLI_winstuff.h"
Fix for part of bug #17759: point cache clearing did too much, also clearing other physics systems on the same object.
2008-10-04 21:26:48 +00:00
#endif
code cleanup: defines with braces - end with '(void)0' so callers must end with ';' like normal function. ... without this some editors dont parse the source so well.
2012-05-27 12:21:13 +00:00
#define PTCACHE_DATA_FROM(data, type, from) \
if (data[type]) { \
memcpy(data[type], from, ptcache_data_size[type]); \
} (void)0
#define PTCACHE_DATA_TO(data, type, index, to) \
if (data[type]) { \
memcpy(to, (char *)(data)[type] + ((index) ? (index) * ptcache_data_size[type] : 0), ptcache_data_size[type]); \
} (void)0
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
/* could be made into a pointcache option */
#define DURIAN_POINTCACHE_LIB_OK 1
Compiler warning fixes for the same point cache things Joshua tried to fix earlier, don't know why I didn't notice these before.
2011-01-10 02:40:24 +00:00
static int ptcache_data_size[] = {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
sizeof(unsigned int), // BPHYS_DATA_INDEX
3 * sizeof(float), // BPHYS_DATA_LOCATION
3 * sizeof(float), // BPHYS_DATA_VELOCITY
4 * sizeof(float), // BPHYS_DATA_ROTATION
3 * sizeof(float), // BPHYS_DATA_AVELOCITY / BPHYS_DATA_XCONST
sizeof(float), // BPHYS_DATA_SIZE
3 * sizeof(float), // BPHYS_DATA_TIMES
sizeof(BoidData) // case BPHYS_DATA_BOIDS
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
};
Compiler warning fixes for the same point cache things Joshua tried to fix earlier, don't know why I didn't notice these before.
2011-01-10 02:40:24 +00:00
static int ptcache_extra_datasize[] = {
0,
sizeof(ParticleSpring)
};
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
/* forward declerations */
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static int ptcache_file_compressed_read(PTCacheFile *pf, unsigned char *result, unsigned int len);
static int ptcache_file_compressed_write(PTCacheFile *pf, unsigned char *in, unsigned int in_len, unsigned char *out, int mode);
quiet some warnings from recent merge
2011-11-10 12:28:26 +00:00
static int ptcache_file_write(PTCacheFile *pf, const void *f, unsigned int tot, unsigned int size);
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static int ptcache_file_read(PTCacheFile *pf, void *f, unsigned int tot, unsigned int size);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Common functions */
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_basic_header_read(PTCacheFile *pf)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int error=0;
/* Custom functions should read these basic elements too! */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && !fread(&pf->totpoint, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
error = 1;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && !fread(&pf->data_types, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
error = 1;
return !error;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_basic_header_write(PTCacheFile *pf)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
/* Custom functions should write these basic elements too! */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!fwrite(&pf->totpoint, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!fwrite(&pf->data_types, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
return 1;
}
/* Softbody functions */
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_softbody_write(int index, void *soft_v, void **data, int UNUSED(cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
SoftBody *soft= soft_v;
BodyPoint *bp = soft->bpoint + index;
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_FROM(data, BPHYS_DATA_LOCATION, bp->pos);
PTCACHE_DATA_FROM(data, BPHYS_DATA_VELOCITY, bp->vec);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 1;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_softbody_read(int index, void *soft_v, void **data, float UNUSED(cfra), float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
SoftBody *soft= soft_v;
BodyPoint *bp = soft->bpoint + index;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(bp->pos, data, 3 * sizeof(float));
memcpy(bp->vec, data + 3, 3 * sizeof(float));
}
else {
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_TO(data, BPHYS_DATA_LOCATION, 0, bp->pos);
PTCACHE_DATA_TO(data, BPHYS_DATA_VELOCITY, 0, bp->vec);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_softbody_interpolate(int index, void *soft_v, void **data, float cfra, float cfra1, float cfra2, float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
SoftBody *soft= soft_v;
BodyPoint *bp = soft->bpoint + index;
ParticleKey keys[4];
float dfra;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 == cfra2)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return;
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
copy_v3_v3(keys[1].co, bp->pos);
copy_v3_v3(keys[1].vel, bp->vec);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(keys[2].co, old_data, 3 * sizeof(float));
memcpy(keys[2].vel, old_data + 3, 3 * sizeof(float));
}
else
BKE_ptcache_make_particle_key(keys+2, 0, data, cfra2);
dfra = cfra2 - cfra1;
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
mul_v3_fl(keys[1].vel, dfra);
mul_v3_fl(keys[2].vel, dfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
psys_interpolate_particle(-1, keys, (cfra - cfra1) / dfra, keys, 1);
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
mul_v3_fl(keys->vel, 1.0f / dfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
copy_v3_v3(bp->pos, keys->co);
copy_v3_v3(bp->vec, keys->vel);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_softbody_totpoint(void *soft_v, int UNUSED(cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
SoftBody *soft= soft_v;
return soft->totpoint;
}
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_softbody_error(void *UNUSED(soft_v), const char *UNUSED(message))
{
/* ignored for now */
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Particle functions */
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
void BKE_ptcache_make_particle_key(ParticleKey *key, int index, void **data, float time)
{
PTCACHE_DATA_TO(data, BPHYS_DATA_LOCATION, index, key->co);
PTCACHE_DATA_TO(data, BPHYS_DATA_VELOCITY, index, key->vel);
/* no rotation info, so make something nice up */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_ROTATION]==NULL) {
style cleanup: whitespace / commas
2012-04-29 15:47:02 +00:00
vec_to_quat(key->rot, key->vel, OB_NEGX, OB_POSZ);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
else {
PTCACHE_DATA_TO(data, BPHYS_DATA_ROTATION, index, key->rot);
}
PTCACHE_DATA_TO(data, BPHYS_DATA_AVELOCITY, index, key->ave);
key->time = time;
}
static int ptcache_particle_write(int index, void *psys_v, void **data, int cfra)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ParticleSystem *psys= psys_v;
ParticleData *pa = psys->particles + index;
Some particles cleanup & fixes: - Some big refresh issues with softbody & cloth point cache usage should now be fixed. - Removed sticky objects from particles (better stuff will come back when I get to updating reactor particles). - Some initial easy memory efficiency cleanup for ParticleData struct. The ultimate goal is to get particles less memory hungry -> more particles possible in single scene. - Wrong path timing clamping caused hair particles to seem disappeared when changing between normal<->hair particles. - "Calculate to current frame" in cache buttons baked instead of the intended function. - Boids particle data is now a bit better organized.
2009-09-04 23:06:15 +00:00
BoidParticle *boid = (psys->part->phystype == PART_PHYS_BOIDS) ? pa->boid : NULL;
use c90 compatible static initializers.
2010-11-03 06:31:53 +00:00
float times[3];
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
int step = psys->pointcache->step;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
/* No need to store unborn or died particles outside cache step bounds */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_INDEX] && (cfra < pa->time - step || cfra > pa->dietime + step))
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
return 0;
use c90 compatible static initializers.
2010-11-03 06:31:53 +00:00
style cleanup
2012-10-22 08:15:51 +00:00
times[0] = pa->time;
times[1] = pa->dietime;
times[2] = pa->lifetime;
use c90 compatible static initializers.
2010-11-03 06:31:53 +00:00
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_FROM(data, BPHYS_DATA_INDEX, &index);
PTCACHE_DATA_FROM(data, BPHYS_DATA_LOCATION, pa->state.co);
PTCACHE_DATA_FROM(data, BPHYS_DATA_VELOCITY, pa->state.vel);
PTCACHE_DATA_FROM(data, BPHYS_DATA_ROTATION, pa->state.rot);
PTCACHE_DATA_FROM(data, BPHYS_DATA_AVELOCITY, pa->state.ave);
PTCACHE_DATA_FROM(data, BPHYS_DATA_SIZE, &pa->size);
PTCACHE_DATA_FROM(data, BPHYS_DATA_TIMES, times);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
add missing break to space-buttons listener, also rewind own recent change (uiPopupBlockHandle.retvec size)
2013-07-25 12:46:20 +00:00
if (boid) {
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_FROM(data, BPHYS_DATA_BOIDS, &boid->data);
add missing break to space-buttons listener, also rewind own recent change (uiPopupBlockHandle.retvec size)
2013-07-25 12:46:20 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
/* return flag 1+1=2 for newly born particles to copy exact birth location to previously cached frame */
return 1 + (pa->state.time >= pa->time && pa->prev_state.time <= pa->time);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_particle_read(int index, void *psys_v, void **data, float cfra, float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ParticleSystem *psys= psys_v;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
ParticleData *pa;
BoidParticle *boid;
style cleanup
2013-01-09 03:30:15 +00:00
float timestep = 0.04f * psys->part->timetweak;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (index >= psys->totpart)
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
return;
pa = psys->particles + index;
boid = (psys->part->phystype == PART_PHYS_BOIDS) ? pa->boid : NULL;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra > pa->state.time)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(&pa->prev_state, &pa->state, sizeof(ParticleKey));
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* old format cache */
memcpy(&pa->state, old_data, sizeof(ParticleKey));
return;
}
BKE_ptcache_make_particle_key(&pa->state, 0, data, cfra);
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
/* set frames cached before birth to birth time */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra < pa->time)
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
pa->state.time = pa->time;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (cfra > pa->dietime)
Bug fix: with bigger than 1 cache step dead particles didn't alway stay dead * Particle die time wasn't properly taken into account when reading from the cache.
2011-02-18 15:21:52 +00:00
pa->state.time = pa->dietime;
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
add missing break to space-buttons listener, also rewind own recent change (uiPopupBlockHandle.retvec size)
2013-07-25 12:46:20 +00:00
if (data[BPHYS_DATA_SIZE]) {
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_TO(data, BPHYS_DATA_SIZE, 0, &pa->size);
add missing break to space-buttons listener, also rewind own recent change (uiPopupBlockHandle.retvec size)
2013-07-25 12:46:20 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_TIMES]) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
float times[3];
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_TO(data, BPHYS_DATA_TIMES, 0, &times);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
pa->time = times[0];
pa->dietime = times[1];
pa->lifetime = times[2];
}
code cleanup: remove some paranoid checks which would have crashed anyway earlier on. Also some minor formatting.
2013-01-20 14:50:50 +00:00
if (boid) {
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_TO(data, BPHYS_DATA_BOIDS, 0, &boid->data);
code cleanup: remove some paranoid checks which would have crashed anyway earlier on. Also some minor formatting.
2013-01-20 14:50:50 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* determine velocity from previous location */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_VELOCITY]) {
if (cfra > pa->prev_state.time) {
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
sub_v3_v3v3(pa->state.vel, pa->state.co, pa->prev_state.co);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(pa->state.vel, (cfra - pa->prev_state.time) * timestep);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
else {
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
sub_v3_v3v3(pa->state.vel, pa->prev_state.co, pa->state.co);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(pa->state.vel, (pa->prev_state.time - cfra) * timestep);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
}
"Fix" for [#30098] Particle rotation wrong / explode modifier - The main problem was that in order to be accurate all particle rotations have to be calculated incrementally so the only working solution is to store rotations to the point cache (previously this was only done for dynamic rotations). This can nearly double the point cache size so it's not ideal to have this as a default as in many cases you don't care about particle rotations. - Particle rotation panel now has a new "enable" checkbox that enables rotation calculations and the storing of rotations to point cache. - Old files will have rotations enabled via do_versions so that in the worst case old files will only get bigger point caches, but no sudden loss of particle rotations.
2012-03-18 21:33:00 +00:00
/* default to no rotation */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_ROTATION]) {
use math vector init functions
2012-07-17 12:05:15 +00:00
unit_qt(pa->state.rot);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_particle_interpolate(int index, void *psys_v, void **data, float cfra, float cfra1, float cfra2, float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ParticleSystem *psys= psys_v;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
ParticleData *pa;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
ParticleKey keys[4];
style cleanup
2013-01-09 03:30:15 +00:00
float dfra, timestep = 0.04f * psys->part->timetweak;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (index >= psys->totpart)
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
return;
pa = psys->particles + index;
/* particle wasn't read from first cache so can't interpolate */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if ((int)cfra1 < pa->time - psys->pointcache->step || (int)cfra1 > pa->dietime + psys->pointcache->step)
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
return;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
cfra = MIN2(cfra, pa->dietime);
cfra1 = MIN2(cfra1, pa->dietime);
cfra2 = MIN2(cfra2, pa->dietime);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 == cfra2)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return;
memcpy(keys+1, &pa->state, sizeof(ParticleKey));
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(keys+2, old_data, sizeof(ParticleKey));
else
BKE_ptcache_make_particle_key(keys+2, 0, data, cfra2);
General particle bug fixes + few small goodies The goodies: * Curves can be used as normal dynamic effectors too with the new "curve" field shape. * Group visualization has optional duplication counts for each object in the specified group. * Object & group visualizations, which are done without taking the dupliobject's global position into account (unless the whole group is used). This is much nicer than the previous behavior, but I added a "Use Global Location" option for those who want to use it the old way. * The active particle system's particles are now drawn a with theme coloured outline instead of pure white. * Added object aligned velocity factors (buttons categorized and re-organized too). Bug fixes: * Absorption didn't work as the ui toggle button was forgotten. * Some other force field ui tweaks. * Crash after adding children and changing trails count. * Display types "cross" and "axis" crashed. * Particles weren't drawn with correct coloring. * Billboards didn't update properly in viewport to camera location changes. * Particle rotation wasn't recreated correctly from point cache. * Changing particles amount crashed sometimes. * Some files with child hair crashed on loading. * Compiler warning fixes. * Adding boids crashed on frame 1;
2009-10-05 13:25:56 +00:00
/* determine velocity from previous location */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_VELOCITY]) {
if (keys[1].time > keys[2].time) {
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
sub_v3_v3v3(keys[2].vel, keys[1].co, keys[2].co);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(keys[2].vel, (keys[1].time - keys[2].time) * timestep);
General particle bug fixes + few small goodies The goodies: * Curves can be used as normal dynamic effectors too with the new "curve" field shape. * Group visualization has optional duplication counts for each object in the specified group. * Object & group visualizations, which are done without taking the dupliobject's global position into account (unless the whole group is used). This is much nicer than the previous behavior, but I added a "Use Global Location" option for those who want to use it the old way. * The active particle system's particles are now drawn a with theme coloured outline instead of pure white. * Added object aligned velocity factors (buttons categorized and re-organized too). Bug fixes: * Absorption didn't work as the ui toggle button was forgotten. * Some other force field ui tweaks. * Crash after adding children and changing trails count. * Display types "cross" and "axis" crashed. * Particles weren't drawn with correct coloring. * Billboards didn't update properly in viewport to camera location changes. * Particle rotation wasn't recreated correctly from point cache. * Changing particles amount crashed sometimes. * Some files with child hair crashed on loading. * Compiler warning fixes. * Adding boids crashed on frame 1;
2009-10-05 13:25:56 +00:00
}
else {
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
sub_v3_v3v3(keys[2].vel, keys[2].co, keys[1].co);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(keys[2].vel, (keys[2].time - keys[1].time) * timestep);
General particle bug fixes + few small goodies The goodies: * Curves can be used as normal dynamic effectors too with the new "curve" field shape. * Group visualization has optional duplication counts for each object in the specified group. * Object & group visualizations, which are done without taking the dupliobject's global position into account (unless the whole group is used). This is much nicer than the previous behavior, but I added a "Use Global Location" option for those who want to use it the old way. * The active particle system's particles are now drawn a with theme coloured outline instead of pure white. * Added object aligned velocity factors (buttons categorized and re-organized too). Bug fixes: * Absorption didn't work as the ui toggle button was forgotten. * Some other force field ui tweaks. * Crash after adding children and changing trails count. * Display types "cross" and "axis" crashed. * Particles weren't drawn with correct coloring. * Billboards didn't update properly in viewport to camera location changes. * Particle rotation wasn't recreated correctly from point cache. * Changing particles amount crashed sometimes. * Some files with child hair crashed on loading. * Compiler warning fixes. * Adding boids crashed on frame 1;
2009-10-05 13:25:56 +00:00
}
}
Fix T45936: invalid cycles motion blur for particle rotation and children. Commit rB709ca0ece changed how rotation was handled for particles so that unless actual rotation physics is enabled, there is no rotation. However it only updated ptcache_particle_read, forgetting to change exactly the same code in ptcache_particle_interpolate. This means that for subframes old code that computes a rotation from velocity is used, resulting in completely different rotation than for integer frames. This causes rotational motion blur by itself, and also mangles motion blur paths of child particles. Reviewers: sergey, lukastoenne Maniphest Tasks: T45936 Differential Revision: https://developer.blender.org/D2124
2016-07-26 18:00:07 +02:00
/* default to no rotation */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_ROTATION]) {
Fix T45936: invalid cycles motion blur for particle rotation and children. Commit rB709ca0ece changed how rotation was handled for particles so that unless actual rotation physics is enabled, there is no rotation. However it only updated ptcache_particle_read, forgetting to change exactly the same code in ptcache_particle_interpolate. This means that for subframes old code that computes a rotation from velocity is used, resulting in completely different rotation than for integer frames. This causes rotational motion blur by itself, and also mangles motion blur paths of child particles. Reviewers: sergey, lukastoenne Maniphest Tasks: T45936 Differential Revision: https://developer.blender.org/D2124
2016-07-26 18:00:07 +02:00
unit_qt(keys[2].rot);
General particle bug fixes + few small goodies The goodies: * Curves can be used as normal dynamic effectors too with the new "curve" field shape. * Group visualization has optional duplication counts for each object in the specified group. * Object & group visualizations, which are done without taking the dupliobject's global position into account (unless the whole group is used). This is much nicer than the previous behavior, but I added a "Use Global Location" option for those who want to use it the old way. * The active particle system's particles are now drawn a with theme coloured outline instead of pure white. * Added object aligned velocity factors (buttons categorized and re-organized too). Bug fixes: * Absorption didn't work as the ui toggle button was forgotten. * Some other force field ui tweaks. * Crash after adding children and changing trails count. * Display types "cross" and "axis" crashed. * Particles weren't drawn with correct coloring. * Billboards didn't update properly in viewport to camera location changes. * Particle rotation wasn't recreated correctly from point cache. * Changing particles amount crashed sometimes. * Some files with child hair crashed on loading. * Compiler warning fixes. * Adding boids crashed on frame 1;
2009-10-05 13:25:56 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra > pa->time)
Unified effector functionality for particles, cloth and softbody * Unified scene wide gravity (currently in scene buttons) instead of each simulation having it's own gravity. * Weight parameters for all effectors and an effector group setting. * Every effector can use noise. * Most effectors have "shapes" point, plane, surface, every point. - "Point" is most like the old effectors and uses the effector location as the effector point. - "Plane" uses the closest point on effectors local xy-plane as the effector point. - "Surface" uses the closest point on an effector object's surface as the effector point. - "Every Point" uses every point in a mesh effector object as an effector point. - The falloff is calculated from this point, so for example with "surface" shape and "use only negative z axis" it's possible to apply force only "inside" the effector object. * Spherical effector is now renamed as "force" as it's no longer just spherical. * New effector parameter "flow", which makes the effector act as surrounding air velocity, so the resulting force is proportional to the velocity difference of the point and "air velocity". For example a wind field with flow=1.0 results in proper non-accelerating wind. * New effector fields "turbulence", which creates nice random flow paths, and "drag", which slows the points down. * Much improved vortex field. * Effectors can now effect particle rotation as well as location. * Use full, or only positive/negative z-axis to apply force (note. the z-axis is the surface normal in the case of effector shape "surface") * New "force field" submenu in add menu, which adds an empty with the chosen effector (curve object for corve guides). * Other dynamics should be quite easy to add to the effector system too if wanted. * "Unified" doesn't mean that force fields give the exact same results for particles, softbody & cloth, since their final effect depends on many external factors, like for example the surface area of the effected faces. Code changes * Subversion bump for correct handling of global gravity. * Separate ui py file for common dynamics stuff. * Particle settings updating is flushed with it's id through DAG_id_flush_update(..). Known issues * Curve guides don't yet have all ui buttons in place, but they should work none the less. * Hair dynamics don't yet respect force fields. Other changes * Particle emission defaults now to frames 1-200 with life of 50 frames to fill the whole default timeline. * Many particles drawing related crashes fixed. * Sometimes particles didn't update on first frame properly. * Hair with object/group visualization didn't work properly. * Memory leaks with PointCacheID lists (Genscher, remember to free pidlists after use :).
2009-09-30 22:10:14 +00:00
cfra1 = MAX2(cfra1, pa->time);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
dfra = cfra2 - cfra1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(keys[1].vel, dfra * timestep);
mul_v3_fl(keys[2].vel, dfra * timestep);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
psys_interpolate_particle(-1, keys, (cfra - cfra1) / dfra, &pa->state, 1);
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
interp_qt_qtqt(pa->state.rot, keys[1].rot, keys[2].rot, (cfra - cfra1) / dfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
mul_v3_fl(pa->state.vel, 1.f / (dfra * timestep));
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
pa->state.time = cfra;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_particle_totpoint(void *psys_v, int UNUSED(cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ParticleSystem *psys = psys_v;
return psys->totpart;
}
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_particle_error(void *UNUSED(psys_v), const char *UNUSED(message))
{
/* ignored for now */
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_particle_totwrite(void *psys_v, int cfra)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ParticleSystem *psys = psys_v;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
ParticleData *pa= psys->particles;
int p, step = psys->pointcache->step;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int totwrite = 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra == 0)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
return psys->totpart;
style cleanup: whitespace / commas
2012-04-29 15:47:02 +00:00
for (p=0; p<psys->totpart; p++, pa++)
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
totwrite += (cfra >= pa->time - step && cfra <= pa->dietime + step);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return totwrite;
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
static void ptcache_particle_extra_write(void *psys_v, PTCacheMem *pm, int UNUSED(cfra))
{
ParticleSystem *psys = psys_v;
PTCacheExtra *extra = NULL;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->phystype == PART_PHYS_FLUID &&
Cleanup: indentation
2018-04-16 17:08:27 +02:00
psys->part->fluid && psys->part->fluid->flag & SPH_VISCOELASTIC_SPRINGS &&
psys->tot_fluidsprings && psys->fluid_springs)
{
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
extra = MEM_callocN(sizeof(PTCacheExtra), "Point cache: fluid extra data");
extra->type = BPHYS_EXTRA_FLUID_SPRINGS;
extra->totdata = psys->tot_fluidsprings;
extra->data = MEM_callocN(extra->totdata * ptcache_extra_datasize[extra->type], "Point cache: extra data");
memcpy(extra->data, psys->fluid_springs, extra->totdata * ptcache_extra_datasize[extra->type]);
BLI_addtail(&pm->extradata, extra);
}
}
callback function ptcache_particle_extra_read() was incorrect type, also fix some other compiler warnings with recent commits.
2011-01-10 05:17:53 +00:00
static void ptcache_particle_extra_read(void *psys_v, PTCacheMem *pm, float UNUSED(cfra))
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
{
ParticleSystem *psys = psys_v;
PTCacheExtra *extra = pm->extradata.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; extra; extra=extra->next) {
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
switch (extra->type) {
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
case BPHYS_EXTRA_FLUID_SPRINGS:
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->fluid_springs)
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
MEM_freeN(psys->fluid_springs);
psys->fluid_springs = MEM_dupallocN(extra->data);
psys->tot_fluidsprings = psys->alloc_fluidsprings = extra->totdata;
break;
}
}
}
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Cloth functions */
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_cloth_write(int index, void *cloth_v, void **data, int UNUSED(cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ClothModifierData *clmd= cloth_v;
Cloth *cloth= clmd->clothObject;
ClothVertex *vert = cloth->verts + index;
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_FROM(data, BPHYS_DATA_LOCATION, vert->x);
PTCACHE_DATA_FROM(data, BPHYS_DATA_VELOCITY, vert->v);
PTCACHE_DATA_FROM(data, BPHYS_DATA_XCONST, vert->xconst);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 1;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_cloth_read(int index, void *cloth_v, void **data, float UNUSED(cfra), float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ClothModifierData *clmd= cloth_v;
Cloth *cloth= clmd->clothObject;
ClothVertex *vert = cloth->verts + index;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(vert->x, data, 3 * sizeof(float));
memcpy(vert->xconst, data + 3, 3 * sizeof(float));
memcpy(vert->v, data + 6, 3 * sizeof(float));
}
else {
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
PTCACHE_DATA_TO(data, BPHYS_DATA_LOCATION, 0, vert->x);
PTCACHE_DATA_TO(data, BPHYS_DATA_VELOCITY, 0, vert->v);
PTCACHE_DATA_TO(data, BPHYS_DATA_XCONST, 0, vert->xconst);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_cloth_interpolate(int index, void *cloth_v, void **data, float cfra, float cfra1, float cfra2, float *old_data)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ClothModifierData *clmd= cloth_v;
Cloth *cloth= clmd->clothObject;
ClothVertex *vert = cloth->verts + index;
ParticleKey keys[4];
float dfra;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 == cfra2)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return;
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
copy_v3_v3(keys[1].co, vert->x);
copy_v3_v3(keys[1].vel, vert->v);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (old_data) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
memcpy(keys[2].co, old_data, 3 * sizeof(float));
memcpy(keys[2].vel, old_data + 6, 3 * sizeof(float));
}
else
BKE_ptcache_make_particle_key(keys+2, 0, data, cfra2);
dfra = cfra2 - cfra1;
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
mul_v3_fl(keys[1].vel, dfra);
mul_v3_fl(keys[2].vel, dfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
psys_interpolate_particle(-1, keys, (cfra - cfra1) / dfra, keys, 1);
Math Lib * Convert all code to use new functions. * Branch maintainers may want to skip this commit, and run this conversion script instead, if they use a lot of math functions in new code: http://www.pasteall.org/9052/python
2009-11-10 20:43:45 +00:00
mul_v3_fl(keys->vel, 1.0f / dfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
copy_v3_v3(vert->x, keys->co);
copy_v3_v3(vert->v, keys->vel);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* should vert->xconst be interpolated somehow too? - jahka */
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_cloth_totpoint(void *cloth_v, int UNUSED(cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
ClothModifierData *clmd= cloth_v;
Replace MFace w/ vert-tri's for collision modifier Note that the collision modifier doesn't have any use for Loop indices, so to avoid duplicating the loop array too, MVertTri has been added which simply stores vertex indices (runtime only).
2015-07-31 14:00:07 +10:00
return clmd->clothObject ? clmd->clothObject->mvert_num : 0;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_cloth_error(void *cloth_v, const char *message)
{
ClothModifierData *clmd= cloth_v;
modifier_setError(&clmd->modifier, "%s", message);
}
cmake option to build without smoke sim: WITH_MOD_SMOKE
2011-07-13 18:40:21 +00:00
#ifdef WITH_SMOKE
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
/* Smoke functions */
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_smoke_totpoint(void *smoke_v, int UNUSED(cfra))
{
SmokeModifierData *smd= (SmokeModifierData *)smoke_v;
SmokeDomainSettings *sds = smd->domain;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->fluid) {
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
return sds->base_res[0]*sds->base_res[1]*sds->base_res[2];
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
else
return 0;
}
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_smoke_error(void *smoke_v, const char *message)
{
SmokeModifierData *smd= (SmokeModifierData *)smoke_v;
modifier_setError(&smd->modifier, "%s", message);
}
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
#define SMOKE_CACHE_VERSION "1.04"
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_smoke_write(PTCacheFile *pf, void *smoke_v)
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
{
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
SmokeModifierData *smd= (SmokeModifierData *)smoke_v;
SmokeDomainSettings *sds = smd->domain;
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
int ret = 0;
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
int fluid_fields = smoke_get_data_flags(sds);
/* version header */
ptcache_file_write(pf, SMOKE_CACHE_VERSION, 4, sizeof(char));
ptcache_file_write(pf, &fluid_fields, 1, sizeof(int));
ptcache_file_write(pf, &sds->active_fields, 1, sizeof(int));
ptcache_file_write(pf, &sds->res, 3, sizeof(int));
ptcache_file_write(pf, &sds->dx, 1, sizeof(float));
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->fluid) {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
size_t res = sds->res[0]*sds->res[1]*sds->res[2];
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
float dt, dx, *dens, *react, *fuel, *flame, *heat, *heatold, *vx, *vy, *vz, *r, *g, *b;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
unsigned char *obstacles;
unsigned int in_len = sizeof(float)*(unsigned int)res;
style cleanup
2013-01-08 02:06:16 +00:00
unsigned char *out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len) * 4, "pointcache_lzo_buffer");
Smoke: The well known Miika Hämäläinen (aka MiikaH) patch (http://blenderartists.org/forum/showthread.php?t=158317&page=42) * Better (and windows enabled) OpenMP handling (> 2x-5x speed) * More Volumetric Texture mapping options (heat, etc) <-- Matt if that's not to your liking, just revert that part, it's separate anyway * Initial velocity taken from particle settings (no more slow starting) * Option to select compression method (there seem to be a bug in my high compression usage, at least it's been reported to result in exploding smoke - better use low compression for the time being) It's been tested since a while but as usual please report any (new!) bugs. ;-)
2010-01-25 15:10:14 +00:00
//int mode = res >= 1000000 ? 2 : 1;
int mode=1; // light
if (sds->cache_comp == SM_CACHE_HEAVY) mode=2; // heavy
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_export(sds->fluid, &dt, &dx, &dens, &react, &flame, &fuel, &heat, &heatold, &vx, &vy, &vz, &r, &g, &b, &obstacles);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)sds->shadow, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)dens, in_len, out, mode);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
if (fluid_fields & SM_ACTIVE_HEAT) {
ptcache_file_compressed_write(pf, (unsigned char *)heat, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)heatold, in_len, out, mode);
}
if (fluid_fields & SM_ACTIVE_FIRE) {
ptcache_file_compressed_write(pf, (unsigned char *)flame, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)fuel, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)react, in_len, out, mode);
}
if (fluid_fields & SM_ACTIVE_COLORS) {
ptcache_file_compressed_write(pf, (unsigned char *)r, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)g, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)b, in_len, out, mode);
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)vx, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)vy, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)vz, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)obstacles, (unsigned int)res, out, mode);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
ptcache_file_write(pf, &dt, 1, sizeof(float));
ptcache_file_write(pf, &dx, 1, sizeof(float));
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
ptcache_file_write(pf, &sds->p0, 3, sizeof(float));
ptcache_file_write(pf, &sds->p1, 3, sizeof(float));
ptcache_file_write(pf, &sds->dp0, 3, sizeof(float));
ptcache_file_write(pf, &sds->shift, 3, sizeof(int));
ptcache_file_write(pf, &sds->obj_shift_f, 3, sizeof(float));
ptcache_file_write(pf, &sds->obmat, 16, sizeof(float));
ptcache_file_write(pf, &sds->base_res, 3, sizeof(int));
ptcache_file_write(pf, &sds->res_min, 3, sizeof(int));
ptcache_file_write(pf, &sds->res_max, 3, sizeof(int));
ptcache_file_write(pf, &sds->active_color, 3, sizeof(float));
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
MEM_freeN(out);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
ret = 1;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->wt) {
Smoke: * Fixing some gcc warnings
2009-09-14 16:46:34 +00:00
int res_big_array[3];
int res_big;
int res = sds->res[0]*sds->res[1]*sds->res[2];
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
float *dens, *react, *fuel, *flame, *tcu, *tcv, *tcw, *r, *g, *b;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
unsigned int in_len = sizeof(float)*(unsigned int)res;
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
unsigned int in_len_big;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
unsigned char *out;
int mode;
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
smoke_turbulence_get_res(sds->wt, res_big_array);
res_big = res_big_array[0]*res_big_array[1]*res_big_array[2];
Smoke: The well known Miika Hämäläinen (aka MiikaH) patch (http://blenderartists.org/forum/showthread.php?t=158317&page=42) * Better (and windows enabled) OpenMP handling (> 2x-5x speed) * More Volumetric Texture mapping options (heat, etc) <-- Matt if that's not to your liking, just revert that part, it's separate anyway * Initial velocity taken from particle settings (no more slow starting) * Option to select compression method (there seem to be a bug in my high compression usage, at least it's been reported to result in exploding smoke - better use low compression for the time being) It's been tested since a while but as usual please report any (new!) bugs. ;-)
2010-01-25 15:10:14 +00:00
//mode = res_big >= 1000000 ? 2 : 1;
mode = 1; // light
if (sds->cache_high_comp == SM_CACHE_HEAVY) mode=2; // heavy
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
in_len_big = sizeof(float) * (unsigned int)res_big;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_turbulence_export(sds->wt, &dens, &react, &flame, &fuel, &r, &g, &b, &tcu, &tcv, &tcw);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len_big), "pointcache_lzo_buffer");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)dens, in_len_big, out, mode);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
if (fluid_fields & SM_ACTIVE_FIRE) {
ptcache_file_compressed_write(pf, (unsigned char *)flame, in_len_big, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)fuel, in_len_big, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)react, in_len_big, out, mode);
}
if (fluid_fields & SM_ACTIVE_COLORS) {
ptcache_file_compressed_write(pf, (unsigned char *)r, in_len_big, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)g, in_len_big, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)b, in_len_big, out, mode);
}
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
MEM_freeN(out);
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len), "pointcache_lzo_buffer");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)tcu, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)tcv, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)tcw, in_len, out, mode);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
MEM_freeN(out);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
ret = 1;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
}
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
return ret;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
}
Smoke Bugfix /enhancement: Load pre 2.65 pointcaches. Warning: Just make sure that you DON'T free the cache at any point. This patch can only display existing pointcaches from e.g. 2.64
2013-03-18 21:33:48 +00:00
/* read old smoke cache from 2.64 */
static int ptcache_smoke_read_old(PTCacheFile *pf, void *smoke_v)
{
SmokeModifierData *smd= (SmokeModifierData *)smoke_v;
SmokeDomainSettings *sds = smd->domain;
if (sds->fluid) {
Cleanup: shadowing (blenkernel)
2015-11-23 11:27:02 +11:00
const size_t res = sds->res[0] * sds->res[1] * sds->res[2];
const unsigned int out_len = (unsigned int)res * sizeof(float);
Smoke Bugfix /enhancement: Load pre 2.65 pointcaches. Warning: Just make sure that you DON'T free the cache at any point. This patch can only display existing pointcaches from e.g. 2.64
2013-03-18 21:33:48 +00:00
float dt, dx, *dens, *heat, *heatold, *vx, *vy, *vz;
unsigned char *obstacles;
float *tmp_array = MEM_callocN(out_len, "Smoke old cache tmp");
int fluid_fields = smoke_get_data_flags(sds);
/* Part part of the new cache header */
sds->active_color[0] = 0.7f;
sds->active_color[1] = 0.7f;
sds->active_color[2] = 0.7f;
smoke_export(sds->fluid, &dt, &dx, &dens, NULL, NULL, NULL, &heat, &heatold, &vx, &vy, &vz, NULL, NULL, NULL, &obstacles);
ptcache_file_compressed_read(pf, (unsigned char *)sds->shadow, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)dens, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
if (fluid_fields & SM_ACTIVE_HEAT)
{
ptcache_file_compressed_read(pf, (unsigned char*)heat, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)heatold, out_len);
}
else
{
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
}
ptcache_file_compressed_read(pf, (unsigned char*)vx, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)vy, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)vz, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)obstacles, (unsigned int)res);
ptcache_file_read(pf, &dt, 1, sizeof(float));
ptcache_file_read(pf, &dx, 1, sizeof(float));
MEM_freeN(tmp_array);
if (pf->data_types & (1<<BPHYS_DATA_SMOKE_HIGH) && sds->wt) {
int res_big, res_big_array[3];
Cleanup: shadowing (blenkernel)
2015-11-23 11:27:02 +11:00
float *tcu, *tcv, *tcw;
Smoke Bugfix /enhancement: Load pre 2.65 pointcaches. Warning: Just make sure that you DON'T free the cache at any point. This patch can only display existing pointcaches from e.g. 2.64
2013-03-18 21:33:48 +00:00
unsigned int out_len_big;
unsigned char *tmp_array_big;
smoke_turbulence_get_res(sds->wt, res_big_array);
res_big = res_big_array[0]*res_big_array[1]*res_big_array[2];
out_len_big = sizeof(float) * (unsigned int)res_big;
tmp_array_big = MEM_callocN(out_len_big, "Smoke old cache tmp");
smoke_turbulence_export(sds->wt, &dens, NULL, NULL, NULL, NULL, NULL, NULL, &tcu, &tcv, &tcw);
ptcache_file_compressed_read(pf, (unsigned char*)dens, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char*)tmp_array_big, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char*)tcu, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tcv, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)tcw, out_len);
MEM_freeN(tmp_array_big);
}
}
return 1;
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
static int ptcache_smoke_read(PTCacheFile *pf, void *smoke_v)
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
{
SmokeModifierData *smd= (SmokeModifierData *)smoke_v;
SmokeDomainSettings *sds = smd->domain;
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
char version[4];
int ch_res[3];
float ch_dx;
int fluid_fields = smoke_get_data_flags(sds);
int cache_fields = 0;
int active_fields = 0;
int reallocate = 0;
/* version header */
ptcache_file_read(pf, version, 4, sizeof(char));
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (!STREQLEN(version, SMOKE_CACHE_VERSION, 4))
Smoke Bugfix /enhancement: Load pre 2.65 pointcaches. Warning: Just make sure that you DON'T free the cache at any point. This patch can only display existing pointcaches from e.g. 2.64
2013-03-18 21:33:48 +00:00
{
/* reset file pointer */
fseek(pf->fp, -4, SEEK_CUR);
return ptcache_smoke_read_old(pf, smoke_v);
}
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
/* fluid info */
ptcache_file_read(pf, &cache_fields, 1, sizeof(int));
ptcache_file_read(pf, &active_fields, 1, sizeof(int));
ptcache_file_read(pf, &ch_res, 3, sizeof(int));
ptcache_file_read(pf, &ch_dx, 1, sizeof(float));
/* check if resolution has changed */
if (sds->res[0] != ch_res[0] ||
Cleanup: indentation
2018-04-16 17:08:27 +02:00
sds->res[1] != ch_res[1] ||
sds->res[2] != ch_res[2])
{
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
if (sds->flags & MOD_SMOKE_ADAPTIVE_DOMAIN)
reallocate = 1;
else
return 0;
}
/* check if active fields have changed */
if (fluid_fields != cache_fields ||
active_fields != sds->active_fields)
reallocate = 1;
/* reallocate fluid if needed*/
if (reallocate) {
Fix #33353: Smoke color problems with adaptive domain High resolution cache didn't always read all written data fields which caused colors to get messed up.
2012-11-30 11:01:14 +00:00
sds->active_fields = active_fields | cache_fields;
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_reallocate_fluid(sds, ch_dx, ch_res, 1);
sds->dx = ch_dx;
VECCOPY(sds->res, ch_res);
sds->total_cells = ch_res[0]*ch_res[1]*ch_res[2];
style cleanup: bge, switch statements mostly. also left bmesh decimator on in previous commit.
2012-10-21 07:58:38 +00:00
if (sds->flags & MOD_SMOKE_HIGHRES) {
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_reallocate_highres_fluid(sds, ch_dx, ch_res, 1);
}
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->fluid) {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
size_t res = sds->res[0]*sds->res[1]*sds->res[2];
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
float dt, dx, *dens, *react, *fuel, *flame, *heat, *heatold, *vx, *vy, *vz, *r, *g, *b;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
unsigned char *obstacles;
unsigned int out_len = (unsigned int)res * sizeof(float);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_export(sds->fluid, &dt, &dx, &dens, &react, &flame, &fuel, &heat, &heatold, &vx, &vy, &vz, &r, &g, &b, &obstacles);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)sds->shadow, out_len);
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)dens, out_len);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
if (cache_fields & SM_ACTIVE_HEAT) {
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)heat, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)heatold, out_len);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
}
if (cache_fields & SM_ACTIVE_FIRE) {
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)flame, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)fuel, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)react, out_len);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
}
if (cache_fields & SM_ACTIVE_COLORS) {
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)r, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)g, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)b, out_len);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
}
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)vx, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)vy, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)vz, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)obstacles, (unsigned int)res);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
ptcache_file_read(pf, &dt, 1, sizeof(float));
ptcache_file_read(pf, &dx, 1, sizeof(float));
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
ptcache_file_read(pf, &sds->p0, 3, sizeof(float));
ptcache_file_read(pf, &sds->p1, 3, sizeof(float));
ptcache_file_read(pf, &sds->dp0, 3, sizeof(float));
ptcache_file_read(pf, &sds->shift, 3, sizeof(int));
ptcache_file_read(pf, &sds->obj_shift_f, 3, sizeof(float));
ptcache_file_read(pf, &sds->obmat, 16, sizeof(float));
ptcache_file_read(pf, &sds->base_res, 3, sizeof(int));
ptcache_file_read(pf, &sds->res_min, 3, sizeof(int));
ptcache_file_read(pf, &sds->res_max, 3, sizeof(int));
ptcache_file_read(pf, &sds->active_color, 3, sizeof(float));
}
if (pf->data_types & (1<<BPHYS_DATA_SMOKE_HIGH) && sds->wt) {
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
int res = sds->res[0]*sds->res[1]*sds->res[2];
int res_big, res_big_array[3];
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
float *dens, *react, *fuel, *flame, *tcu, *tcv, *tcw, *r, *g, *b;
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
unsigned int out_len = sizeof(float)*(unsigned int)res;
unsigned int out_len_big;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
smoke_turbulence_get_res(sds->wt, res_big_array);
res_big = res_big_array[0]*res_big_array[1]*res_big_array[2];
out_len_big = sizeof(float) * (unsigned int)res_big;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
smoke_turbulence_export(sds->wt, &dens, &react, &flame, &fuel, &r, &g, &b, &tcu, &tcv, &tcw);
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)dens, out_len_big);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
if (cache_fields & SM_ACTIVE_FIRE) {
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)flame, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char *)fuel, out_len_big);
Fix #33353: Smoke color problems with adaptive domain High resolution cache didn't always read all written data fields which caused colors to get messed up.
2012-11-30 11:01:14 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)react, out_len_big);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
}
if (cache_fields & SM_ACTIVE_COLORS) {
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)r, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char *)g, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char *)b, out_len_big);
Google Summer of Code project: "Smoke Simulator Improvements & Fire". Documentation & Test blend files: ------------------ http://wiki.blender.org/index.php/User:MiikaH/GSoC-2012-Smoke-Simulator-Improvements Credits: ------------------ Miika Hamalainen (MiikaH): Student / Main programmer Daniel Genrich (Genscher): Mentor / Programmer of merged patches from Smoke2 branch Google: For Google Summer of Code 2012
2012-10-10 13:18:07 +00:00
}
Smoke: * Enable cache for high res + new preview * Bugfix for smoke banding (in cooperation with N_T) Hint: Work-in-progress regarding collision objects so can be broken, didn't test Hint2: jahka enabled a general particle panel but * bake button doesn't work * step is not supported for cloth * several other things there ;)
2009-09-09 18:39:40 +00:00
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)tcu, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)tcv, out_len);
ptcache_file_compressed_read(pf, (unsigned char *)tcw, out_len);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
return 1;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
}
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
#ifdef WITH_OPENVDB
/**
* Construct matrices which represent the fluid object, for low and high res:
* <pre>
* vs 0 0 0
* 0 vs 0 0
* 0 0 vs 0
* px py pz 1
* </pre>
*
* with `vs` = voxel size, and `px, py, pz`,
* the min position of the domain's bounding box.
*/
static void compute_fluid_matrices(SmokeDomainSettings *sds)
{
float bbox_min[3];
copy_v3_v3(bbox_min, sds->p0);
if (sds->flags & MOD_SMOKE_ADAPTIVE_DOMAIN) {
bbox_min[0] += (sds->cell_size[0] * (float)sds->res_min[0]);
bbox_min[1] += (sds->cell_size[1] * (float)sds->res_min[1]);
bbox_min[2] += (sds->cell_size[2] * (float)sds->res_min[2]);
add_v3_v3(bbox_min, sds->obj_shift_f);
}
/* construct low res matrix */
size_to_mat4(sds->fluidmat, sds->cell_size);
copy_v3_v3(sds->fluidmat[3], bbox_min);
/* The smoke simulator stores voxels cell-centered, whilst VDB is node
* centered, so we offset the matrix by half a voxel to compensate. */
madd_v3_v3fl(sds->fluidmat[3], sds->cell_size, 0.5f);
mul_m4_m4m4(sds->fluidmat, sds->obmat, sds->fluidmat);
if (sds->wt) {
float voxel_size_high[3];
/* construct high res matrix */
mul_v3_v3fl(voxel_size_high, sds->cell_size, 1.0f / (float)(sds->amplify + 1));
size_to_mat4(sds->fluidmat_wt, voxel_size_high);
copy_v3_v3(sds->fluidmat_wt[3], bbox_min);
/* Same here, add half a voxel to adjust the position of the fluid. */
madd_v3_v3fl(sds->fluidmat_wt[3], voxel_size_high, 0.5f);
mul_m4_m4m4(sds->fluidmat_wt, sds->obmat, sds->fluidmat_wt);
}
}
static int ptcache_smoke_openvdb_write(struct OpenVDBWriter *writer, void *smoke_v)
{
SmokeModifierData *smd = (SmokeModifierData *)smoke_v;
SmokeDomainSettings *sds = smd->domain;
OpenVDBWriter_set_flags(writer, sds->openvdb_comp, (sds->data_depth == 16));
OpenVDBWriter_add_meta_int(writer, "blender/smoke/active_fields", sds->active_fields);
OpenVDBWriter_add_meta_v3_int(writer, "blender/smoke/resolution", sds->res);
OpenVDBWriter_add_meta_v3_int(writer, "blender/smoke/min_resolution", sds->res_min);
OpenVDBWriter_add_meta_v3_int(writer, "blender/smoke/max_resolution", sds->res_max);
OpenVDBWriter_add_meta_v3_int(writer, "blender/smoke/base_resolution", sds->base_res);
OpenVDBWriter_add_meta_v3(writer, "blender/smoke/min_bbox", sds->p0);
OpenVDBWriter_add_meta_v3(writer, "blender/smoke/max_bbox", sds->p1);
OpenVDBWriter_add_meta_v3(writer, "blender/smoke/dp0", sds->dp0);
OpenVDBWriter_add_meta_v3_int(writer, "blender/smoke/shift", sds->shift);
OpenVDBWriter_add_meta_v3(writer, "blender/smoke/obj_shift_f", sds->obj_shift_f);
OpenVDBWriter_add_meta_v3(writer, "blender/smoke/active_color", sds->active_color);
OpenVDBWriter_add_meta_mat4(writer, "blender/smoke/obmat", sds->obmat);
int fluid_fields = smoke_get_data_flags(sds);
struct OpenVDBFloatGrid *clip_grid = NULL;
compute_fluid_matrices(sds);
OpenVDBWriter_add_meta_int(writer, "blender/smoke/fluid_fields", fluid_fields);
if (sds->wt) {
struct OpenVDBFloatGrid *wt_density_grid;
float *dens, *react, *fuel, *flame, *tcu, *tcv, *tcw, *r, *g, *b;
smoke_turbulence_export(sds->wt, &dens, &react, &flame, &fuel, &r, &g, &b, &tcu, &tcv, &tcw);
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
wt_density_grid = OpenVDB_export_grid_fl(writer, "density", dens, sds->res_wt, sds->fluidmat_wt, sds->clipping, NULL);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
clip_grid = wt_density_grid;
if (fluid_fields & SM_ACTIVE_FIRE) {
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, "flame", flame, sds->res_wt, sds->fluidmat_wt, sds->clipping, wt_density_grid);
OpenVDB_export_grid_fl(writer, "fuel", fuel, sds->res_wt, sds->fluidmat_wt, sds->clipping, wt_density_grid);
OpenVDB_export_grid_fl(writer, "react", react, sds->res_wt, sds->fluidmat_wt, sds->clipping, wt_density_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
if (fluid_fields & SM_ACTIVE_COLORS) {
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_vec(writer, "color", r, g, b, sds->res_wt, sds->fluidmat_wt, VEC_INVARIANT, true, sds->clipping, wt_density_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_vec(writer, "texture coordinates", tcu, tcv, tcw, sds->res, sds->fluidmat, VEC_INVARIANT, false, sds->clipping, wt_density_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
if (sds->fluid) {
struct OpenVDBFloatGrid *density_grid;
float dt, dx, *dens, *react, *fuel, *flame, *heat, *heatold, *vx, *vy, *vz, *r, *g, *b;
unsigned char *obstacles;
smoke_export(sds->fluid, &dt, &dx, &dens, &react, &flame, &fuel, &heat,
&heatold, &vx, &vy, &vz, &r, &g, &b, &obstacles);
OpenVDBWriter_add_meta_fl(writer, "blender/smoke/dx", dx);
OpenVDBWriter_add_meta_fl(writer, "blender/smoke/dt", dt);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
const char *name = (!sds->wt) ? "density" : "density_low";
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
density_grid = OpenVDB_export_grid_fl(writer, name, dens, sds->res, sds->fluidmat, sds->clipping, NULL);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
clip_grid = sds->wt ? clip_grid : density_grid;
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, "shadow", sds->shadow, sds->res, sds->fluidmat, sds->clipping, NULL);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (fluid_fields & SM_ACTIVE_HEAT) {
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, "heat", heat, sds->res, sds->fluidmat, sds->clipping, clip_grid);
OpenVDB_export_grid_fl(writer, "heat_old", heatold, sds->res, sds->fluidmat, sds->clipping, clip_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
if (fluid_fields & SM_ACTIVE_FIRE) {
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "flame" : "flame_low";
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, name, flame, sds->res, sds->fluidmat, sds->clipping, density_grid);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "fuel" : "fuel_low";
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, name, fuel, sds->res, sds->fluidmat, sds->clipping, density_grid);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "react" : "react_low";
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_fl(writer, name, react, sds->res, sds->fluidmat, sds->clipping, density_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
if (fluid_fields & SM_ACTIVE_COLORS) {
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "color" : "color_low";
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_vec(writer, name, r, g, b, sds->res, sds->fluidmat, VEC_INVARIANT, true, sds->clipping, density_grid);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
Smoke: expose empty space clipping property to the UI. This is used to determine which voxels are to be considered empty space. Previously it was hardcoded for converting dense grids to OpenVDB grids to reduce disk space usage. This value is also useful for rendering engines to know, i.e. to optimize ray marching.
2018-02-22 16:26:50 +01:00
OpenVDB_export_grid_vec(writer, "velocity", vx, vy, vz, sds->res, sds->fluidmat, VEC_CONTRAVARIANT_RELATIVE, false, sds->clipping, clip_grid);
OpenVDB_export_grid_ch(writer, "obstacles", obstacles, sds->res, sds->fluidmat, sds->clipping, NULL);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
return 1;
}
static int ptcache_smoke_openvdb_read(struct OpenVDBReader *reader, void *smoke_v)
{
SmokeModifierData *smd = (SmokeModifierData *)smoke_v;
if (!smd) {
return 0;
}
SmokeDomainSettings *sds = smd->domain;
int fluid_fields = smoke_get_data_flags(sds);
int active_fields, cache_fields = 0;
int cache_res[3];
float cache_dx;
bool reallocate = false;
OpenVDBReader_get_meta_v3_int(reader, "blender/smoke/min_resolution", sds->res_min);
OpenVDBReader_get_meta_v3_int(reader, "blender/smoke/max_resolution", sds->res_max);
OpenVDBReader_get_meta_v3_int(reader, "blender/smoke/base_resolution", sds->base_res);
OpenVDBReader_get_meta_v3(reader, "blender/smoke/min_bbox", sds->p0);
OpenVDBReader_get_meta_v3(reader, "blender/smoke/max_bbox", sds->p1);
OpenVDBReader_get_meta_v3(reader, "blender/smoke/dp0", sds->dp0);
OpenVDBReader_get_meta_v3_int(reader, "blender/smoke/shift", sds->shift);
OpenVDBReader_get_meta_v3(reader, "blender/smoke/obj_shift_f", sds->obj_shift_f);
OpenVDBReader_get_meta_v3(reader, "blender/smoke/active_color", sds->active_color);
OpenVDBReader_get_meta_mat4(reader, "blender/smoke/obmat", sds->obmat);
OpenVDBReader_get_meta_int(reader, "blender/smoke/fluid_fields", &cache_fields);
OpenVDBReader_get_meta_int(reader, "blender/smoke/active_fields", &active_fields);
OpenVDBReader_get_meta_fl(reader, "blender/smoke/dx", &cache_dx);
OpenVDBReader_get_meta_v3_int(reader, "blender/smoke/resolution", cache_res);
/* check if resolution has changed */
if (sds->res[0] != cache_res[0] ||
sds->res[1] != cache_res[1] ||
sds->res[2] != cache_res[2])
{
if (sds->flags & MOD_SMOKE_ADAPTIVE_DOMAIN) {
reallocate = true;
}
else {
return 0;
}
}
/* check if active fields have changed */
if ((fluid_fields != cache_fields) || (active_fields != sds->active_fields)) {
reallocate = true;
}
/* reallocate fluid if needed*/
if (reallocate) {
sds->active_fields = active_fields | cache_fields;
smoke_reallocate_fluid(sds, cache_dx, cache_res, 1);
sds->dx = cache_dx;
copy_v3_v3_int(sds->res, cache_res);
sds->total_cells = cache_res[0] * cache_res[1] * cache_res[2];
if (sds->flags & MOD_SMOKE_HIGHRES) {
smoke_reallocate_highres_fluid(sds, cache_dx, cache_res, 1);
}
}
if (sds->fluid) {
float dt, dx, *dens, *react, *fuel, *flame, *heat, *heatold, *vx, *vy, *vz, *r, *g, *b;
unsigned char *obstacles;
smoke_export(sds->fluid, &dt, &dx, &dens, &react, &flame, &fuel, &heat,
&heatold, &vx, &vy, &vz, &r, &g, &b, &obstacles);
OpenVDBReader_get_meta_fl(reader, "blender/smoke/dt", &dt);
OpenVDB_import_grid_fl(reader, "shadow", &sds->shadow, sds->res);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
const char *name = (!sds->wt) ? "density" : "density_low";
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, name, &dens, sds->res);
Fix T47559: Crash when scrubbing in time on a OpenVDB Smoke. Cache reading was not using the right simulation flags (outdated), which made it lookup non-existent grids, and crash on de-referencing null pointers.
2016-02-25 22:44:16 +01:00
if (cache_fields & SM_ACTIVE_HEAT) {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, "heat", &heat, sds->res);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
OpenVDB_import_grid_fl(reader, "heat_old", &heatold, sds->res);
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
}
Fix T47559: Crash when scrubbing in time on a OpenVDB Smoke. Cache reading was not using the right simulation flags (outdated), which made it lookup non-existent grids, and crash on de-referencing null pointers.
2016-02-25 22:44:16 +01:00
if (cache_fields & SM_ACTIVE_FIRE) {
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "flame" : "flame_low";
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, name, &flame, sds->res);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "fuel" : "fuel_low";
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, name, &fuel, sds->res);
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "react" : "react_low";
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, name, &react, sds->res);
}
Fix T47559: Crash when scrubbing in time on a OpenVDB Smoke. Cache reading was not using the right simulation flags (outdated), which made it lookup non-existent grids, and crash on de-referencing null pointers.
2016-02-25 22:44:16 +01:00
if (cache_fields & SM_ACTIVE_COLORS) {
OpenVDB : use underscores instead of spaces in grid names. Some other software cannot handle grid names with spaces in them. We still check for names with spaces so as to not break old files. This fixes T53802.
2018-01-29 18:37:35 +01:00
name = (!sds->wt) ? "color" : "color_low";
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_vec(reader, name, &r, &g, &b, sds->res);
}
OpenVDB_import_grid_vec(reader, "velocity", &vx, &vy, &vz, sds->res);
OpenVDB_import_grid_ch(reader, "obstacles", &obstacles, sds->res);
}
if (sds->wt) {
float *dens, *react, *fuel, *flame, *tcu, *tcv, *tcw, *r, *g, *b;
smoke_turbulence_export(sds->wt, &dens, &react, &flame, &fuel, &r, &g, &b, &tcu, &tcv, &tcw);
OpenVDB_import_grid_fl(reader, "density", &dens, sds->res_wt);
Fix crash in OpenVDB highres smoke cache when seeking to frame 0. This code obviously should also use the cache_fields flag variable, like the code for reading the lowres data in the same function. This is because fluid_fields actually represents the old state before smoke was reallocated to match cache_fields read from the file, and if it has some fields enabled that aren't allocated any more, it crashes. This also fixes a reverse glitch: when a file was loaded with the current frame in the middle of a baked smoke+fire simulation, smoke appeared immediately, but the fire didn't until the frame was changed. The reason is the same: after file load no fields are initially allocated and thus fluid_fields is 0.
2016-08-20 12:03:40 +03:00
if (cache_fields & SM_ACTIVE_FIRE) {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_fl(reader, "flame", &flame, sds->res_wt);
OpenVDB_import_grid_fl(reader, "fuel", &fuel, sds->res_wt);
OpenVDB_import_grid_fl(reader, "react", &react, sds->res_wt);
}
Fix crash in OpenVDB highres smoke cache when seeking to frame 0. This code obviously should also use the cache_fields flag variable, like the code for reading the lowres data in the same function. This is because fluid_fields actually represents the old state before smoke was reallocated to match cache_fields read from the file, and if it has some fields enabled that aren't allocated any more, it crashes. This also fixes a reverse glitch: when a file was loaded with the current frame in the middle of a baked smoke+fire simulation, smoke appeared immediately, but the fire didn't until the frame was changed. The reason is the same: after file load no fields are initially allocated and thus fluid_fields is 0.
2016-08-20 12:03:40 +03:00
if (cache_fields & SM_ACTIVE_COLORS) {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
OpenVDB_import_grid_vec(reader, "color", &r, &g, &b, sds->res_wt);
}
OpenVDB_import_grid_vec(reader, "texture coordinates", &tcu, &tcv, &tcw, sds->res);
}
OpenVDBReader_free(reader);
return 1;
}
#endif
cmake option to build without smoke sim: WITH_MOD_SMOKE
2011-07-13 18:40:21 +00:00
#else // WITH_SMOKE
pep/style edits & quiet some warnings
2011-11-11 03:28:46 +00:00
static int ptcache_smoke_totpoint(void *UNUSED(smoke_v), int UNUSED(cfra)) { return 0; }
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_smoke_error(void *UNUSED(smoke_v), const char *UNUSED(message)) { }
quiet some warnings from recent merge
2011-11-10 12:28:26 +00:00
static int ptcache_smoke_read(PTCacheFile *UNUSED(pf), void *UNUSED(smoke_v)) { return 0; }
cmake option to build without smoke sim: WITH_MOD_SMOKE
2011-07-13 18:40:21 +00:00
static int ptcache_smoke_write(PTCacheFile *UNUSED(pf), void *UNUSED(smoke_v)) { return 0; }
#endif // WITH_SMOKE
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
#if !defined(WITH_SMOKE) || !defined(WITH_OPENVDB)
static int ptcache_smoke_openvdb_write(struct OpenVDBWriter *writer, void *smoke_v)
{
UNUSED_VARS(writer, smoke_v);
return 0;
}
static int ptcache_smoke_openvdb_read(struct OpenVDBReader *reader, void *smoke_v)
{
UNUSED_VARS(reader, smoke_v);
return 0;
}
#endif
quiet some warnings from recent merge
2011-11-10 12:28:26 +00:00
static int ptcache_dynamicpaint_totpoint(void *sd, int UNUSED(cfra))
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
{
DynamicPaintSurface *surface = (DynamicPaintSurface*)sd;
if (!surface->data) return 0;
else return surface->data->total_points;
}
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_dynamicpaint_error(void *UNUSED(sd), const char *UNUSED(message))
{
/* ignored for now */
}
Dynamic Paint merge: Commit Dynamic Paint from "soc-2011-carrot" branch into trunk. End-user documentation: http://wiki.blender.org/index.php/Doc:2.6/Manual/Modifiers/Simulation/Dynamic_Paint GSoC wiki page: http://wiki.blender.org/index.php/User:MiikaH/GSoC-2011-DynamicPaint
2011-11-10 10:24:34 +00:00
#define DPAINT_CACHE_VERSION "1.01"
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
static int ptcache_dynamicpaint_write(PTCacheFile *pf, void *dp_v)
{
DynamicPaintSurface *surface = (DynamicPaintSurface*)dp_v;
int cache_compress = 1;
/* version header */
style cleanup
2013-01-08 02:06:16 +00:00
ptcache_file_write(pf, DPAINT_CACHE_VERSION, 1, sizeof(char) * 4);
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (surface->format != MOD_DPAINT_SURFACE_F_IMAGESEQ && surface->data) {
Dynamic Paint: * Added BKE_dynamicpaint.h to CMake list. * Changed some unsigned integers back to signed, as OpenMP can't iterate unsigned values anyway. * Fixed more warnings.
2011-11-10 13:00:27 +00:00
int total_points=surface->data->total_points;
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
unsigned int in_len;
unsigned char *out;
/* cache type */
ptcache_file_write(pf, &surface->type, 1, sizeof(int));
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (surface->type == MOD_DPAINT_SURFACE_T_PAINT) {
in_len = sizeof(PaintPoint) * total_points;
}
Dynamic Paint: * Added vertex weight painting. * Added dissolve/fade setting for every surface type. * Restored image format and displace type selection for image sequences. * Fixed a possible crash when OpenMP enabled. * Attempt to fix surface rna paths. * Fixed compile warnings.
2011-06-18 18:41:20 +00:00
else if (surface->type == MOD_DPAINT_SURFACE_T_DISPLACE ||
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
surface->type == MOD_DPAINT_SURFACE_T_WEIGHT)
{
in_len = sizeof(float) * total_points;
}
else if (surface->type == MOD_DPAINT_SURFACE_T_WAVE) {
in_len = sizeof(PaintWavePoint) * total_points;
}
else {
return 0;
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
* Added a new surface type "Waves" that simulates simple 2D wave motion. * Optimized adjacency point searching for vertex surfaces. * Separated adjacency code from effect system. * Overall code cleaning and tweaking.
2011-07-02 18:06:39 +00:00
out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len), "pointcache_lzo_buffer");
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)surface->data->type_data, in_len, out, cache_compress);
MEM_freeN(out);
}
return 1;
}
static int ptcache_dynamicpaint_read(PTCacheFile *pf, void *dp_v)
{
DynamicPaintSurface *surface = (DynamicPaintSurface*)dp_v;
char version[4];
/* version header */
style cleanup
2013-01-08 02:06:16 +00:00
ptcache_file_read(pf, version, 1, sizeof(char) * 4);
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (!STREQLEN(version, DPAINT_CACHE_VERSION, 4)) {
fix for possible out of bounds read, printing a warning with pointcache header.
2013-07-27 09:19:03 +00:00
printf("Dynamic Paint: Invalid cache version: '%c%c%c%c'!\n", UNPACK4(version));
Temp fix for stupid -Wformat-security issue. Also makes more use of BLI_snprintf, and makes more consistent code acrross all "header strings" generation in transform.c.
2013-03-12 13:18:39 +00:00
return 0;
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (surface->format != MOD_DPAINT_SURFACE_F_IMAGESEQ && surface->data) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
unsigned int data_len;
int surface_type;
/* cache type */
ptcache_file_read(pf, &surface_type, 1, sizeof(int));
if (surface_type != surface->type)
return 0;
/* read surface data */
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (surface->type == MOD_DPAINT_SURFACE_T_PAINT) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
data_len = sizeof(PaintPoint);
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
}
Dynamic Paint: * Added vertex weight painting. * Added dissolve/fade setting for every surface type. * Restored image format and displace type selection for image sequences. * Fixed a possible crash when OpenMP enabled. * Attempt to fix surface rna paths. * Fixed compile warnings.
2011-06-18 18:41:20 +00:00
else if (surface->type == MOD_DPAINT_SURFACE_T_DISPLACE ||
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
surface->type == MOD_DPAINT_SURFACE_T_WEIGHT)
{
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
data_len = sizeof(float);
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
}
else if (surface->type == MOD_DPAINT_SURFACE_T_WAVE) {
* Added a new surface type "Waves" that simulates simple 2D wave motion. * Optimized adjacency point searching for vertex surfaces. * Separated adjacency code from effect system. * Overall code cleaning and tweaking.
2011-07-02 18:06:39 +00:00
data_len = sizeof(PaintWavePoint);
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
}
else {
return 0;
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)surface->data->type_data, data_len*surface->data->total_points);
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
}
return 1;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
}
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
/* Rigid Body functions */
static int ptcache_rigidbody_write(int index, void *rb_v, void **data, int UNUSED(cfra))
{
RigidBodyWorld *rbw = rb_v;
Object *ob = NULL;
if (rbw->objects)
ob = rbw->objects[index];
if (ob && ob->rigidbody_object) {
RigidBodyOb *rbo = ob->rigidbody_object;
if (rbo->type == RBO_TYPE_ACTIVE) {
make bullet optional again
2013-01-23 07:26:39 +00:00
#ifdef WITH_BULLET
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
RB_body_get_position(rbo->physics_object, rbo->pos);
RB_body_get_orientation(rbo->physics_object, rbo->orn);
make bullet optional again
2013-01-23 07:26:39 +00:00
#endif
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
PTCACHE_DATA_FROM(data, BPHYS_DATA_LOCATION, rbo->pos);
PTCACHE_DATA_FROM(data, BPHYS_DATA_ROTATION, rbo->orn);
}
}
return 1;
}
static void ptcache_rigidbody_read(int index, void *rb_v, void **data, float UNUSED(cfra), float *old_data)
{
RigidBodyWorld *rbw = rb_v;
Object *ob = NULL;
if (rbw->objects)
ob = rbw->objects[index];
if (ob && ob->rigidbody_object) {
RigidBodyOb *rbo = ob->rigidbody_object;
if (rbo->type == RBO_TYPE_ACTIVE) {
if (old_data) {
memcpy(rbo->pos, data, 3 * sizeof(float));
memcpy(rbo->orn, data + 3, 4 * sizeof(float));
}
else {
PTCACHE_DATA_TO(data, BPHYS_DATA_LOCATION, 0, rbo->pos);
PTCACHE_DATA_TO(data, BPHYS_DATA_ROTATION, 0, rbo->orn);
}
}
}
}
static void ptcache_rigidbody_interpolate(int index, void *rb_v, void **data, float cfra, float cfra1, float cfra2, float *old_data)
{
RigidBodyWorld *rbw = rb_v;
Object *ob = NULL;
if (rbw->objects)
ob = rbw->objects[index];
if (ob && ob->rigidbody_object) {
RigidBodyOb *rbo = ob->rigidbody_object;
if (rbo->type == RBO_TYPE_ACTIVE) {
Fix T42294: Bullet rigidbody point cache reading was using uninitialized key values for velocity.
2014-10-29 14:59:29 +01:00
ParticleKey keys[4];
ParticleKey result;
float dfra;
memset(keys, 0, sizeof(keys));
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
copy_v3_v3(keys[1].co, rbo->pos);
Fix for rigidbody treating the quat as a v3 with pointcache
2014-04-11 18:16:48 +10:00
copy_qt_qt(keys[1].rot, rbo->orn);
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
if (old_data) {
memcpy(keys[2].co, data, 3 * sizeof(float));
memcpy(keys[2].rot, data + 3, 4 * sizeof(float));
}
else {
Fix T42294: Bullet rigidbody point cache reading was using uninitialized key values for velocity.
2014-10-29 14:59:29 +01:00
BKE_ptcache_make_particle_key(&keys[2], 0, data, cfra2);
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
}
dfra = cfra2 - cfra1;
Fix T42294: Bullet rigidbody point cache reading was using uninitialized key values for velocity.
2014-10-29 14:59:29 +01:00
/* note: keys[0] and keys[3] unused for type < 1 (crappy) */
psys_interpolate_particle(-1, keys, (cfra - cfra1) / dfra, &result, true);
interp_qt_qtqt(result.rot, keys[1].rot, keys[2].rot, (cfra - cfra1) / dfra);
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
Fix T42294: Bullet rigidbody point cache reading was using uninitialized key values for velocity.
2014-10-29 14:59:29 +01:00
copy_v3_v3(rbo->pos, result.co);
copy_qt_qt(rbo->orn, result.rot);
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
}
}
}
static int ptcache_rigidbody_totpoint(void *rb_v, int UNUSED(cfra))
{
RigidBodyWorld *rbw = rb_v;
return rbw->numbodies;
}
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
static void ptcache_rigidbody_error(void *UNUSED(rb_v), const char *UNUSED(message))
{
/* ignored for now */
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
/* Creating ID's */
void BKE_ptcache_id_from_softbody(PTCacheID *pid, Object *ob, SoftBody *sb)
{
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= sb;
pid->type= PTCACHE_TYPE_SOFTBODY;
pid->cache= sb->pointcache;
pid->cache_ptr= &sb->pointcache;
pid->ptcaches= &sb->ptcaches;
pid->totpoint= pid->totwrite= ptcache_softbody_totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_softbody_error;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_point = ptcache_softbody_write;
pid->read_point = ptcache_softbody_read;
pid->interpolate_point = ptcache_softbody_interpolate;
pid->write_stream = NULL;
pid->read_stream = NULL;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = NULL;
pid->read_openvdb_stream = NULL;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->data_types= (1<<BPHYS_DATA_LOCATION) | (1<<BPHYS_DATA_VELOCITY);
pid->info_types= 0;
pid->stack_index = pid->cache->index;
Fix part of #29310: secondary point caches for cloth did not get steps set to correct default, was 10 when supposed to be 1.
2011-11-24 20:46:00 +00:00
pid->default_step = 10;
pid->max_step = 20;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = PTCACHE_FILE_PTCACHE;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
void BKE_ptcache_id_from_particles(PTCacheID *pid, Object *ob, ParticleSystem *psys)
{
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= psys;
pid->type= PTCACHE_TYPE_PARTICLES;
pid->stack_index= psys->pointcache->index;
pid->cache= psys->pointcache;
pid->cache_ptr= &psys->pointcache;
pid->ptcaches= &psys->ptcaches;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->type != PART_HAIR)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->flag |= PTCACHE_VEL_PER_SEC;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->totpoint = ptcache_particle_totpoint;
pid->totwrite = ptcache_particle_totwrite;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_particle_error;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_point = ptcache_particle_write;
pid->read_point = ptcache_particle_read;
pid->interpolate_point = ptcache_particle_interpolate;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_stream = NULL;
pid->read_stream = NULL;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = NULL;
pid->read_openvdb_stream = NULL;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
pid->data_types = (1<<BPHYS_DATA_LOCATION) | (1<<BPHYS_DATA_VELOCITY) | (1<<BPHYS_DATA_INDEX);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->phystype == PART_PHYS_BOIDS)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->data_types|= (1<<BPHYS_DATA_AVELOCITY) | (1<<BPHYS_DATA_ROTATION) | (1<<BPHYS_DATA_BOIDS);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (psys->part->phystype == PART_PHYS_FLUID && psys->part->fluid && psys->part->fluid->flag & SPH_VISCOELASTIC_SPRINGS) {
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
pid->write_extra_data = ptcache_particle_extra_write;
pid->read_extra_data = ptcache_particle_extra_read;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->flag & PART_ROTATIONS) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->data_types|= (1<<BPHYS_DATA_ROTATION);
style cleanup: multi-line if statements.
2012-04-21 15:11:03 +00:00
if (psys->part->rotmode != PART_ROT_VEL ||
psys->part->avemode == PART_AVE_RAND ||
style cleanyp
2012-06-23 23:22:19 +00:00
psys->part->avefac != 0.0f)
style cleanup: multi-line if statements.
2012-04-21 15:11:03 +00:00
{
pid->data_types |= (1 << BPHYS_DATA_AVELOCITY);
}
"Fix" for [#30098] Particle rotation wrong / explode modifier - The main problem was that in order to be accurate all particle rotations have to be calculated incrementally so the only working solution is to store rotations to the point cache (previously this was only done for dynamic rotations). This can nearly double the point cache size so it's not ideal to have this as a default as in many cases you don't care about particle rotations. - Particle rotation panel now has a new "enable" checkbox that enables rotation calculations and the storing of rotations to point cache. - Old files will have rotations enabled via do_versions so that in the worst case old files will only get bigger point caches, but no sudden loss of particle rotations.
2012-03-18 21:33:00 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->info_types= (1<<BPHYS_DATA_TIMES);
Fix part of #29310: secondary point caches for cloth did not get steps set to correct default, was 10 when supposed to be 1.
2011-11-24 20:46:00 +00:00
pid->default_step = 10;
pid->max_step = 20;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = PTCACHE_FILE_PTCACHE;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
void BKE_ptcache_id_from_cloth(PTCacheID *pid, Object *ob, ClothModifierData *clmd)
{
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= clmd;
pid->type= PTCACHE_TYPE_CLOTH;
pid->stack_index= clmd->point_cache->index;
pid->cache= clmd->point_cache;
pid->cache_ptr= &clmd->point_cache;
pid->ptcaches= &clmd->ptcaches;
pid->totpoint= pid->totwrite= ptcache_cloth_totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_cloth_error;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_point = ptcache_cloth_write;
pid->read_point = ptcache_cloth_read;
pid->interpolate_point = ptcache_cloth_interpolate;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = NULL;
pid->read_openvdb_stream = NULL;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_stream = NULL;
pid->read_stream = NULL;
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->data_types= (1<<BPHYS_DATA_LOCATION) | (1<<BPHYS_DATA_VELOCITY) | (1<<BPHYS_DATA_XCONST);
pid->info_types= 0;
Fix part of #29310: secondary point caches for cloth did not get steps set to correct default, was 10 when supposed to be 1.
2011-11-24 20:46:00 +00:00
pid->default_step = 1;
pid->max_step = 1;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = PTCACHE_FILE_PTCACHE;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
void BKE_ptcache_id_from_smoke(PTCacheID *pid, struct Object *ob, struct SmokeModifierData *smd)
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
{
SmokeDomainSettings *sds = smd->domain;
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= smd;
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
pid->type= PTCACHE_TYPE_SMOKE_DOMAIN;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
pid->stack_index= sds->point_cache[0]->index;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
pid->cache= sds->point_cache[0];
pid->cache_ptr= &(sds->point_cache[0]);
pid->ptcaches= &(sds->ptcaches[0]);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->totpoint= pid->totwrite= ptcache_smoke_totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_smoke_error;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_point = NULL;
pid->read_point = NULL;
pid->interpolate_point = NULL;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->read_stream = ptcache_smoke_read;
pid->write_stream = ptcache_smoke_write;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = ptcache_smoke_openvdb_write;
pid->read_openvdb_stream = ptcache_smoke_openvdb_read;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
pid->data_types= 0;
Smoke: *Bugfix for crash on using a plane as smoke domain (reported by DingTo) * Bringing slowly high res back, not yet working
2009-08-25 23:39:49 +00:00
pid->info_types= 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->fluid)
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
pid->data_types |= (1<<BPHYS_DATA_SMOKE_LOW);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (sds->wt)
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
pid->data_types |= (1<<BPHYS_DATA_SMOKE_HIGH);
Fix part of #29310: secondary point caches for cloth did not get steps set to correct default, was 10 when supposed to be 1.
2011-11-24 20:46:00 +00:00
pid->default_step = 1;
pid->max_step = 1;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = smd->domain->cache_file_format;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
}
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
void BKE_ptcache_id_from_dynamicpaint(PTCacheID *pid, Object *ob, DynamicPaintSurface *surface)
{
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= surface;
pid->type= PTCACHE_TYPE_DYNAMICPAINT;
pid->cache= surface->pointcache;
pid->cache_ptr= &surface->pointcache;
pid->ptcaches= &surface->ptcaches;
pid->totpoint= pid->totwrite= ptcache_dynamicpaint_totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_dynamicpaint_error;
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
pid->write_point = NULL;
pid->read_point = NULL;
pid->interpolate_point = NULL;
pid->write_stream = ptcache_dynamicpaint_write;
pid->read_stream = ptcache_dynamicpaint_read;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = NULL;
pid->read_openvdb_stream = NULL;
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
pid->data_types= BPHYS_DATA_DYNAMICPAINT;
pid->info_types= 0;
pid->stack_index = pid->cache->index;
Fix part of #29310: secondary point caches for cloth did not get steps set to correct default, was 10 when supposed to be 1.
2011-11-24 20:46:00 +00:00
pid->default_step = 1;
pid->max_step = 1;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = PTCACHE_FILE_PTCACHE;
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
}
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
void BKE_ptcache_id_from_rigidbody(PTCacheID *pid, Object *ob, RigidBodyWorld *rbw)
{
memset(pid, 0, sizeof(PTCacheID));
pid->ob= ob;
pid->calldata= rbw;
pid->type= PTCACHE_TYPE_RIGIDBODY;
pid->cache= rbw->pointcache;
pid->cache_ptr= &rbw->pointcache;
pid->ptcaches= &rbw->ptcaches;
pid->totpoint= pid->totwrite= ptcache_rigidbody_totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
pid->error = ptcache_rigidbody_error;
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
pid->write_point = ptcache_rigidbody_write;
pid->read_point = ptcache_rigidbody_read;
pid->interpolate_point = ptcache_rigidbody_interpolate;
pid->write_stream = NULL;
pid->read_stream = NULL;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->write_openvdb_stream = NULL;
pid->read_openvdb_stream = NULL;
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
pid->write_extra_data = NULL;
pid->read_extra_data = NULL;
pid->interpolate_extra_data = NULL;
pid->write_header = ptcache_basic_header_write;
pid->read_header = ptcache_basic_header_read;
pid->data_types= (1<<BPHYS_DATA_LOCATION) | (1<<BPHYS_DATA_ROTATION);
pid->info_types= 0;
pid->stack_index = pid->cache->index;
pid->default_step = 1;
pid->max_step = 1;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
pid->file_type = PTCACHE_FILE_PTCACHE;
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
}
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
void BKE_ptcache_ids_from_object(Main *bmain, ListBase *lb, Object *ob, Scene *scene, int duplis)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
PTCacheID *pid;
ParticleSystem *psys;
ModifierData *md;
lb->first= lb->last= NULL;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (ob->soft) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_softbody(pid, ob, ob->soft);
BLI_addtail(lb, pid);
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (psys=ob->particlesystem.first; psys; psys=psys->next) {
if (psys->part==NULL)
Fix for [#24031] Baked Physics with Phsics set to NO can't be freed * Should fix the cause ("no physics" gets baked) and old files that are effected (ui allows freeing if cache isn't really used)
2010-09-28 08:47:59 +00:00
continue;
/* check to make sure point cache is actually used by the particles */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (ELEM(psys->part->phystype, PART_PHYS_NO, PART_PHYS_KEYED))
Fix for [#24031] Baked Physics with Phsics set to NO can't be freed * Should fix the cause ("no physics" gets baked) and old files that are effected (ui allows freeing if cache isn't really used)
2010-09-28 08:47:59 +00:00
continue;
Fix for [#24118] Hair particles can not be edited * Own mistake from a previous fix.
2010-10-04 08:48:50 +00:00
/* hair needs to be included in id-list for cache edit mode to work */
style cleanup: follow style guide for/with/if spacing
2012-03-24 07:52:14 +00:00
/* if (psys->part->type == PART_HAIR && (psys->flag & PSYS_HAIR_DYNAMICS)==0) */
Fix for [#24118] Hair particles can not be edited * Own mistake from a previous fix.
2010-10-04 08:48:50 +00:00
/* continue; */
Fix for [#24031] Baked Physics with Phsics set to NO can't be freed * Should fix the cause ("no physics" gets baked) and old files that are effected (ui allows freeing if cache isn't really used)
2010-09-28 08:47:59 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->type == PART_FLUID)
Fix for [#24031] Baked Physics with Phsics set to NO can't be freed * Should fix the cause ("no physics" gets baked) and old files that are effected (ui allows freeing if cache isn't really used)
2010-09-28 08:47:59 +00:00
continue;
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_particles(pid, ob, psys);
BLI_addtail(lb, pid);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (md=ob->modifiers.first; md; md=md->next) {
if (md->type == eModifierType_Cloth) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_cloth(pid, ob, (ClothModifierData*)md);
BLI_addtail(lb, pid);
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (md->type == eModifierType_Smoke) {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
SmokeModifierData *smd = (SmokeModifierData *)md;
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (smd->type & MOD_SMOKE_TYPE_DOMAIN) {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
BKE_ptcache_id_from_smoke(pid, ob, (SmokeModifierData*)md);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
BLI_addtail(lb, pid);
}
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (md->type == eModifierType_DynamicPaint) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
DynamicPaintModifierData *pmd = (DynamicPaintModifierData *)md;
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (pmd->canvas) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
DynamicPaintSurface *surface = pmd->canvas->surfaces.first;
for (; surface; surface=surface->next) {
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_dynamicpaint(pid, ob, surface);
BLI_addtail(lb, pid);
}
}
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
if (scene && ob->rigidbody_object && scene->rigidbody_world) {
pid = MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_rigidbody(pid, ob, scene->rigidbody_world);
BLI_addtail(lb, pid);
}
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (scene && (duplis-- > 0) && (ob->transflag & OB_DUPLI)) {
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
ListBase *lb_dupli_ob;
style cleanup: some warnigs & spelling.
2013-02-06 14:02:19 +00:00
/* don't update the dupli groups, we only want their pid's */
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
if ((lb_dupli_ob = object_duplilist_ex(bmain, bmain->eval_ctx, scene, ob, false))) {
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
DupliObject *dob;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (dob= lb_dupli_ob->first; dob; dob= dob->next) {
if (dob->ob != ob) { /* avoids recursive loops with dupliframes: bug 22988 */
bugfix [#22988] Duplicating Lights by Frame Causes Hangup
2010-07-24 02:51:51 +00:00
ListBase lb_dupli_pid;
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
BKE_ptcache_ids_from_object(bmain, &lb_dupli_pid, dob->ob, scene, duplis);
rename addlisttolist() to BLI_movelisttolist() name was misleading because the list items were removed from the source list. (no functional changes)
2010-12-21 14:49:34 +00:00
BLI_movelisttolist(lb, &lb_dupli_pid);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (lb_dupli_pid.first)
bugfix [#22988] Duplicating Lights by Frame Causes Hangup
2010-07-24 02:51:51 +00:00
printf("Adding Dupli\n");
}
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
}
free_object_duplilist(lb_dupli_ob); /* does restore */
}
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* File handling */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
static const char *ptcache_file_extension(const PTCacheID *pid)
{
switch (pid->file_type) {
default:
case PTCACHE_FILE_PTCACHE:
return PTCACHE_EXT;
case PTCACHE_FILE_OPENVDB:
return ".vdb";
}
}
/**
* Similar to #BLI_path_frame_get, but takes into account the stack-index which is after the frame.
*/
static int ptcache_frame_from_filename(const char *filename, const char *ext)
{
const int frame_len = 6;
const int ext_len = frame_len + strlen(ext);
const int len = strlen(filename);
/* could crash if trying to copy a string out of this range */
if (len > ext_len) {
/* using frame_len here gives compile error (vla) */
char num[/* frame_len */6 + 1];
BLI_strncpy(num, filename + len - ext_len, sizeof(num));
return atoi(num);
}
return -1;
}
style cleanup / comment formatting for bli/bke/bmesh
2012-03-03 20:19:11 +00:00
/* Takes an Object ID and returns a unique name
* - id: object id
* - cfra: frame for the cache, can be negative
Fix for [#34898] Typo in error message of mathutils.Vector * Also fixed some more cases of "more then" -> "more than".
2013-04-07 15:09:06 +00:00
* - stack_index: index in the modifier stack. we can have cache for more than one stack_index
style cleanup / comment formatting for bli/bke/bmesh
2012-03-03 20:19:11 +00:00
*/
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
Fix for part of bug #17759: point cache clearing did too much, also clearing other physics systems on the same object.
2008-10-04 21:26:48 +00:00
#define MAX_PTCACHE_PATH FILE_MAX
style cleanup
2013-01-08 02:06:16 +00:00
#define MAX_PTCACHE_FILE (FILE_MAX * 2)
Fix for part of bug #17759: point cache clearing did too much, also clearing other physics systems on the same object.
2008-10-04 21:26:48 +00:00
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
static int ptcache_path(PTCacheID *pid, char *filename)
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
{
style cleanup
2013-03-31 03:28:46 +00:00
Library *lib = (pid->ob) ? pid->ob->id.lib : NULL;
Cleanup: use new accessors to blendfile path (Main.name).
2018-06-05 15:10:33 +02:00
const char *blendfilename= (lib && (pid->cache->flag & PTCACHE_IGNORE_LIBPATH)==0) ? lib->filepath: BKE_main_blendfile_path_from_global();
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
size_t i;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_EXTERNAL) {
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
strcpy(filename, pid->cache->path);
pointcache support for relative external paths with the useual // prefix as well as library path option.
2010-07-14 07:47:03 +00:00
code cleanup: add utility function BLI_path_is_rel()
2012-08-29 10:32:38 +00:00
if (BLI_path_is_rel(filename)) {
pointcache support for relative external paths with the useual // prefix as well as library path option.
2010-07-14 07:47:03 +00:00
BLI_path_abs(filename, blendfilename);
code cleanup: add utility function BLI_path_is_rel()
2012-08-29 10:32:38 +00:00
}
pointcache support for relative external paths with the useual // prefix as well as library path option.
2010-07-14 07:47:03 +00:00
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
return BLI_add_slash(filename); /* new strlen() */
}
else if (G.relbase_valid || lib) {
spelling cleanup
2012-03-18 07:38:51 +00:00
char file[MAX_PTCACHE_PATH]; /* we don't want the dir, only the file */
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
- add convenience functions BLI_split_dir_part / BLI_split_file_part, which just call BLI_split_dirfile(). - add a fixed value for bprogdir (the dir of bprogname), since it was being used for resource lookups.
2011-10-21 02:13:36 +00:00
BLI_split_file_part(blendfilename, file, sizeof(file));
Pointcache for unsaved files is saved in /tmp/blendcache_untitled/... instead of not generating pointcache at all, be careful since the temp path can be NULL or missing. This is messy because the files can be left there when blender quits, however saveBuffers and sessions alredy do :/ Some cleanup function needs to Blender that deals with this!
2008-02-12 17:11:45 +00:00
i = strlen(file);
/* remove .blend */
if (i > 6)
file[i-6] = '\0';
Fixes for snprintf usage: * replace by BLI_snprintf in various places, note _snprintf on windows does not properly null terminate the string. * fix overflow in sequencer proxy code due to buffer being smaller than specified size. * fix some usage of snprintf as strcpy, this is will go wrong if the string contains % characters. * remove BLI_dynstr_printf function in gpu module, use BLI_dynstr_appendf
2011-08-30 10:07:50 +00:00
BLI_snprintf(filename, MAX_PTCACHE_PATH, "//"PTCACHE_PATH"%s", file); /* add blend file name to pointcache dir */
rename some functions to use easier to understand names. 'BLI_makestringcode' --> 'BLI_path_rel' 'BLI_convertstringcwd' --> 'BLI_path_cwd' 'BLI_convertstringframe' --> 'BLI_path_frame' 'BLI_convertstringframe_range' --> 'BLI_path_frame_range' 'BLI_make_cwdpath' --> 'BLI_path_cwd'
2010-03-09 17:36:23 +00:00
BLI_path_abs(filename, blendfilename);
- Support for importing python packages. (directories of python scripts containing an __init__.py) - BLI_add_slash returns the new string length. - BLI_where_am_i() would often have /./ in the path (not incorrect but annoying, got into python exceptions) - release/ui/space_image.py, py error referencing invalid keyword args.
2009-07-19 17:45:14 +00:00
return BLI_add_slash(filename); /* new strlen() */
Pointcache for unsaved files is saved in /tmp/blendcache_untitled/... instead of not generating pointcache at all, be careful since the temp path can be NULL or missing. This is messy because the files can be left there when blender quits, however saveBuffers and sessions alredy do :/ Some cleanup function needs to Blender that deals with this!
2008-02-12 17:11:45 +00:00
}
anim.c, buttons_object.c, readfile.c, BKE_blender.h - dupliFace scale option, needed for leaves. modifier.c, BKE_modifier.h - flag for modifiers to say they use pointcache, also new func modifiers_usesPointCache renamed //pointcache to //blendcache_blendfilename so blendfiles in the same dir dont conflict, and other to show this dir isnt limited to pointcache only (nodes way want to use this) wizard_curve2tree.py - better defaults for pretty tree's
2007-11-30 10:38:59 +00:00
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
/* use the temp path. this is weak but better then not using point cache at all */
Code cleanup: don't use btempdir/bprogdir/bprogname globals anymore, but wrap in BLI_ functions.
2011-10-21 17:37:38 +00:00
/* temporary directory is assumed to exist and ALWAYS has a trailing slash */
Refactor: BLI_path_util (part 2) Use BKE_appdir/tempdir naming prefix for functions extracted from BLI_path_util
2014-11-23 15:54:29 +01:00
BLI_snprintf(filename, MAX_PTCACHE_PATH, "%s"PTCACHE_PATH, BKE_tempdir_session());
- Support for importing python packages. (directories of python scripts containing an __init__.py) - BLI_add_slash returns the new string length. - BLI_where_am_i() would often have /./ in the path (not incorrect but annoying, got into python exceptions) - release/ui/space_image.py, py error referencing invalid keyword args.
2009-07-19 17:45:14 +00:00
return BLI_add_slash(filename); /* new strlen() */
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_filename(PTCacheID *pid, char *filename, int cfra, short do_path, short do_ext)
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
{
int len=0;
char *idname;
char *newname;
filename[0] = '\0';
newname = filename;
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
if (!G.relbase_valid && (pid->cache->flag & PTCACHE_EXTERNAL)==0) return 0; /* save blend file before using disk pointcache */
anim.c, buttons_object.c, readfile.c, BKE_blender.h - dupliFace scale option, needed for leaves. modifier.c, BKE_modifier.h - flag for modifiers to say they use pointcache, also new func modifiers_usesPointCache renamed //pointcache to //blendcache_blendfilename so blendfiles in the same dir dont conflict, and other to show this dir isnt limited to pointcache only (nodes way want to use this) wizard_curve2tree.py - better defaults for pretty tree's
2007-11-30 10:38:59 +00:00
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
/* start with temp dir */
if (do_path) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
len = ptcache_path(pid, filename);
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
newname += len;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->name[0] == '\0' && (pid->cache->flag & PTCACHE_EXTERNAL)==0) {
code cleanup
2013-03-18 18:25:05 +00:00
idname = (pid->ob->id.name + 2);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* convert chars to hex so they are always a valid filename */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while ('\0' != *idname) {
Code cleanup: quiet warnings & style
2014-04-13 11:26:31 +10:00
BLI_snprintf(newname, MAX_PTCACHE_FILE, "%02X", (unsigned int)(*idname++));
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
newname+=2;
len += 2;
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
else {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
int temp = (int)strlen(pid->cache->name);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
strcpy(newname, pid->cache->name);
newname+=temp;
len += temp;
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
if (do_ext) {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->index < 0)
code cleanup: function naming, use BKE_*type* prefix.
2012-05-05 14:03:12 +00:00
pid->cache->index = pid->stack_index = BKE_object_insert_ptcache(pid->ob);
Pointcache: *introducing unique ID's following brechts hint from ML Enhancements resulting from this: * multiple caches per modifier stack position
2009-08-25 18:41:36 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const char *ext = ptcache_file_extension(pid);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_EXTERNAL) {
if (pid->cache->index >= 0)
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_snprintf(newname, MAX_PTCACHE_FILE, "_%06d_%02u%s", cfra, pid->stack_index, ext); /* always 6 chars */
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
else
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_snprintf(newname, MAX_PTCACHE_FILE, "_%06d%s", cfra, ext); /* always 6 chars */
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
}
else {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_snprintf(newname, MAX_PTCACHE_FILE, "_%06d_%02u%s", cfra, pid->stack_index, ext); /* always 6 chars */
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
len += 16;
}
return len; /* make sure the above string is always 16 chars */
}
/* youll need to close yourself after! */
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
static PTCacheFile *ptcache_file_open(PTCacheID *pid, int mode, int cfra)
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
{
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
PTCacheFile *pf;
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
FILE *fp = NULL;
style cleanup
2013-01-08 02:06:16 +00:00
char filename[FILE_MAX * 2];
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
#ifndef DURIAN_POINTCACHE_LIB_OK
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
/* don't allow writing for linked objects */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->ob->id.lib && mode == PTCACHE_FILE_WRITE)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return NULL;
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
#endif
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
if (!G.relbase_valid && (pid->cache->flag & PTCACHE_EXTERNAL)==0) return NULL; /* save blend file before using disk pointcache */
anim.c, buttons_object.c, readfile.c, BKE_blender.h - dupliFace scale option, needed for leaves. modifier.c, BKE_modifier.h - flag for modifiers to say they use pointcache, also new func modifiers_usesPointCache renamed //pointcache to //blendcache_blendfilename so blendfiles in the same dir dont conflict, and other to show this dir isnt limited to pointcache only (nodes way want to use this) wizard_curve2tree.py - better defaults for pretty tree's
2007-11-30 10:38:59 +00:00
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_filename(pid, filename, cfra, 1, 1);
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
if (mode==PTCACHE_FILE_READ) {
Adds support for utf paths on Windows. Not all file formats/calls are supported yet. It will be expended. Please from now on use BLI_fopen, BLI_* for file manipulations. For non-windows systems BLI_fopen just calls fopen. For Windows, the utf-8 string is translated to utf-16 string in order to call UTF version of the function.
2012-03-20 02:17:37 +00:00
fp = BLI_fopen(filename, "rb");
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else if (mode==PTCACHE_FILE_WRITE) {
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
BLI_make_existing_file(filename); /* will create the dir if needs be, same as //textures is created */
Adds support for utf paths on Windows. Not all file formats/calls are supported yet. It will be expended. Please from now on use BLI_fopen, BLI_* for file manipulations. For non-windows systems BLI_fopen just calls fopen. For Windows, the utf-8 string is translated to utf-16 string in order to call UTF version of the function.
2012-03-20 02:17:37 +00:00
fp = BLI_fopen(filename, "wb");
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else if (mode==PTCACHE_FILE_UPDATE) {
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
BLI_make_existing_file(filename);
Adds support for utf paths on Windows. Not all file formats/calls are supported yet. It will be expended. Please from now on use BLI_fopen, BLI_* for file manipulations. For non-windows systems BLI_fopen just calls fopen. For Windows, the utf-8 string is translated to utf-16 string in order to call UTF version of the function.
2012-03-20 02:17:37 +00:00
fp = BLI_fopen(filename, "rb+");
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
converted more mixed tab/space indentations to tabs. only whitespace changes.
2011-04-21 15:53:30 +00:00
if (!fp)
return NULL;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
pf= MEM_mallocN(sizeof(PTCacheFile), "PTCacheFile");
pf->fp= fp;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pf->old_format = 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->frame = cfra;
converted more mixed tab/space indentations to tabs. only whitespace changes.
2011-04-21 15:53:30 +00:00
return pf;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
static void ptcache_file_close(PTCacheFile *pf)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
fclose(pf->fp);
MEM_freeN(pf);
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static int ptcache_file_compressed_read(PTCacheFile *pf, unsigned char *result, unsigned int len)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
{
int r = 0;
unsigned char compressed = 0;
size_t in_len;
#ifdef WITH_LZO
size_t out_len = len;
#endif
unsigned char *in;
style cleanup
2013-01-08 02:06:16 +00:00
unsigned char *props = MEM_callocN(16 * sizeof(char), "tmp");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_read(pf, &compressed, 1, sizeof(unsigned char));
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (compressed) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int size;
ptcache_file_read(pf, &size, 1, sizeof(unsigned int));
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
in_len = (size_t)size;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (in_len==0) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* do nothing */
}
else {
in = (unsigned char *)MEM_callocN(sizeof(unsigned char)*in_len, "pointcache_compressed_buffer");
ptcache_file_read(pf, in, in_len, sizeof(unsigned char));
#ifdef WITH_LZO
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (compressed == 1)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
r = lzo1x_decompress_safe(in, (lzo_uint)in_len, result, (lzo_uint *)&out_len, NULL);
#endif
#ifdef WITH_LZMA
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (compressed == 2) {
quiet warning & minor pep8 edits.
2011-04-04 22:42:43 +00:00
size_t sizeOfIt;
Fixed compile error when building with WITH_LZMA and not WITH_LZO for pointcache compression.
2012-04-09 01:24:56 +00:00
size_t leni = in_len, leno = len;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_read(pf, &size, 1, sizeof(unsigned int));
sizeOfIt = (size_t)size;
ptcache_file_read(pf, props, sizeOfIt, sizeof(unsigned char));
r = LzmaUncompress(result, &leno, in, &leni, props, sizeOfIt);
}
#endif
MEM_freeN(in);
}
}
else {
ptcache_file_read(pf, result, len, sizeof(unsigned char));
}
MEM_freeN(props);
return r;
}
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static int ptcache_file_compressed_write(PTCacheFile *pf, unsigned char *in, unsigned int in_len, unsigned char *out, int mode)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
{
int r = 0;
unsigned char compressed = 0;
size_t out_len= 0;
style cleanup
2013-01-08 02:06:16 +00:00
unsigned char *props = MEM_callocN(16 * sizeof(char), "tmp");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
size_t sizeOfIt = 5;
(void)mode; /* unused when building w/o compression */
#ifdef WITH_LZO
out_len= LZO_OUT_LEN(in_len);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == 1) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
LZO_HEAP_ALLOC(wrkmem, LZO1X_MEM_COMPRESS);
style cleanup: trailing tabs & expand some non prefix tabs into spaces.
2012-10-21 05:46:41 +00:00
r = lzo1x_1_compress(in, (lzo_uint)in_len, out, (lzo_uint *)&out_len, wrkmem);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
if (!(r == LZO_E_OK) || (out_len >= in_len))
compressed = 0;
else
compressed = 1;
}
#endif
#ifdef WITH_LZMA
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == 2) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: whitespace / commas
2012-04-29 15:47:02 +00:00
r = LzmaCompress(out, &out_len, in, in_len, //assume sizeof(char)==1....
Temp fix for stupid -Wformat-security issue. Also makes more use of BLI_snprintf, and makes more consistent code acrross all "header strings" generation in transform.c.
2013-03-12 13:18:39 +00:00
props, &sizeOfIt, 5, 1 << 24, 3, 0, 2, 32, 2);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!(r == SZ_OK) || (out_len >= in_len))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
compressed = 0;
else
compressed = 2;
}
#endif
ptcache_file_write(pf, &compressed, 1, sizeof(unsigned char));
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (compressed) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int size = out_len;
ptcache_file_write(pf, &size, 1, sizeof(unsigned int));
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_write(pf, out, out_len, sizeof(unsigned char));
}
else
ptcache_file_write(pf, in, in_len, sizeof(unsigned char));
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (compressed == 2) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int size = sizeOfIt;
ptcache_file_write(pf, &sizeOfIt, 1, sizeof(unsigned int));
ptcache_file_write(pf, props, size, sizeof(unsigned char));
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
MEM_freeN(props);
return r;
}
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static int ptcache_file_read(PTCacheFile *pf, void *f, unsigned int tot, unsigned int size)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return (fread(f, size, tot, pf->fp) == tot);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
quiet some warnings from recent merge
2011-11-10 12:28:26 +00:00
static int ptcache_file_write(PTCacheFile *pf, const void *f, unsigned int tot, unsigned int size)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return (fwrite(f, size, tot, pf->fp) == tot);
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_file_data_read(PTCacheFile *pf)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int i;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
if ((pf->data_types & (1<<i)) && !ptcache_file_read(pf, pf->cur[i], 1, ptcache_data_size[i]))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 1;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_file_data_write(PTCacheFile *pf)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int i;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
if ((pf->data_types & (1<<i)) && !ptcache_file_write(pf, pf->cur[i], 1, ptcache_data_size[i]))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
Keyed physics refresh: - Keyed targets in one list instead of "chaining", this opens up many more possibilities than before and is much less obscure. - Better keyed timing possibilities (time & duration in frames). - Looping over keyed targets list. Other changes: - New child setting "length" with threshold (great for guard & underfur with a single particle system) - Modularization of path interpolation code. - Cleared "animateable" flags from many particle settings that shouldn't be animateable. Fixes: - Keyed particles weren't copied properly (ancient bug). - Hair rotations depended on global z-axis for root rotation so downward facing strands could flip rotation randomly. Now initial hair rotation is derived from face dependent hair matrix. (This caused for example ugly flipping of child strands on some cases). - Children from faces weren't calculated straight after activating them. - Multiple disk cache fixes: * Disk cache didn't work correctly with frame steps. * Conversion from memory cache to disk cache didn't work with cloth. * Disk cache crashed on some frames trying to close an already closed cache file. * Trails didn't work with disk cached particles. - Child rough effects were effected by emitter object loc/rot making them next to useless with animation, why didn't anybody tell me this!! - Lots of random code cleanup.
2009-07-12 23:38:47 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 1;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_file_header_begin_read(PTCacheFile *pf)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int typeflag=0;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int error=0;
char bphysics[8];
pf->data_types = 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (fread(bphysics, sizeof(char), 8, pf->fp) != 8)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
error = 1;
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (!error && !STREQLEN(bphysics, "BPHYSICS", 8))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
error = 1;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && !fread(&typeflag, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
error = 1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->type = (typeflag & PTCACHE_TYPEFLAG_TYPEMASK);
pf->flag = (typeflag & PTCACHE_TYPEFLAG_FLAGMASK);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* if there was an error set file as it was */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (error)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
fseek(pf->fp, 0, SEEK_SET);
return !error;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_file_header_begin_write(PTCacheFile *pf)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
Enabled GCC -Wwrite-strings warning for CMake and replaced many 'char's for 'const char's,. Only one functional change where Transform orientations passed "" to BIF_createTransformOrientation() which could then have the value written into.
2010-12-03 17:05:21 +00:00
const char *bphysics = "BPHYSICS";
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int typeflag = pf->type + pf->flag;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (fwrite(bphysics, sizeof(char), 8, pf->fp) != 8)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!fwrite(&typeflag, sizeof(unsigned int), 1, pf->fp))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 0;
return 1;
}
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Data pointer handling */
int BKE_ptcache_data_size(int data_type)
{
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
return ptcache_data_size[data_type];
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static void ptcache_file_pointers_init(PTCacheFile *pf)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int data_types = pf->data_types;
Possible fix for the issue that came up in [#24890] Vector Blur node is Buggy * Apparently some compilers don't respect proper operator precedence, so added some parentheses around to make inline conditionals unambiguous.
2010-11-25 11:38:55 +00:00
pf->cur[BPHYS_DATA_INDEX] = (data_types & (1<<BPHYS_DATA_INDEX)) ? &pf->data.index : NULL;
pf->cur[BPHYS_DATA_LOCATION] = (data_types & (1<<BPHYS_DATA_LOCATION)) ? &pf->data.loc : NULL;
pf->cur[BPHYS_DATA_VELOCITY] = (data_types & (1<<BPHYS_DATA_VELOCITY)) ? &pf->data.vel : NULL;
pf->cur[BPHYS_DATA_ROTATION] = (data_types & (1<<BPHYS_DATA_ROTATION)) ? &pf->data.rot : NULL;
pf->cur[BPHYS_DATA_AVELOCITY] = (data_types & (1<<BPHYS_DATA_AVELOCITY))? &pf->data.ave : NULL;
pf->cur[BPHYS_DATA_SIZE] = (data_types & (1<<BPHYS_DATA_SIZE)) ? &pf->data.size : NULL;
pf->cur[BPHYS_DATA_TIMES] = (data_types & (1<<BPHYS_DATA_TIMES)) ? &pf->data.times : NULL;
pf->cur[BPHYS_DATA_BOIDS] = (data_types & (1<<BPHYS_DATA_BOIDS)) ? &pf->data.boids : NULL;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
/* Check to see if point number "index" is in pm, uses binary search for index data. */
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
int BKE_ptcache_mem_index_find(PTCacheMem *pm, unsigned int index)
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
{
Fix T37712: Point cache index lookup crashed with 0 points stored. The BKE_ptcache_mem_index_find is using unsigned ints for binary search "high" values - but this leads to integer overflow if the totpoint number is 0 and causes invalid array access.
2013-12-09 11:02:41 +01:00
if (pm->totpoint > 0 && pm->data[BPHYS_DATA_INDEX]) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int *data = pm->data[BPHYS_DATA_INDEX];
unsigned int mid, low = 0, high = pm->totpoint - 1;
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (index < *data || index > *(data+high))
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
return -1;
/* check simple case for continuous indexes first */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (index-*data < high && data[index-*data] == index)
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
return index-*data;
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (low <= high) {
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
mid= (low + high)/2;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (data[mid] > index)
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
high = mid - 1;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (data[mid] < index)
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
low = mid + 1;
else
return mid;
}
return -1;
}
else {
return (index < pm->totpoint ? index : -1);
}
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
void BKE_ptcache_mem_pointers_init(PTCacheMem *pm)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int data_types = pm->data_types;
int i;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++)
Possible fix for the issue that came up in [#24890] Vector Blur node is Buggy * Apparently some compilers don't respect proper operator precedence, so added some parentheses around to make inline conditionals unambiguous.
2010-11-25 11:38:55 +00:00
pm->cur[i] = ((data_types & (1<<i)) ? pm->data[i] : NULL);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
void BKE_ptcache_mem_pointers_incr(PTCacheMem *pm)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int i;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
if (pm->cur[i])
style cleanup
2012-12-28 14:19:05 +00:00
pm->cur[i] = (char *)pm->cur[i] + ptcache_data_size[i];
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int BKE_ptcache_mem_pointers_seek(int point_index, PTCacheMem *pm)
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
{
int data_types = pm->data_types;
Fix for [#25544] Blender crashes when changing the particles emission amount * I've getting bad feelings about the point cache index_array for a while (cause for this bug too), so from now on memory cache uses a simple binary search directly on the index data to handle queries to specific data points. * This is a bit slower than just checking from a dedicated array, but it's much less error prone, uses less memory and makes the code more readable too, so it's not a tough choice.
2011-01-09 07:41:51 +00:00
int i, index = BKE_ptcache_mem_index_find(pm, point_index);
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (index < 0) {
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
/* Can't give proper location without reallocation, so don't give any location.
* Some points will be cached improperly, but this only happens with simulation
* steps bigger than cache->step, so the cache has to be recalculated anyways
* at some point.
*/
return 0;
}
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++)
style cleanup
2012-12-28 14:19:05 +00:00
pm->cur[i] = data_types & (1<<i) ? (char *)pm->data[i] + index * ptcache_data_size[i] : NULL;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
return 1;
Fix for [#19955] 2.5 Particles are calculated wrong when object has any transform animation * The exact location of particle birth wasn't cached if cache step was greater than 1, so the interpolation from an unborn particle to a living one wasn't accurate at all. * Although not ideal, the current solution is to copy the birth location to the previous cached frame so it's always exact. This should work properly for memory & disk cache. * This fix shouldn't cause any problems, but exhaustive tests of caching are difficult so be sure to let me know if there's strange behavior.
2010-02-25 00:03:16 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static void ptcache_data_alloc(PTCacheMem *pm)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int data_types = pm->data_types;
int totpoint = pm->totpoint;
int i;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
if (data_types & (1<<i))
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
pm->data[i] = MEM_callocN(totpoint * ptcache_data_size[i], "PTCache Data");
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static void ptcache_data_free(PTCacheMem *pm)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
void **data = pm->data;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int i;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
if (data[i])
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
MEM_freeN(data[i]);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static void ptcache_data_copy(void *from[], void *to[])
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
int i;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<BPHYS_TOT_DATA; i++) {
Cleanup: indentation
2018-04-16 17:08:27 +02:00
/* note, durian file 03.4b_comp crashes if to[i] is not tested
* its NULL, not sure if this should be fixed elsewhere but for now its needed */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (from[i] && to[i])
Particles cleanup, optimizations and some small new stuff. New stuff - Bending springs for hair dynamics. Code cleanup & optimization - Disabled reactor particles temporarily for cleanup, it's a clumsy system that will be replaced with something better. - Removed child seams, something better will come here too :) - Normal particle drawing data is now saved between redraws if the particles don't move between redraws. * For example rotating the 3d view is now realtime even with 1M particles. - Many random values for particles now come from a lookup table making things much faster. - Most accessed small point cache functions are now much faster as macros. - Lot's of general code cleanup. - Nothing big should have changed so if something doesn't work like it used to it's probably just a typo somewhere :)
2009-09-17 22:00:49 +00:00
memcpy(to[i], from[i], ptcache_data_size[i]);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static void ptcache_extra_free(PTCacheMem *pm)
{
PTCacheExtra *extra = pm->extradata.first;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (extra) {
for (; extra; extra=extra->next) {
if (extra->data)
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
MEM_freeN(extra->data);
}
BLI_freelistN(&pm->extradata);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_old_elemsize(PTCacheID *pid)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->type==PTCACHE_TYPE_SOFTBODY)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 6 * sizeof(float);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type==PTCACHE_TYPE_PARTICLES)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return sizeof(ParticleKey);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type==PTCACHE_TYPE_CLOTH)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return 9 * sizeof(float);
return 0;
}
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
static void ptcache_find_frames_around(PTCacheID *pid, unsigned int frame, int *fra1, int *fra2)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Fixed point cache subframe interpolation issue. Point cache interpolation was using too early start frame in cases when target frame was a subframe. This appeared as random ghost particles when rendering particle animation with full sample motion blur enabled.
2011-12-16 09:22:28 +00:00
int cfra1=frame, cfra2=frame+1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (cfra1 >= pid->cache->startframe && !BKE_ptcache_id_exist(pid, cfra1))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cfra1--;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 < pid->cache->startframe)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cfra1 = 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (cfra2 <= pid->cache->endframe && !BKE_ptcache_id_exist(pid, cfra2))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cfra2++;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra2 > pid->cache->endframe)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cfra2 = 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 && !cfra2) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
*fra1 = 0;
*fra2 = cfra1;
}
else {
*fra1 = cfra1;
*fra2 = cfra2;
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->cache->mem_cache.first) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
PTCacheMem *pm = pid->cache->mem_cache.first;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheMem *pm2 = pid->cache->mem_cache.last;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (pm->next && pm->next->frame <= frame)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm= pm->next;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm2->frame < frame) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm2 = NULL;
quiet some clang warnings & fix for bugs in exceptional cases. - ghost C api, BLI_get_folder_version() could assign garbage values. - pointcache ptcache_find_frames_around() had a superfluous NULL check which would have crashed anyway if actually NULL.
2011-04-23 12:57:03 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
else {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (pm2->prev && pm2->prev->frame > frame) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm2= pm2->prev;
quiet some clang warnings & fix for bugs in exceptional cases. - ghost C api, BLI_get_folder_version() could assign garbage values. - pointcache ptcache_find_frames_around() had a superfluous NULL check which would have crashed anyway if actually NULL.
2011-04-23 12:57:03 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!pm2) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
*fra1 = 0;
*fra2 = pm->frame;
}
else {
*fra1 = pm->frame;
*fra2 = pm2->frame;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static PTCacheMem *ptcache_disk_frame_to_mem(PTCacheID *pid, int cfra)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
{
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheFile *pf = ptcache_file_open(pid, PTCACHE_FILE_READ, cfra);
PTCacheMem *pm = NULL;
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int i, error = 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf == NULL)
replace 0 with NULL when used as a pointer
2011-03-03 17:58:06 +00:00
return NULL;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!ptcache_file_header_begin_read(pf))
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && (pf->type != pid->type || !pid->read_header(pf)))
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
pm = MEM_callocN(sizeof(PTCacheMem), "Pointcache mem");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
pm->totpoint = pf->totpoint;
pm->data_types = pf->data_types;
pm->frame = pf->frame;
ptcache_data_alloc(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf->flag & PTCACHE_TYPEFLAG_COMPRESS) {
for (i=0; i<BPHYS_TOT_DATA; i++) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
unsigned int out_len = pm->totpoint*ptcache_data_size[i];
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf->data_types & (1<<i))
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)(pm->data[i]), out_len);
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
else {
BKE_ptcache_mem_pointers_init(pm);
ptcache_file_pointers_init(pf);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<pm->totpoint; i++) {
if (!ptcache_file_data_read(pf)) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
break;
}
ptcache_data_copy(pf->cur, pm->cur);
BKE_ptcache_mem_pointers_incr(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && pf->flag & PTCACHE_TYPEFLAG_EXTRADATA) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int extratype = 0;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (ptcache_file_read(pf, &extratype, 1, sizeof(unsigned int))) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheExtra *extra = MEM_callocN(sizeof(PTCacheExtra), "Pointcache extradata");
extra->type = extratype;
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
ptcache_file_read(pf, &extra->totdata, 1, sizeof(unsigned int));
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
extra->data = MEM_callocN(extra->totdata * ptcache_extra_datasize[extra->type], "Pointcache extradata->data");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf->flag & PTCACHE_TYPEFLAG_COMPRESS)
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_read(pf, (unsigned char *)(extra->data), extra->totdata*ptcache_extra_datasize[extra->type]);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
else
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
ptcache_file_read(pf, extra->data, extra->totdata, ptcache_extra_datasize[extra->type]);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BLI_addtail(&pm->extradata, extra);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (error && pm) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_data_free(pm);
ptcache_extra_free(pm);
MEM_freeN(pm);
pm = NULL;
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
ptcache_file_close(pf);
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (error && G.debug & G_DEBUG)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
printf("Error reading from disk cache\n");
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
return pm;
}
static int ptcache_mem_frame_to_disk(PTCacheID *pid, PTCacheMem *pm)
{
PTCacheFile *pf = NULL;
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int i, error = 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_FRAME, pm->frame);
pf = ptcache_file_open(pid, PTCACHE_FILE_WRITE, pm->frame);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf==NULL) {
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (G.debug & G_DEBUG)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
printf("Error opening disk cache file for writing\n");
return 0;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->data_types = pm->data_types;
pf->totpoint = pm->totpoint;
pf->type = pid->type;
pf->flag = 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm->extradata.first)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->flag |= PTCACHE_TYPEFLAG_EXTRADATA;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->compression)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->flag |= PTCACHE_TYPEFLAG_COMPRESS;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!ptcache_file_header_begin_write(pf) || !pid->write_header(pf))
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error) {
if (pid->cache->compression) {
for (i=0; i<BPHYS_TOT_DATA; i++) {
if (pm->data[i]) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
unsigned int in_len = pm->totpoint*ptcache_data_size[i];
style cleanup
2013-01-08 02:06:16 +00:00
unsigned char *out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len) * 4, "pointcache_lzo_buffer");
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)(pm->data[i]), in_len, out, pid->cache->compression);
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
MEM_freeN(out);
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
}
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
else {
BKE_ptcache_mem_pointers_init(pm);
ptcache_file_pointers_init(pf);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<pm->totpoint; i++) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
ptcache_data_copy(pm->cur, pf->cur);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!ptcache_file_data_write(pf)) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
break;
}
BKE_ptcache_mem_pointers_incr(pm);
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && pm->extradata.first) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheExtra *extra = pm->extradata.first;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; extra; extra=extra->next) {
if (extra->data == NULL || extra->totdata == 0)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
continue;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
ptcache_file_write(pf, &extra->type, 1, sizeof(unsigned int));
ptcache_file_write(pf, &extra->totdata, 1, sizeof(unsigned int));
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->compression) {
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
unsigned int in_len = extra->totdata * ptcache_extra_datasize[extra->type];
style cleanup
2013-01-08 02:06:16 +00:00
unsigned char *out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len) * 4, "pointcache_lzo_buffer");
style cleanup
2012-10-20 18:46:57 +00:00
ptcache_file_compressed_write(pf, (unsigned char *)(extra->data), in_len, out, pid->cache->compression);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
MEM_freeN(out);
}
else {
Viscoelastic springs for sph particle fluids, original patch by Stephen Whitehorn (chickencoop) * Viscoelastic springs between the fluid particles can simulate all kinds of viscous and elastic substances, such as jelly and honey. This is achieved by creating springs dynamically between neighboring particles and adjusting their rest length based on stretching/compression. * This nearly completes the currently intended functionality for particle fluids. The last missing thing is a surfacing extraction algorithm, which is needed for a proper representation of a sph fluid. * I also cleaned up and renamed some of the fluid parameters to make the ui a bit easier to understand. * One addition to the patch is an option to use "initial rest length" for the springs, which uses the lengths between the particles at the time of spring creation as the spring rest lengths instead of interaction radius/2. This makes the fluid keep it's original shape better (good for very viscoelastic materials), but can create large density differences inside the fluid (not really physically correct for a fluid). * Viscoelastic springs are stored in point cache as extra data.
2011-01-09 19:09:41 +00:00
ptcache_file_write(pf, extra->data, extra->totdata, ptcache_extra_datasize[extra->type]);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_file_close(pf);
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (error && G.debug & G_DEBUG)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
printf("Error writing to disk cache\n");
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
return error==0;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
Quiet warnings from last commit. Janne: could you check if ptcache_file_pointers_seek() & ptcache_read_init() can be removed?
2011-01-02 10:13:17 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static int ptcache_read_stream(PTCacheID *pid, int cfra)
{
PTCacheFile *pf = ptcache_file_open(pid, PTCACHE_FILE_READ, cfra);
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
int error = 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->read_stream == NULL)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
return 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf == NULL) {
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (G.debug & G_DEBUG)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
printf("Error opening disk cache file for reading\n");
return 0;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
if (!ptcache_file_header_begin_read(pf)) {
pid->error(pid->calldata, "Failed to read point cache file");
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
}
else if (pf->type != pid->type) {
pid->error(pid->calldata, "Point cache file has wrong type");
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
}
else if (!pid->read_header(pf)) {
pid->error(pid->calldata, "Failed to read point cache file header");
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
}
else if (pf->totpoint != pid->totpoint(pid->calldata, cfra)) {
pid->error(pid->calldata, "Number of points in cache does not match mesh");
error = 1;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
ptcache_file_pointers_init(pf);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
// we have stream reading here
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
if (!pid->read_stream(pf, pid->calldata)) {
pid->error(pid->calldata, "Failed to read point cache file data");
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
error = 1;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
}
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
ptcache_file_close(pf);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
return error == 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
}
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
static int ptcache_read_openvdb_stream(PTCacheID *pid, int cfra)
{
#ifdef WITH_OPENVDB
char filename[FILE_MAX * 2];
Cleanup: style, whitespace, doxy filepaths
2016-06-19 06:25:54 +10:00
/* save blend file before using disk pointcache */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (!G.relbase_valid && (pid->cache->flag & PTCACHE_EXTERNAL) == 0)
return 0;
ptcache_filename(pid, filename, cfra, 1, 1);
if (!BLI_exists(filename)) {
return 0;
}
struct OpenVDBReader *reader = OpenVDBReader_create();
OpenVDBReader_open(reader, filename);
if (!pid->read_openvdb_stream(reader, pid->calldata)) {
return 0;
}
return 1;
#else
UNUSED_VARS(pid, cfra);
return 0;
#endif
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static int ptcache_read(PTCacheID *pid, int cfra)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
{
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheMem *pm = NULL;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int i;
int *index = &i;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* get a memory cache to read from */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = ptcache_disk_frame_to_mem(pid, cfra);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
else {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = pid->cache->mem_cache.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (pm && pm->frame != cfra)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = pm->next;
}
/* read the cache */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
int totpoint = pm->totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
if ((pid->data_types & (1<<BPHYS_DATA_INDEX)) == 0) {
int pid_totpoint = pid->totpoint(pid->calldata, cfra);
if (totpoint != pid_totpoint) {
pid->error(pid->calldata, "Number of points in cache does not match mesh");
totpoint = MIN2(totpoint, pid_totpoint);
}
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_mem_pointers_init(pm);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<totpoint; i++) {
if (pm->data_types & (1<<BPHYS_DATA_INDEX))
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
index = pm->cur[BPHYS_DATA_INDEX];
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->read_point(*index, pid->calldata, pm->cur, (float)pm->frame, NULL);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
BKE_ptcache_mem_pointers_incr(pm);
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->read_extra_data && pm->extradata.first)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->read_extra_data(pid->calldata, pm, (float)pm->frame);
/* clean up temporary memory cache */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_data_free(pm);
ptcache_extra_free(pm);
MEM_freeN(pm);
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
return 1;
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static int ptcache_interpolate(PTCacheID *pid, float cfra, int cfra1, int cfra2)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
{
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheMem *pm = NULL;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int i;
int *index = &i;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* get a memory cache to read from */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = ptcache_disk_frame_to_mem(pid, cfra2);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
else {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = pid->cache->mem_cache.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (pm && pm->frame != cfra2)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = pm->next;
}
/* read the cache */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
int totpoint = pm->totpoint;
Related to #36532: show message in modifier when cloth point cache read fails, for example due to a subsurf modifier preceding a cloth modifier.
2013-09-03 19:58:09 +00:00
if ((pid->data_types & (1<<BPHYS_DATA_INDEX)) == 0) {
int pid_totpoint = pid->totpoint(pid->calldata, (int)cfra);
if (totpoint != pid_totpoint) {
pid->error(pid->calldata, "Number of points in cache does not match mesh");
totpoint = MIN2(totpoint, pid_totpoint);
}
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_mem_pointers_init(pm);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (i=0; i<totpoint; i++) {
if (pm->data_types & (1<<BPHYS_DATA_INDEX))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
index = pm->cur[BPHYS_DATA_INDEX];
pid->interpolate_point(*index, pid->calldata, pm->cur, cfra, (float)cfra1, (float)cfra2, NULL);
BKE_ptcache_mem_pointers_incr(pm);
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->interpolate_extra_data && pm->extradata.first)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->interpolate_extra_data(pid->calldata, pm, cfra, (float)cfra1, (float)cfra2);
/* clean up temporary memory cache */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_data_free(pm);
ptcache_extra_free(pm);
MEM_freeN(pm);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 1;
}
/* reads cache from disk or memory */
/* possible to get old or interpolated result */
Fix T49004 and motion blur of cloth, softbody, etc past the end frame. Point cache read code contains checks designed to prevent it reading stale data when the relevant simulation code should instead compute the next frame from the previous one. However in some situations like motion blur subframes the simulation can't possibly do it and just exits. This causes completely incorrect motion blur at or after the last cached frame. To fix, add a parameter that tells the cache code whether it should apply the checks and exit, or read what it can even if stale (true means exactly same as old behavior). Doing this in cache rather than clamping the frame number better in the caller lets it handle the case of incomplete cache that stops before the official last frame. Reviewed By: mont29, lukastoenne Maniphest Tasks: T49004 Differential Revision: https://developer.blender.org/D2144
2016-08-11 13:36:29 +03:00
int BKE_ptcache_read(PTCacheID *pid, float cfra, bool no_extrapolate_old)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
{
Fixed point cache subframe interpolation issue. Point cache interpolation was using too early start frame in cases when target frame was a subframe. This appeared as random ghost particles when rendering particle animation with full sample motion blur enabled.
2011-12-16 09:22:28 +00:00
int cfrai = (int)floor(cfra), cfra1=0, cfra2=0;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int ret = 0;
/* nothing to read to */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->totpoint(pid->calldata, cfrai) == 0)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_READ_INFO) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
pid->cache->flag &= ~PTCACHE_READ_INFO;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_read(pid, 0);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
}
/* first check if we have the actual frame cached */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra == (float)cfrai && BKE_ptcache_id_exist(pid, cfrai))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cfra1 = cfrai;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
/* no exact cache frame found so try to find cached frames around cfra */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 == 0)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_find_frames_around(pid, cfrai, &cfra1, &cfra2);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1 == 0 && cfra2 == 0)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 0;
spelling errors, no real changes to code.
2010-01-14 10:59:42 +00:00
/* don't read old cache if already simulated past cached frame */
Fix T49004 and motion blur of cloth, softbody, etc past the end frame. Point cache read code contains checks designed to prevent it reading stale data when the relevant simulation code should instead compute the next frame from the previous one. However in some situations like motion blur subframes the simulation can't possibly do it and just exits. This causes completely incorrect motion blur at or after the last cached frame. To fix, add a parameter that tells the cache code whether it should apply the checks and exit, or read what it can even if stale (true means exactly same as old behavior). Doing this in cache rather than clamping the frame number better in the caller lets it handle the case of incomplete cache that stops before the official last frame. Reviewed By: mont29, lukastoenne Maniphest Tasks: T49004 Differential Revision: https://developer.blender.org/D2144
2016-08-11 13:36:29 +03:00
if (no_extrapolate_old) {
if (cfra1 == 0 && cfra2 && cfra2 <= pid->cache->simframe)
return 0;
if (cfra1 && cfra1 == cfra2)
return 0;
}
else {
/* avoid calling interpolate between the same frame values */
if (cfra1 && cfra1 == cfra2)
cfra1 = 0;
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1) {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (pid->file_type == PTCACHE_FILE_OPENVDB && pid->read_openvdb_stream) {
if (!ptcache_read_openvdb_stream(pid, cfra1)) {
return 0;
}
}
else if (pid->read_stream) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
if (!ptcache_read_stream(pid, cfra1))
return 0;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->read_point)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_read(pid, cfra1);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra2) {
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (pid->file_type == PTCACHE_FILE_OPENVDB && pid->read_openvdb_stream) {
if (!ptcache_read_openvdb_stream(pid, cfra2)) {
return 0;
}
}
else if (pid->read_stream) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
if (!ptcache_read_stream(pid, cfra2))
return 0;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->read_point) {
if (cfra1 && cfra2 && pid->interpolate_point)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_interpolate(pid, cfra, cfra1, cfra2);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
else
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_read(pid, cfra2);
Keyed physics refresh: - Keyed targets in one list instead of "chaining", this opens up many more possibilities than before and is much less obscure. - Better keyed timing possibilities (time & duration in frames). - Looping over keyed targets list. Other changes: - New child setting "length" with threshold (great for guard & underfur with a single particle system) - Modularization of path interpolation code. - Cleared "animateable" flags from many particle settings that shouldn't be animateable. Fixes: - Keyed particles weren't copied properly (ancient bug). - Hair rotations depended on global z-axis for root rotation so downward facing strands could flip rotation randomly. Now initial hair rotation is derived from face dependent hair matrix. (This caused for example ugly flipping of child strands on some cases). - Children from faces weren't calculated straight after activating them. - Multiple disk cache fixes: * Disk cache didn't work correctly with frame steps. * Conversion from memory cache to disk cache didn't work with cloth. * Disk cache crashed on some frames trying to close an already closed cache file. * Trails didn't work with disk cached particles. - Child rough effects were effected by emitter object loc/rot making them next to useless with animation, why didn't anybody tell me this!! - Lots of random code cleanup.
2009-07-12 23:38:47 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra1)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ret = (cfra2 ? PTCACHE_READ_INTERPOLATED : PTCACHE_READ_EXACT);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (cfra2) {
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
ret = PTCACHE_READ_OLD;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->cache->simframe = cfra2;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Related to previous commit I did: Removed checks for PTCACHE_QUICK_CACHE for exception handling. This to ensure normal cache works as if old PTCACHE_QUICK_CACHE wasn't set. Thanks Campbell for pointing at this!
2012-10-23 15:01:41 +00:00
cfrai = (int)cfra;
/* clear invalid cache frames so that better stuff can be simulated */
if (pid->cache->flag & PTCACHE_OUTDATED) {
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_AFTER, cfrai);
}
else if (pid->cache->flag & PTCACHE_FRAMES_SKIPPED) {
if (cfra <= pid->cache->last_exact)
pid->cache->flag &= ~PTCACHE_FRAMES_SKIPPED;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
Related to previous commit I did: Removed checks for PTCACHE_QUICK_CACHE for exception handling. This to ensure normal cache works as if old PTCACHE_QUICK_CACHE wasn't set. Thanks Campbell for pointing at this!
2012-10-23 15:01:41 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_AFTER, MAX2(cfrai, pid->cache->last_exact));
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return ret;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static int ptcache_write_stream(PTCacheID *pid, int cfra, int totpoint)
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
{
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheFile *pf = NULL;
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
int error = 0;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_FRAME, cfra);
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf = ptcache_file_open(pid, PTCACHE_FILE_WRITE, cfra);
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf==NULL) {
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (G.debug & G_DEBUG)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
printf("Error opening disk cache file for writing\n");
return 0;
}
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pf->data_types = pid->data_types;
pf->totpoint = totpoint;
pf->type = pid->type;
pf->flag = 0;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && (!ptcache_file_header_begin_write(pf) || !pid->write_header(pf)))
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
error = 1;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!error && pid->write_stream)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_stream(pf, pid->calldata);
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
ptcache_file_close(pf);
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (error && G.debug & G_DEBUG)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
printf("Error writing to disk cache\n");
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
return error == 0;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
}
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
static int ptcache_write_openvdb_stream(PTCacheID *pid, int cfra)
{
#ifdef WITH_OPENVDB
struct OpenVDBWriter *writer = OpenVDBWriter_create();
char filename[FILE_MAX * 2];
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_FRAME, cfra);
ptcache_filename(pid, filename, cfra, 1, 1);
BLI_make_existing_file(filename);
int error = pid->write_openvdb_stream(writer, pid->calldata);
OpenVDBWriter_write(writer, filename);
OpenVDBWriter_free(writer);
return error == 0;
#else
UNUSED_VARS(pid, cfra);
return 0;
#endif
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
static int ptcache_write(PTCacheID *pid, int cfra, int overwrite)
{
PointCache *cache = pid->cache;
PTCacheMem *pm=NULL, *pm2=NULL;
int totpoint = pid->totpoint(pid->calldata, cfra);
int i, error = 0;
pm = MEM_callocN(sizeof(PTCacheMem), "Pointcache mem");
pm->totpoint = pid->totwrite(pid->calldata, cfra);
pm->data_types = cfra ? pid->data_types : pid->info_types;
ptcache_data_alloc(pm);
BKE_ptcache_mem_pointers_init(pm);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (overwrite) {
if (cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
int fra = cfra-1;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (fra >= cache->startframe && !BKE_ptcache_id_exist(pid, fra))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
fra--;
pm2 = ptcache_disk_frame_to_mem(pid, fra);
}
else
pm2 = cache->mem_cache.last;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->write_point) {
for (i=0; i<totpoint; i++) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
int write = pid->write_point(i, pid->calldata, pm->cur, cfra);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (write) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_mem_pointers_incr(pm);
/* newly born particles have to be copied to previous cached frame */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (overwrite && write == 2 && pm2 && BKE_ptcache_mem_pointers_seek(i, pm2))
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_point(i, pid->calldata, pm2->cur, cfra);
}
}
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->write_extra_data)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pid->write_extra_data(pid->calldata, pm, cfra);
pm->frame = cfra;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_DISK_CACHE) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
error += !ptcache_mem_frame_to_disk(pid, pm);
style cleanup: follow style guide for/with/if spacing
2012-03-24 07:52:14 +00:00
// if (pm) /* pm is always set */
corrections for redundant null checks & transform printing a string into its self.
2011-04-29 06:59:18 +00:00
{
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_data_free(pm);
ptcache_extra_free(pm);
MEM_freeN(pm);
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm2) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
error += !ptcache_mem_frame_to_disk(pid, pm2);
ptcache_data_free(pm2);
ptcache_extra_free(pm2);
MEM_freeN(pm2);
}
}
else {
BLI_addtail(&cache->mem_cache, pm);
}
return error;
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
static int ptcache_write_needed(PTCacheID *pid, int cfra, int *overwrite)
{
PointCache *cache = pid->cache;
int ofra = 0, efra = cache->endframe;
cleanup: style/spelling
2015-02-02 23:29:53 +11:00
/* always start from scratch on the first frame */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra && cfra == cache->startframe) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, cfra);
cache->flag &= ~PTCACHE_REDO_NEEDED;
return 1;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
if (cfra==0 && cache->startframe > 0)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 1;
/* find last cached frame */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (efra > cache->startframe && !BKE_ptcache_id_exist(pid, efra))
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
efra--;
/* find second last cached frame */
ofra = efra-1;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (ofra > cache->startframe && !BKE_ptcache_id_exist(pid, ofra))
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ofra--;
}
else {
PTCacheMem *pm = cache->mem_cache.last;
/* don't write info file in memory */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra == 0)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm == NULL)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 1;
efra = pm->frame;
ofra = (pm->prev ? pm->prev->frame : efra - cache->step);
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (efra >= cache->startframe && cfra > efra) {
if (ofra >= cache->startframe && efra - ofra < cache->step) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
/* overwrite previous frame */
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_FRAME, efra);
*overwrite = 1;
}
return 1;
}
return 0;
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* writes cache to disk or memory */
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
int BKE_ptcache_write(PTCacheID *pid, unsigned int cfra)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
PointCache *cache = pid->cache;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
int totpoint = pid->totpoint(pid->calldata, cfra);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
int overwrite = 0, error = 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (totpoint == 0 || (cfra ? pid->data_types == 0 : pid->info_types == 0))
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
return 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (ptcache_write_needed(pid, cfra, &overwrite)==0)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
return 0;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (pid->file_type == PTCACHE_FILE_OPENVDB && pid->write_openvdb_stream) {
ptcache_write_openvdb_stream(pid, cfra);
}
else if (pid->write_stream) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_write_stream(pid, cfra, totpoint);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->write_point) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
error += ptcache_write(pid, cfra, overwrite);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* Mark frames skipped if more than 1 frame forwards since last non-skipped frame. */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra - cache->last_exact == 1 || cfra == cache->startframe) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
cache->last_exact = cfra;
cache->flag &= ~PTCACHE_FRAMES_SKIPPED;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* Don't mark skipped when writing info file (frame 0) */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (cfra)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
cache->flag |= PTCACHE_FRAMES_SKIPPED;
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* Update timeline cache display */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra && cache->cached_frames)
Fix for [#24134] pointcache memory error (crash)
2010-10-06 07:57:55 +00:00
cache->cached_frames[cfra-cache->startframe] = 1;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
BKE_ptcache_update_info(pid);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
return !error;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
/* youll need to close yourself after!
Cleanup: trailing whitespace (comment blocks) Strip unindented comment blocks - mainly headers to avoid conflicts.
2018-06-01 18:19:39 +02:00
* mode - PTCACHE_CLEAR_ALL,
style cleanup / comment formatting for bli/bke/bmesh
2012-03-03 20:19:11 +00:00
*/
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Clears & resets */
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
void BKE_ptcache_id_clear(PTCacheID *pid, int mode, unsigned int cfra)
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
{
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int len; /* store the length of the string */
unsigned int sta, end;
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
/* mode is same as fopen's modes */
DIR *dir;
struct dirent *de;
Fix for part of bug #17759: point cache clearing did too much, also clearing other physics systems on the same object.
2008-10-04 21:26:48 +00:00
char path[MAX_PTCACHE_PATH];
char filename[MAX_PTCACHE_FILE];
char path_full[MAX_PTCACHE_FILE];
char ext[MAX_PTCACHE_PATH];
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!pid || !pid->cache || pid->cache->flag & PTCACHE_BAKED)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return;
Dynamic Paint: * Added option to preview surface wetmap instead of paint. * Changed default paint color to blue for better visibility. * Fix: Random sized particles didn't work anymore. * Fix: Particle initial velocity issues when using canvas substeps and a particle brush.
2011-08-28 16:36:47 +00:00
if (pid->cache->flag & PTCACHE_IGNORE_CLEAR)
return;
Fix for [#25218] No smoke is emitted when particle system starts and ends on same frame * Depsgraph wasn't updated properly for smoke flow collision object dependencies. * Smoke also wasn't properly using the actual emission frame of the flow particles. * There was a lot of bloated logic in some parts of particle code so this fix turned into a small scale cleanup operation. ** As a result particle updating and cache usage should be a bit more stable too.
2010-12-15 17:05:34 +00:00
sta = pid->cache->startframe;
end = pid->cache->endframe;
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
#ifndef DURIAN_POINTCACHE_LIB_OK
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
/* don't allow clearing for linked objects */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->ob->id.lib)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return;
Baking for dupligroup & linked library objects - library data allows pointcache writing (hard to know how this should work long term so ifdef'd for now) - changing the frame now updates the dupligroup objects - BKE_ptcache_ids_from_object(), option to get the id's from duplis note! scene_update_tagged() is called from the main() loop, and runs BKE_ptcache_quick_cache_all(), this could become a performance issue, especially with duplis, should probably not call BKE_ptcache_quick_cache_all() all the time, even when not playing back animation. (commits 27856 by Campbell from render25 branch)
2010-03-30 12:23:13 +00:00
#endif
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
Pointcache for unsaved files is saved in /tmp/blendcache_untitled/... instead of not generating pointcache at all, be careful since the temp path can be NULL or missing. This is messy because the files can be left there when blender quits, however saveBuffers and sessions alredy do :/ Some cleanup function needs to Blender that deals with this!
2008-02-12 17:11:45 +00:00
/*if (!G.relbase_valid) return; *//* save blend file before using pointcache */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const char *fext = ptcache_file_extension(pid);
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
/* clear all files in the temp dir with the prefix of the ID and the ".bphys" suffix */
switch (mode) {
case PTCACHE_CLEAR_ALL:
style cleanup: trailing tabs & expand some non prefix tabs into spaces.
2012-10-21 05:46:41 +00:00
case PTCACHE_CLEAR_BEFORE:
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
case PTCACHE_CLEAR_AFTER:
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
ptcache_path(pid, path);
dir = opendir(path);
if (dir==NULL)
return;
Fix for bad file name string matching in point caches, leading to deletion of baked caches. This happens when objects use file names with matching prefixes: "CubeX" -> not baked "CubeXYZ" -> baked The first objects cache should be discarded up to the current frame on file load, but the second should be left intact. But because the cache file names for both use the same prefix as well (based on hex name representation) they both match the "CubeX" name and get discarded. Adding the underscore terminator solves this issue, because it is never part of the hex file name string. WARNING: this solution does not work with custom names for point caches. This feature is pretty much broken, users have to ensure their names are unique themselves. Due to the possibility of underscores in names and the ambiguity of point cache suffixes there is no reliable way to encode filenames in that case.
2014-12-19 12:49:27 +01:00
len = ptcache_filename(pid, filename, cfra, 0, 0); /* no path */
/* append underscore terminator to ensure we don't match similar names
* from objects whose names start with the same prefix
*/
if (len < sizeof(filename) - 2) {
BLI_strncpy(filename + len, "_", sizeof(filename) - 2 - len);
len += 1;
}
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_snprintf(ext, sizeof(ext), "_%02u%s", pid->stack_index, fext);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (STREQLEN(filename, de->d_name, len)) { /* do we have the right prefix */
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
if (mode == PTCACHE_CLEAR_ALL) {
merge of last commit to trunk
2010-05-11 20:06:20 +00:00
pid->cache->last_exact = MIN2(pid->cache->startframe, 0);
enforce string limits (reported by pedantic checking tools & some developers). mostly replace strcpy with BLI_strncpy and multiple strcat's with a BLI_snprintf(). also fix possible crash if CWD isnt available.
2011-02-13 03:21:27 +00:00
BLI_join_dirfile(path_full, sizeof(path_full), path, de->d_name);
patch [#34103] fileops_1.patch from Lawrence D'Oliveiro (ldo) Add comments and use of bool type in fileops.c
2013-03-05 03:53:22 +00:00
BLI_delete(path_full, false, false);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* read the number of the file */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const int frame = ptcache_frame_from_filename(de->d_name, ext);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (frame != -1) {
style cleanup: brace placement/newlines
2012-05-24 16:35:45 +00:00
if ((mode == PTCACHE_CLEAR_BEFORE && frame < cfra) ||
(mode == PTCACHE_CLEAR_AFTER && frame > cfra))
{
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
enforce string limits (reported by pedantic checking tools & some developers). mostly replace strcpy with BLI_strncpy and multiple strcat's with a BLI_snprintf(). also fix possible crash if CWD isnt available.
2011-02-13 03:21:27 +00:00
BLI_join_dirfile(path_full, sizeof(path_full), path, de->d_name);
patch [#34103] fileops_1.patch from Lawrence D'Oliveiro (ldo) Add comments and use of bool type in fileops.c
2013-03-05 03:53:22 +00:00
BLI_delete(path_full, false, false);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->cached_frames && frame >=sta && frame <= end)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
pid->cache->cached_frames[frame-sta] = 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
}
}
}
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
closedir(dir);
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == PTCACHE_CLEAR_ALL && pid->cache->cached_frames)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
memset(pid->cache->cached_frames, 0, MEM_allocN_len(pid->cache->cached_frames));
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
else {
PTCacheMem *pm= pid->cache->mem_cache.first;
PTCacheMem *link= NULL;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == PTCACHE_CLEAR_ALL) {
merge of last commit to trunk
2010-05-11 20:06:20 +00:00
/*we want startframe if the cache starts before zero*/
pid->cache->last_exact = MIN2(pid->cache->startframe, 0);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; pm; pm=pm->next) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_data_free(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_extra_free(pm);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
BLI_freelistN(&pid->cache->mem_cache);
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->cached_frames)
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
memset(pid->cache->cached_frames, 0, MEM_allocN_len(pid->cache->cached_frames));
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else {
while (pm) {
style cleanup: brace placement/newlines
2012-05-24 16:35:45 +00:00
if ((mode == PTCACHE_CLEAR_BEFORE && pm->frame < cfra) ||
(mode == PTCACHE_CLEAR_AFTER && pm->frame > cfra))
{
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
link = pm;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->cached_frames && pm->frame >=sta && pm->frame <= end)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
pid->cache->cached_frames[pm->frame-sta] = 0;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_data_free(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_extra_free(pm);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
pm = pm->next;
BLI_freelinkN(&pid->cache->mem_cache, link);
}
else
pm = pm->next;
}
}
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
break;
case PTCACHE_CLEAR_FRAME:
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
if (BKE_ptcache_id_exist(pid, cfra)) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_filename(pid, filename, cfra, 1, 1); /* no path */
patch [#34103] fileops_1.patch from Lawrence D'Oliveiro (ldo) Add comments and use of bool type in fileops.c
2013-03-05 03:53:22 +00:00
BLI_delete(filename, false, false);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
}
else {
PTCacheMem *pm = pid->cache->mem_cache.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; pm; pm=pm->next) {
if (pm->frame == cfra) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_data_free(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_extra_free(pm);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
BLI_freelinkN(&pid->cache->mem_cache, pm);
break;
}
}
}
style cleanup: also rename bmesh_decimate.c --> bmesh_decimate_collapse.c
2012-10-23 03:38:26 +00:00
if (pid->cache->cached_frames && cfra >= sta && cfra <= end)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
pid->cache->cached_frames[cfra-sta] = 0;
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
break;
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
BKE_ptcache_update_info(pid);
Particles ========= Merge of the famous particle patch by Janne Karhu, a full rewrite of the Blender particle system. This includes: - Emitter, Hair and Reactor particle types. - Newtonian, Keyed and Boids physics. - Various particle visualisation and rendering types. - Vertex group and texture control for various properties. - Interpolated child particles from parents. - Hair editing with combing, growing, cutting, .. . - Explode modifier. - Harmonic, Magnetic fields, and multiple falloff types. .. and lots of other things, some more info is here: http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite http://wiki.blender.org/index.php/BlenderDev/Particles_Rewrite_Doc The new particle system cannot be backwards compatible. Old particle systems are being converted to the new system, but will require tweaking to get them looking the same as before. Point Cache =========== The new system to replace manual baking, based on automatic caching on disk. This is currently used by softbodies and the particle system. See the Cache API section on: http://wiki.blender.org/index.php/BlenderDev/PhysicsSprint Documentation ============= These new features still need good docs for the release logs, help for this is appreciated.
2007-11-26 22:09:57 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int BKE_ptcache_id_exist(PTCacheID *pid, int cfra)
2nd try at committing to trunk (instead of particles branch) :) Fixes for bugs in #7833 - Hair softbody cache was reset on rendering. - When Alt-a playback is synced to audio frames are skipped. This caused dynamics caching to go crazy. Now the first loop of playback goes through all frames if there are non cached dynamic systems. Invalid point caches weren't cleared when new systems were created. This caused messed up shapes when softbody was added.
2007-11-29 21:09:16 +00:00
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!pid->cache)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return 0;
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cfra<pid->cache->startframe || cfra > pid->cache->endframe)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
return 0;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->cached_frames && pid->cache->cached_frames[cfra-pid->cache->startframe]==0)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
return 0;
2nd try at committing to trunk (instead of particles branch) :) Fixes for bugs in #7833 - Hair softbody cache was reset on rendering. - When Alt-a playback is synced to audio frames are skipped. This caused dynamics caching to go crazy. Now the first loop of playback goes through all frames if there are non cached dynamic systems. Invalid point caches weren't cleared when new systems were created. This caused messed up shapes when softbody was added.
2007-11-29 21:09:16 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
char filename[MAX_PTCACHE_FILE];
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_filename(pid, filename, cfra, 1, 1);
2nd try at committing to trunk (instead of particles branch) :) Fixes for bugs in #7833 - Hair softbody cache was reset on rendering. - When Alt-a playback is synced to audio frames are skipped. This caused dynamics caching to go crazy. Now the first loop of playback goes through all frames if there are non cached dynamic systems. Invalid point caches weren't cleared when new systems were created. This caused messed up shapes when softbody was added.
2007-11-29 21:09:16 +00:00
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
return BLI_exists(filename);
}
else {
PTCacheMem *pm = pid->cache->mem_cache.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; pm; pm=pm->next) {
if (pm->frame==cfra)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
return 1;
}
return 0;
}
2nd try at committing to trunk (instead of particles branch) :) Fixes for bugs in #7833 - Hair softbody cache was reset on rendering. - When Alt-a playback is synced to audio frames are skipped. This caused dynamics caching to go crazy. Now the first loop of playback goes through all frames if there are non cached dynamic systems. Invalid point caches weren't cleared when new systems were created. This caused messed up shapes when softbody was added.
2007-11-29 21:09:16 +00:00
}
2.5 Think global, act local! The old favorite G.scene gone! Man... that took almost 2 days. Also removed G.curscreen and G.edbo. Not everything could get solved; here's some notes. - modifiers now store current scene in ModifierData. This is not meant for permanent, but it can probably stick there until we cleaned the anim system and depsgraph to cope better with timing issues. - Game engine G.scene should become an argument for staring it. Didn't solve this yet. - Texture nodes should get scene cfra, but the current implementation is too tightly wrapped to do it easily.
2009-01-04 14:14:06 +00:00
void BKE_ptcache_id_time(PTCacheID *pid, Scene *scene, float cfra, int *startframe, int *endframe, float *timescale)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
/* Object *ob; */ /* UNUSED */
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
PointCache *cache;
Fix for [#27579] Particles Cache Problem * Horrors from the ancient world of deprecated code: object animation offset can't really work correctly with particles unless point cache takes full control of particle system timing. * Disabled the non-working offset control from effecting particles so that for now particles will work consistently and the offset is only applied to the object.
2011-07-05 02:18:55 +00:00
/* float offset; unused for now */
float time, nexttime;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
patch [#30227] Various MSVC (32-bit) Warning and Typo Fixes made some small edits - removed changes to AVI reading since the data types are apart of the format spec. - absf -> abs for a double value in render code.
2012-02-17 19:21:47 +00:00
/* TODO: this has to be sorted out once bsystem_time gets redone, */
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
/* now caches can handle interpolating etc. too - jahka */
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
/* time handling for point cache:
* - simulation time is scaled by result of bsystem_time
* - for offsetting time only time offset is taken into account, since
* that's always the same and can't be animated. a timeoffset which
code cleanup: spelling
2012-07-16 23:23:33 +00:00
* varies over time is not simple to support.
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
* - field and motion blur offsets are currently ignored, proper solution
* is probably to interpolate results from two frames for that ..
*/
replace VECCOPY with copy_v3_v3, same for 2d copy, also added vec copy functions for int & char.
2011-11-07 01:38:32 +00:00
/* ob= pid->ob; */ /* UNUSED */
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
cache= pid->cache;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (timescale) {
code cleanup: BKE_scene api naming. also stop numpy from being found in /usr/include with cmake.
2012-05-05 14:33:36 +00:00
time= BKE_scene_frame_get(scene);
nexttime = BKE_scene_frame_get_from_ctime(scene, CFRA + 1.0f);
Second attempt at getting rid of bsystem_time() Hopefully this fixes Collada. Can't really compile that to check here...
2011-11-06 12:12:14 +00:00
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
*timescale= MAX2(nexttime - time, 0.0f);
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (startframe && endframe) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
*startframe= cache->startframe;
*endframe= cache->endframe;
Fix for [#27579] Particles Cache Problem * Horrors from the ancient world of deprecated code: object animation offset can't really work correctly with particles unless point cache takes full control of particle system timing. * Disabled the non-working offset control from effecting particles so that for now particles will work consistently and the offset is only applied to the object.
2011-07-05 02:18:55 +00:00
/* TODO: time handling with object offsets and simulated vs. cached
* particles isn't particularly easy, so for now what you see is what
* you get. In the future point cache could handle the whole particle
* system timing. */
#if 0
if ((ob->partype & PARSLOW)==0) {
As discussed on the mailing list, removing the non-functional, incompatible, and unmaintainable Time Offset cruft. - Slow Parenting lives another day (just), although it now carries appropriate cautionary disclaimers. It's only really for the Game Engine nowadays, as that's the only place where it can possibly work with any reliability. - "Animation Hacks" panel is now "Relations Extras". I could've merged the two panels, though I figured these options weren't that frequently used to justify taking up screen-space by default along with the panel
2011-11-06 05:46:45 +00:00
offset= ob->sf;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
*startframe += (int)(offset+0.5f);
*endframe += (int)(offset+0.5f);
}
Fix for [#27579] Particles Cache Problem * Horrors from the ancient world of deprecated code: object animation offset can't really work correctly with particles unless point cache takes full control of particle system timing. * Disabled the non-working offset control from effecting particles so that for now particles will work consistently and the offset is only applied to the object.
2011-07-05 02:18:55 +00:00
#endif
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
/* verify cached_frames array is up to date */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->cached_frames) {
if (MEM_allocN_len(cache->cached_frames) != sizeof(char) * (cache->endframe-cache->startframe+1)) {
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
MEM_freeN(cache->cached_frames);
cache->cached_frames = NULL;
style cleanup: trailing tabs & expand some non prefix tabs into spaces.
2012-10-21 05:46:41 +00:00
}
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->cached_frames==NULL && cache->endframe > cache->startframe) {
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int sta=cache->startframe;
unsigned int end=cache->endframe;
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
cache->cached_frames = MEM_callocN(sizeof(char) * (cache->endframe-cache->startframe+1), "cached frames array");
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->cache->flag & PTCACHE_DISK_CACHE) {
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
/* mode is same as fopen's modes */
DIR *dir;
struct dirent *de;
char path[MAX_PTCACHE_PATH];
char filename[MAX_PTCACHE_FILE];
char ext[MAX_PTCACHE_PATH];
Replace uint32_t in pointcache code with unsigned int as it's supported in dna * Not strictly necessary right now, but better for future. * Struct data (only boids at the moment) is now written as structs (with dna) so they work between 64 and 32 bit machines too.
2011-01-09 18:23:41 +00:00
unsigned int len; /* store the length of the string */
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
ptcache_path(pid, path);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
len = ptcache_filename(pid, filename, (int)cfra, 0, 0); /* no path */
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
dir = opendir(path);
if (dir==NULL)
return;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const char *fext = ptcache_file_extension(pid);
BLI_snprintf(ext, sizeof(ext), "_%02u%s", pid->stack_index, fext);
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (STREQLEN(filename, de->d_name, len)) { /* do we have the right prefix */
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
/* read the number of the file */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const int frame = ptcache_frame_from_filename(de->d_name, ext);
if ((frame != -1) && (frame >= sta && frame <= end)) {
cache->cached_frames[frame-sta] = 1;
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
}
}
}
}
closedir(dir);
}
else {
PTCacheMem *pm= pid->cache->mem_cache.first;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
while (pm) {
if (pm->frame >= sta && pm->frame <= end)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
cache->cached_frames[pm->frame-sta] = 1;
pm = pm->next;
}
}
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int BKE_ptcache_id_reset(Scene *scene, PTCacheID *pid, int mode)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
PointCache *cache;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
int reset, clear, after;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!pid->cache)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return 0;
cache= pid->cache;
reset= 0;
clear= 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
after= 0;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == PTCACHE_RESET_DEPSGRAPH) {
sim: Remove "continue physics" code This was left over from 2.4x days and is not used anymore. Now simulations are always interactive.
2013-01-22 20:47:03 +00:00
if (!(cache->flag & PTCACHE_BAKED)) {
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
after= 1;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
cache->flag |= PTCACHE_OUTDATED;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (mode == PTCACHE_RESET_BAKED) {
sim: Remove "continue physics" code This was left over from 2.4x days and is not used anymore. Now simulations are always interactive.
2013-01-22 20:47:03 +00:00
cache->flag |= PTCACHE_OUTDATED;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (mode == PTCACHE_RESET_OUTDATED) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
reset = 1;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_OUTDATED && !(cache->flag & PTCACHE_BAKED)) {
Hair dynamics with cloth simulation - Hair dynamics have their own panel in particle settings with the settings from cloth panel that apply to hair. - Basic internal friction force to quickly emulate self collisions and volume preservation. (Still very early code, but gives some idea of what's possible). - Softbody simulation is no longer used for hair. * Old files with sb dynamics should just load the hair without dynamics so new dynamics can be applied. * Invasion of particles exceptions in sb code is finally over. - Collisions with other objects are disabled for now and will be worked out in the future. Other changes/fixes: - Particle mode editing flag wasn't saved properly. - Some old files with edited hair didn't load correctly. - Disabled delete & specials menu in particle mode for non-hair editing. - Fixed yet one more cloth & softbody pointcache update issue. - Disconnect/connect hair now uses only the deformed mesh so it works correctly also for subsurfed emitters. - Hair editing now updates correctly with a moving emitter.
2009-09-10 22:32:33 +00:00
clear= 1;
cache->flag &= ~PTCACHE_OUTDATED;
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (reset) {
Big cleanup of particle system core, also some minor pointcache cleanup. Shouldn't contain any functional changes.
2010-03-21 20:36:06 +00:00
BKE_ptcache_invalidate(cache);
cache->flag &= ~PTCACHE_REDO_NEEDED;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->type == PTCACHE_TYPE_CLOTH)
most unused arg warnings corrected. - removed deprecated bitmap arg from IMB_allocImBuf (plugins will need updating). - mostly tagged UNUSED() since some of these functions look like they may need to have the arguments used later.
2010-10-16 14:32:17 +00:00
cloth_free_modifier(pid->calldata);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type == PTCACHE_TYPE_SOFTBODY)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
sbFreeSimulation(pid->calldata);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type == PTCACHE_TYPE_PARTICLES)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
psys_reset(pid->calldata, PSYS_RESET_DEPSGRAPH);
style cleanup: comments & spelling
2012-11-18 01:22:31 +00:00
#if 0
else if (pid->type == PTCACHE_TYPE_SMOKE_DOMAIN)
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
smokeModifier_reset(pid->calldata);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type == PTCACHE_TYPE_SMOKE_HIGHRES)
style cleanup: comments & spelling
2012-11-18 01:22:31 +00:00
smokeModifier_reset_turbulence(pid->calldata);
#endif
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type == PTCACHE_TYPE_DYNAMICPAINT)
Fix #36058: Displace Modifier errors using a baked Image and displace baking inconsistency between 2.67/2.68RC and previous versions This was in fact really nasty bug, caused by multitex_nodes function using global variable R (which is a copy of current renderer). this variable is not initialized to anything meaningful for until first rendering (preview or final) happened. Since multitex_nodes might be used outside of render pipeline, made it so whether CM is on or off as an argument to functions multitex_ext_safe and multitex_ext. Now multitex_nodes() is only shall be used for stuff happening from render pipeline! Also needed to make some changes to other places, so all the usages of texture sampling knows for the fact whether CM is on or off. And one more change is related on behavior of dispalcement, wave, warp, weightvg modifiers and smoke. They'll be always using CM off since texture is used for influence, not for color. It's rather bigger patch, but it's mostly straightforward changes, which we really need to be done. Reviewed by Brecht, thanks!
2013-07-15 14:47:58 +00:00
dynamicPaint_clearSurface(scene, (DynamicPaintSurface*)pid->calldata);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (clear)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (after)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_AFTER, CFRA);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
return (reset || clear || after);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int BKE_ptcache_object_reset(Scene *scene, Object *ob, int mode)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
PTCacheID pid;
ParticleSystem *psys;
ModifierData *md;
Fix for bug [#13479] Particle system "corrupts" when changing material colour and used in conjunction with softbodies - changing particlesystem settings for hair particles cleared & recalculated everything even when hair softbody was baked
2008-07-11 23:47:44 +00:00
int reset, skip;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
reset= 0;
Fix for bug [#13479] Particle system "corrupts" when changing material colour and used in conjunction with softbodies - changing particlesystem settings for hair particles cleared & recalculated everything even when hair softbody was baked
2008-07-11 23:47:44 +00:00
skip= 0;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (ob->soft) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
BKE_ptcache_id_from_softbody(&pid, ob, ob->soft);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (psys=ob->particlesystem.first; psys; psys=psys->next) {
code cleanup: spelling
2012-10-20 20:36:51 +00:00
/* children or just redo can be calculated without resetting anything */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->recalc & PSYS_RECALC_REDO || psys->recalc & PSYS_RECALC_CHILD)
New hair child options: * Renamed children to "simple" and "interpolated" as this is easier to explain and more descriptive than "from particles" and "from faces". * Also shuffled the child ui around a bit to make it clearer. * Child seed parameter allows to change the seed for children independent of the main seed value. * Long hair mode for interpolated children: - Making even haircuts was impossible before as the child strand lengths were even, but their root coordinates were not similar in relation to the parent strands. - The "long hair" option uses the tips of the parent strands to calculate the child strand tips. * Hair parting options: - Hair parting can now be calculated dynamically on the fly when in 2.49 there was a cumbersome way of using emitter mesh seams to define parting lines. - For long hair parting can be created by a tip distance/root distance threshold. For example setting the minimum threshold to 2.0 creates partings between children belonging to parents with tip distance of three times the root distance ((1+2)*root distance). - For short hair the parting thresholds are used as angles between the root directions. * New kink parameters: - Kink flatness calculates kink into a shape that would have been achieved with an actual curling iron. - Kink amplitude clump determines how much the main clump value effects the kink amplitude. - The beginning of kink is now smoothed to make the hair look more natural close to the roots. * Some bugs fixed along the way too: - Child parent's were not determined correctly in some cases. - Children didn't always look correct in particle mode. - Changing child parameters caused actual particles to be recalculated. * Also cleaned up some deprecated code. All in all there should be no real changes to how old files look (except perhaps a bit better!), but the new options should make hair/fur creation a bit more enjoyable. I'll try to make a video demonstrating the new stuff shortly.
2011-01-07 11:24:34 +00:00
skip = 1;
/* Baked cloth hair has to be checked too, because we don't want to reset */
Hair dynamics with cloth simulation - Hair dynamics have their own panel in particle settings with the settings from cloth panel that apply to hair. - Basic internal friction force to quickly emulate self collisions and volume preservation. (Still very early code, but gives some idea of what's possible). - Softbody simulation is no longer used for hair. * Old files with sb dynamics should just load the hair without dynamics so new dynamics can be applied. * Invasion of particles exceptions in sb code is finally over. - Collisions with other objects are disabled for now and will be worked out in the future. Other changes/fixes: - Particle mode editing flag wasn't saved properly. - Some old files with edited hair didn't load correctly. - Disabled delete & specials menu in particle mode for non-hair editing. - Fixed yet one more cloth & softbody pointcache update issue. - Disconnect/connect hair now uses only the deformed mesh so it works correctly also for subsurfed emitters. - Hair editing now updates correctly with a moving emitter.
2009-09-10 22:32:33 +00:00
/* particles or cloth in that case -jahka */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (psys->clmd) {
Hair dynamics with cloth simulation - Hair dynamics have their own panel in particle settings with the settings from cloth panel that apply to hair. - Basic internal friction force to quickly emulate self collisions and volume preservation. (Still very early code, but gives some idea of what's possible). - Softbody simulation is no longer used for hair. * Old files with sb dynamics should just load the hair without dynamics so new dynamics can be applied. * Invasion of particles exceptions in sb code is finally over. - Collisions with other objects are disabled for now and will be worked out in the future. Other changes/fixes: - Particle mode editing flag wasn't saved properly. - Some old files with edited hair didn't load correctly. - Disabled delete & specials menu in particle mode for non-hair editing. - Fixed yet one more cloth & softbody pointcache update issue. - Disconnect/connect hair now uses only the deformed mesh so it works correctly also for subsurfed emitters. - Hair editing now updates correctly with a moving emitter.
2009-09-10 22:32:33 +00:00
BKE_ptcache_id_from_cloth(&pid, ob, psys->clmd);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (mode == PSYS_RESET_ALL || !(psys->part->type == PART_HAIR && (pid.cache->flag & PTCACHE_BAKED)))
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
Fix for bug [#13479] Particle system "corrupts" when changing material colour and used in conjunction with softbodies - changing particlesystem settings for hair particles cleared & recalculated everything even when hair softbody was baked
2008-07-11 23:47:44 +00:00
else
skip = 1;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (skip == 0 && psys->part) {
Fix for bug [#13479] Particle system "corrupts" when changing material colour and used in conjunction with softbodies - changing particlesystem settings for hair particles cleared & recalculated everything even when hair softbody was baked
2008-07-11 23:47:44 +00:00
BKE_ptcache_id_from_particles(&pid, ob, psys);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (md=ob->modifiers.first; md; md=md->next) {
if (md->type == eModifierType_Cloth) {
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
BKE_ptcache_id_from_cloth(&pid, ob, (ClothModifierData*)md);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (md->type == eModifierType_Smoke) {
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
SmokeModifierData *smd = (SmokeModifierData *)md;
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (smd->type & MOD_SMOKE_TYPE_DOMAIN) {
Smoke: * cache for low res (deactivating high res for now) * new way of view3d rendering of smoke (no longer 3 axes) -using 3dtexture now (introduced into gpu/intern) * introducing LZO and LZMA libs into extern (makefiles missing for now) * reducing memory usage after simulating for the frame ended (freeing temporary buffers) * splitting smoke into 2 modifier for the cache-sake (it cannot handle more than 1 cache on the same modifier-index) * no color on gui anymore * fixing non-power-of-2 resolutions (hopefully) * fixing select-deselect of domain drawing bug * fixing drawobject.c coding style (making Ton happy) ;-) HINT #1: If scons doesn't work -> cmakefiles are up-to-date, couldn't test scons (but i tried to mantain them, too) CODERS HINT #1: we really need a way to disable adding all modifiers through "Add Modifiers" dropdown! WARNING #1: before applying this commit, deactivate your SMOKE DOMAIN in your old files and save them then. You can open them then savely after that. WARNING #2: File and cache format of smoke can be changed, this is not final!
2009-08-20 00:33:59 +00:00
BKE_ptcache_id_from_smoke(&pid, ob, (SmokeModifierData*)md);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
}
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (md->type == eModifierType_DynamicPaint) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
DynamicPaintModifierData *pmd = (DynamicPaintModifierData *)md;
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (pmd->canvas) {
Dynamic Paint: Committed changes from previous weeks, biggest changes are: * Canvas can now have multiple "surfaces" that each can have specific format, type and settings. * Renewed UI to support this new system. * Aside from old "image sequence" output format, Dynamic Paint can now work on vertex level as well. Currently vertex paint and displace are supported. * Integrated vertex level painting with Point Cache. * Added viewport preview for Point Cache surfaces. Due to massive amount of changes, old Dynamic Paint saves are no longer supported. Also some features are temporarily missing or may not work properly.
2011-06-16 10:41:00 +00:00
DynamicPaintSurface *surface = pmd->canvas->surfaces.first;
for (; surface; surface=surface->next) {
BKE_ptcache_id_from_dynamicpaint(&pid, ob, surface);
reset |= BKE_ptcache_id_reset(scene, &pid, mode);
}
}
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
}
rigidbody: Add rigid body constraints Constraints connect two rigid bodies. Depending on which constraint is used different degrees of freedom are limited, e.g. a hinge constraint only allows the objects to rotate around a common axis. Constraints are implemented as individual objects and bahave similar to rigid bodies in terms of adding/removing/validating. The position and orientation of the constraint object is the pivot point of the constraint. Constraints have their own group in the rigid body world. To make connecting rigid bodies easier, there is a "Connect" operator that creates an empty objects with a rigid body constraint connecting the selected objects to active. Currently the following constraints are implemented: * Fixed * Point * Hinge * Slider * Piston * Generic Note: constraint limits aren't animatable yet).
2013-01-23 05:56:56 +00:00
if (scene->rigidbody_world && (ob->rigidbody_object || ob->rigidbody_constraint)) {
rigidbody: Add point cache support Add read/write/interpolate functions. In order to get rigid body point cache id from object it's now required to pass the scene to BKE_ptcache_ids_from_object(). Rigid body cache is drawn in the orange color of the bullet logo.
2013-01-23 05:56:34 +00:00
if (ob->rigidbody_object)
ob->rigidbody_object->flag |= RBO_FLAG_NEEDS_RESHAPE;
BKE_ptcache_id_from_rigidbody(&pid, ob, scene->rigidbody_world);
/* only flag as outdated, resetting should happen on start frame */
pid.cache->flag |= PTCACHE_OUTDATED;
}
Merge of itasc branch. Project files, scons and cmake should be working. Makefile updated but not tested. Comes with Eigen2 2.0.6 C++ matrix library.
2009-09-24 21:22:24 +00:00
if (ob->type == OB_ARMATURE)
BIK_clear_cache(ob->pose);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return reset;
}
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
/* Use this when quitting blender, with unsaved files */
void BKE_ptcache_remove(void)
{
Fix for part of bug #17759: point cache clearing did too much, also clearing other physics systems on the same object.
2008-10-04 21:26:48 +00:00
char path[MAX_PTCACHE_PATH];
char path_full[MAX_PTCACHE_PATH];
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
int rmdir = 1;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
ptcache_path(NULL, path);
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
Code cleanup: remove BLI_exist, now there is only BLI_exists. One function just called the other, they did the same thing.
2011-10-21 22:33:41 +00:00
if (BLI_exists(path)) {
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
/* The pointcache dir exists? - remove all pointcache */
DIR *dir;
struct dirent *de;
Fix for bug #8930: issue removing temporary pointcache directory on windows.
2008-04-12 12:56:49 +00:00
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
dir = opendir(path);
if (dir==NULL)
return;
while ((de = readdir(dir)) != NULL) {
Cleanup: current/parent paths: add helpers in BLI_path_utils. Also, avoid calling ugly strcmp with '.' or '..', making direct char checks is much cheaper here!
2015-01-26 16:58:02 +01:00
if (FILENAME_IS_CURRPAR(de->d_name)) {
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
/* do nothing */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else if (strstr(de->d_name, PTCACHE_EXT)) { /* do we have the right extension?*/
enforce string limits (reported by pedantic checking tools & some developers). mostly replace strcpy with BLI_strncpy and multiple strcat's with a BLI_snprintf(). also fix possible crash if CWD isnt available.
2011-02-13 03:21:27 +00:00
BLI_join_dirfile(path_full, sizeof(path_full), path, de->d_name);
patch [#34103] fileops_1.patch from Lawrence D'Oliveiro (ldo) Add comments and use of bool type in fileops.c
2013-03-05 03:53:22 +00:00
BLI_delete(path_full, false, false);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else {
spelling cleanup
2012-03-18 07:38:51 +00:00
rmdir = 0; /* unknown file, don't remove the dir */
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
}
}
Fix for bug #8930: issue removing temporary pointcache directory on windows.
2008-04-12 12:56:49 +00:00
closedir(dir);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
}
else {
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
rmdir = 0; /* path dosnt exist */
}
if (rmdir) {
patch [#34103] fileops_1.patch from Lawrence D'Oliveiro (ldo) Add comments and use of bool type in fileops.c
2013-03-05 03:53:22 +00:00
BLI_delete(path, true, false);
* Made BLI_join_dirfile() check before adding a slash between dir and file so as not to get /foo///bar.blend * Pointcache now uses the process id to construct the path for unsaved files. (so 2 or more blender's open wont try to read/write the same pointcache) * Temp pointcache is cleared when existing blender, added BIF_clear_tempfiles() for this. Should also be usedto clear EXR's in the temp dir (TODO), BIF_clear_tempfiles also needs to be added in more places. (On file load for instace)
2008-02-29 15:50:28 +00:00
}
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Point Cache handling */
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
PointCache *BKE_ptcache_add(ListBase *ptcaches)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
PointCache *cache;
cache= MEM_callocN(sizeof(PointCache), "PointCache");
cache->startframe= 1;
cache->endframe= 250;
Always init pointcache frame step to 1. Note it was already the case for many simulations (cloth, rigidbody, etc.), just doing this systematically in BKE_ptcache_add() now, will avoid us some recurring bug reports. Advanced users will know when they can/have to raise this parameter!
2014-04-25 17:55:37 +02:00
cache->step = 1;
Pointcache: *introducing unique ID's following brechts hint from ML Enhancements resulting from this: * multiple caches per modifier stack position
2009-08-25 18:41:36 +00:00
cache->index = -1;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
BLI_addtail(ptcaches, cache);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return cache;
}
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
void BKE_ptcache_free_mem(ListBase *mem_cache)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
PTCacheMem *pm = mem_cache->first;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm) {
for (; pm; pm=pm->next) {
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_data_free(pm);
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
ptcache_extra_free(pm);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
BLI_freelistN(mem_cache);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
void BKE_ptcache_free(PointCache *cache)
{
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
BKE_ptcache_free_mem(&cache->mem_cache);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->edit && cache->free_edit)
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
cache->free_edit(cache->edit);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->cached_frames)
Fixed: Showing pointcached frames in the timeline was terribly slow when using disk cache. * The existence of cached frames was checked each frame causing hundreds of disk operations per frame update. * Pointcache now keeps an updated array of the cached frames for fast "frame exists in cache" queries. * This fix also speeds up some other pointcache operations nicely.
2010-09-27 09:58:37 +00:00
MEM_freeN(cache->cached_frames);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
MEM_freeN(cache);
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
void BKE_ptcache_free_list(ListBase *ptcaches)
{
ghash/bli-listbase edits, rename BLI_ghash_pop -> BLI_ghash_popkey (since it takes a key as an arg and isnt popping any element from the hash as you might expect). add BLI_pophead/tail, since getting the first element from a list and removing it is a common task.
2013-08-26 23:37:08 +00:00
PointCache *cache;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
ghash/bli-listbase edits, rename BLI_ghash_pop -> BLI_ghash_popkey (since it takes a key as an arg and isnt popping any element from the hash as you might expect). add BLI_pophead/tail, since getting the first element from a list and removing it is a common task.
2013-08-26 23:37:08 +00:00
while ((cache = BLI_pophead(ptcaches))) {
Point cache editing: - Baked point caches for particles, cloth and softbody can now be edited in particle mode. * This overwrites the old cloth/sb cache editmode editing. * The type of editable system is chosen from a menu. * For particles the current particle system and it's current cache are used. - Currently this only works for caches that are in memory, but some automatic conversion from disk to memory and back can be implemented later. - All tools from hair editing can't be applied to point caches and are hidden in the tool panel and specials menu. Some functionality like subdividing paths can be later implemented in a slightly different way from how it works for hair. - Code is not yet optimized for speed, so editing might be slow sometimes. Known issues: - Cloth doesn't update properly while in particle mode, due to the way cloth modifier currently works. Daniel can you check on this? - As "particle mode" is not only for particles any more some other name would be in place? - Better icons are needed for the path, point, and tip-modes as the current icons from mesh edit mode are quite misleading. - Direct editing of point velocities is not yet implemented, but will be in the future. Other changes: - Hair editing doesn't require a "make editable" button press any more. - Multiple caches in single particle system disables changing emission properties. - Unified ui code for all point cache panels. * Defined in buttons_particle.py and imported for cloth, smoke & softbody. - Proper disabling of properties in ui after baking point caches. (Daniel could you please make needed disable code for smoke panels as their functionality is not familiar to me.) - Hair weight brush has been removed. Once hair dynamics is re-implemented I'll code a more useable alternative to the functionality. Bug fixes: - Unlinking particle settings crashed. - Deleting the active object with particles in the scene crashed. - Softbody didn't write point caches correctly on save.
2009-08-29 15:20:36 +00:00
BKE_ptcache_free(cache);
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
Refactor ID copying (and to some extent, ID freeing). This will allow much finer controll over how we copy data-blocks, from full copy in Main database, to "lighter" ones (out of Main, inside an already allocated datablock, etc.). This commit also transfers a llot of what was previously handled by per-ID-type custom code to generic ID handling code in BKE_library. Hopefully will avoid in future inconsistencies and missing bits we had all over the codebase in the past. It also adds missing copying handling for a few types, most notably Scene (which where using a fully customized handling previously). Note that the type of allocation used during copying (regular in Main, allocated but outside of Main, or not allocated by ID handling code at all) is stored in ID's, which allows to handle them correctly when freeing. This needs to be taken care of with caution when doing 'weird' unusual things with ID copying and/or allocation! As a final note, while rather noisy, this commit will hopefully not break too much existing branches, old 'API' has been kept for the main part, as a wrapper around new code. Cleaning it up will happen later. Design task : T51804 Phab Diff: D2714
2017-08-07 16:39:55 +02:00
static PointCache *ptcache_copy(PointCache *cache, const bool copy_data)
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
{
PointCache *ncache;
ncache= MEM_dupallocN(cache);
ListBase API: add utility api funcs for clearing and checking empty
2014-02-08 06:07:10 +11:00
BLI_listbase_clear(&ncache->mem_cache);
Keyed physics refresh: - Keyed targets in one list instead of "chaining", this opens up many more possibilities than before and is much less obscure. - Better keyed timing possibilities (time & duration in frames). - Looping over keyed targets list. Other changes: - New child setting "length" with threshold (great for guard & underfur with a single particle system) - Modularization of path interpolation code. - Cleared "animateable" flags from many particle settings that shouldn't be animateable. Fixes: - Keyed particles weren't copied properly (ancient bug). - Hair rotations depended on global z-axis for root rotation so downward facing strands could flip rotation randomly. Now initial hair rotation is derived from face dependent hair matrix. (This caused for example ugly flipping of child strands on some cases). - Children from faces weren't calculated straight after activating them. - Multiple disk cache fixes: * Disk cache didn't work correctly with frame steps. * Conversion from memory cache to disk cache didn't work with cloth. * Disk cache crashed on some frames trying to close an already closed cache file. * Trails didn't work with disk cached particles. - Child rough effects were effected by emitter object loc/rot making them next to useless with animation, why didn't anybody tell me this!! - Lots of random code cleanup.
2009-07-12 23:38:47 +00:00
Code cleanup: remove TRUE/FALSE & WITH_BOOL_COMPAT define
2014-04-01 11:34:00 +11:00
if (copy_data == false) {
Fix #32667: Curve softbodies doesn't render animation (cycles) Issue was caused by cycles being duplicated curve objects before converting them to mesh. This duplication will loose pointcache which resulted in object not being properly deformed.
2012-09-27 14:37:20 +00:00
ncache->cached_frames = NULL;
Fix #33600: copying objects with point caches did not copy the external, disk cache and lib path settings.
2012-12-18 14:05:54 +00:00
/* flag is a mix of user settings and simulator/baking state */
ncache->flag= ncache->flag & (PTCACHE_DISK_CACHE|PTCACHE_EXTERNAL|PTCACHE_IGNORE_LIBPATH);
Fix #32667: Curve softbodies doesn't render animation (cycles) Issue was caused by cycles being duplicated curve objects before converting them to mesh. This duplication will loose pointcache which resulted in object not being properly deformed.
2012-09-27 14:37:20 +00:00
ncache->simframe= 0;
}
else {
PTCacheMem *pm;
for (pm = cache->mem_cache.first; pm; pm = pm->next) {
PTCacheMem *pmn = MEM_dupallocN(pm);
int i;
for (i = 0; i < BPHYS_TOT_DATA; i++) {
if (pmn->data[i])
pmn->data[i] = MEM_dupallocN(pm->data[i]);
}
BKE_ptcache_mem_pointers_init(pm);
BLI_addtail(&ncache->mem_cache, pmn);
}
if (ncache->cached_frames)
ncache->cached_frames = MEM_dupallocN(cache->cached_frames);
}
/* hmm, should these be copied over instead? */
ncache->edit = NULL;
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
return ncache;
}
Fix #32667: Curve softbodies doesn't render animation (cycles) Issue was caused by cycles being duplicated curve objects before converting them to mesh. This duplication will loose pointcache which resulted in object not being properly deformed.
2012-09-27 14:37:20 +00:00
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
/* returns first point cache */
Refactor ID copying (and to some extent, ID freeing). This will allow much finer controll over how we copy data-blocks, from full copy in Main database, to "lighter" ones (out of Main, inside an already allocated datablock, etc.). This commit also transfers a llot of what was previously handled by per-ID-type custom code to generic ID handling code in BKE_library. Hopefully will avoid in future inconsistencies and missing bits we had all over the codebase in the past. It also adds missing copying handling for a few types, most notably Scene (which where using a fully customized handling previously). Note that the type of allocation used during copying (regular in Main, allocated but outside of Main, or not allocated by ID handling code at all) is stored in ID's, which allows to handle them correctly when freeing. This needs to be taken care of with caution when doing 'weird' unusual things with ID copying and/or allocation! As a final note, while rather noisy, this commit will hopefully not break too much existing branches, old 'API' has been kept for the main part, as a wrapper around new code. Cleaning it up will happen later. Design task : T51804 Phab Diff: D2714
2017-08-07 16:39:55 +02:00
PointCache *BKE_ptcache_copy_list(ListBase *ptcaches_new, const ListBase *ptcaches_old, const int flag)
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
{
PointCache *cache = ptcaches_old->first;
ListBase API: add utility api funcs for clearing and checking empty
2014-02-08 06:07:10 +11:00
BLI_listbase_clear(ptcaches_new);
Point Cache Refactoring ======================= Caching and Baking: - The point cache is now cleared on DAG_object_flush_update(), and not cleared for time dependency graph updates. - There is now a Bake button instead of Protect. Also cache start and end frames were added to softbody and particles. - The cloth autoprotect feature was removed. - The Ctrl+B menu now also bakes cloth and particles next to softbody and fluids. Additionally there are now frree bake and free cache menu entries. - The point cache api has been changed. There is now a PTCacheID struct for each point cache type that can be filled and then used to call the point cache functions. - PointCache struct was added to DNA and is automatically allocated for each physics type. - Soft body now supports Bake Editing just like cloth. - Tried to make the systems deal consistently with time ipo's and offsets. Still not sure it all works correct, but too complicated to solve completely now. Library Linking: - Added some more warnings to prevent editing settings on library linked objects. - Linked objects now read from the cache located next to the original library file, and never write to it. This restores old behavior for softbodies. For local simulation the mesh and not the object should be linked. - Dupligroups and proxies can't create local point caches at the moment, how to implement that I'm not sure. We probably need a proxy point cache for that to work (ugh). Physics UI: - Renamed deflection panel to collision for consistency and reorganized the buttons. Also removed some softbody collision buttons from the softbody panel that were duplicated in this panel for cloth. - Tweaked field panel buttons to not jump around when changing options. - Tabbing e.g. Soft Body Collision into the Soft Body panel, it now only shows Collision to make the panel names readable. - I tried to make enabled/disabling physics more consistent, since all three system did things different. Now the two modifier buttons to enable the modifier for the viewport and rendering are also duplicated in the physics panels. Toggling the Soft Body and Cloth buttons now both remove their modifiers. - Fixed modifier error drawing glitch. Particles: - Particles are now recalculated more often than before. Previously it did partial updates based on the changes, but that doesn't work well with DAG_object_flush_update() .. - Fixed memory leak loading keyed particle system. Now keys are not written to file anymore but always created after loading. - Make particle threads work with autothreads. Continue Physics: - The timeline play now has a Continue Physics option in the playback menu, which keeps the simulations going without writing them to the cache. - This doesn't always work that well, some changes are not immediately updated, but this can be improved later. Still it's fun to get a feel for the physics. Todo: - Point cache can get out of sync with and undo and changing a file without saving it. - Change the point cache file format to store a version (so old point cache files can be either converted or at least ignored), and to do correct endian conversion. - Menu item and/or buttons for Ctrl+B. - A system("rm ..") was changed to remove() since the former is very slow for clearing point caches. These system() calls were already giving trouble in a bug in the tracker, but really most use of this system("") should be changed and tested. - The Soft Body Collision and Clot Collision panel titles don't mention there's point cache settings there too, doing that makes them unreadable with the default panel setup.. but may need to make the names longer anyway.
2008-04-10 11:39:20 +00:00
Refactor ID copying (and to some extent, ID freeing). This will allow much finer controll over how we copy data-blocks, from full copy in Main database, to "lighter" ones (out of Main, inside an already allocated datablock, etc.). This commit also transfers a llot of what was previously handled by per-ID-type custom code to generic ID handling code in BKE_library. Hopefully will avoid in future inconsistencies and missing bits we had all over the codebase in the past. It also adds missing copying handling for a few types, most notably Scene (which where using a fully customized handling previously). Note that the type of allocation used during copying (regular in Main, allocated but outside of Main, or not allocated by ID handling code at all) is stored in ID's, which allows to handle them correctly when freeing. This needs to be taken care of with caution when doing 'weird' unusual things with ID copying and/or allocation! As a final note, while rather noisy, this commit will hopefully not break too much existing branches, old 'API' has been kept for the main part, as a wrapper around new code. Cleaning it up will happen later. Design task : T51804 Phab Diff: D2714
2017-08-07 16:39:55 +02:00
for (; cache; cache=cache->next) {
BLI_addtail(ptcaches_new, ptcache_copy(cache, (flag & LIB_ID_COPY_CACHES) != 0));
}
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
return ptcaches_new->first;
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Bugfix 30974 - Disabled "Quick Cache" option. It was causing ridiculous updates on the entire animation system for start-end frame on every user event causing changes here (like during transform). Worst was that you couldn't transform animated objects for that reason either. Most of the code is still there, waiting for a moment to revise it... - Constraint "Follow Track" (marker) wasn't using inverse matrix code in transform, making it wacko to use (wrong pivot, crazyspace)
2012-10-23 14:07:32 +00:00
/* Disabled this code; this is being called on scene_update_tagged, and that in turn gets called on
style cleanup: tabs & whitespace
2012-11-03 15:35:03 +00:00
* every user action changing stuff, and then it runs a complete bake??? (ton) */
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* Baking */
2.5: code changes to reduce the usage of G.main and pass it along or get it from the context instead.
2010-08-01 12:47:49 +00:00
void BKE_ptcache_quick_cache_all(Main *bmain, Scene *scene)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
PTCacheBaker baker;
Fix T47971: rigid body baking crash due to thread race condition.
2016-04-02 04:24:24 +02:00
memset(&baker, 0, sizeof(baker));
Cleanup: remove more G.main from BKE area.
2018-06-13 10:57:10 +02:00
baker.bmain = bmain;
Fix T47971: rigid body baking crash due to thread race condition.
2016-04-02 04:24:24 +02:00
baker.scene = scene;
baker.bake = 0;
baker.render = 0;
Rendering animations with particle trails cached.
2009-07-04 12:09:21 +00:00
baker.anim_init = 0;
Fix T47971: rigid body baking crash due to thread race condition.
2016-04-02 04:24:24 +02:00
baker.quick_step = scene->physics_settings.quick_cache_step;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
BKE_ptcache_bake(&baker);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
Better progress info for physics baking: * Using the job system for physics baking is not yet in the near future, so here's some good old console based progress info to all point cache based physics baking. * The info contains current total bake time, baking time for the current frame, and a simple estimate of completion time. * The info is only shown if the estimated total time for the bake is higher than one minute, so quick bakes don't suffer any performance hits due to console printing.
2011-03-11 02:00:05 +00:00
static void ptcache_dt_to_str(char *str, double dtime)
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (dtime > 60.0) {
if (dtime > 3600.0)
Better progress info for physics baking: * Using the job system for physics baking is not yet in the near future, so here's some good old console based progress info to all point cache based physics baking. * The info contains current total bake time, baking time for the current frame, and a simple estimate of completion time. * The info is only shown if the estimated total time for the bake is higher than one minute, so quick bakes don't suffer any performance hits due to console printing.
2011-03-11 02:00:05 +00:00
sprintf(str, "%ih %im %is", (int)(dtime/3600), ((int)(dtime/60))%60, ((int)dtime) % 60);
else
sprintf(str, "%im %is", ((int)(dtime/60))%60, ((int)dtime) % 60);
}
else
sprintf(str, "%is", ((int)dtime) % 60);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* if bake is not given run simulations to current frame */
style cleanup
2013-03-08 04:00:06 +00:00
void BKE_ptcache_bake(PTCacheBaker *baker)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
{
Cleanup: remove more G.main from BKE area.
2018-06-13 10:57:10 +02:00
Main *bmain = baker->bmain;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Scene *scene = baker->scene;
no functional changes: SETLOOPER macro assumed a scene was defined called 'sce' used to loop over, now make this an argument, helps to make it clear what's going on.
2010-12-17 15:37:59 +00:00
Scene *sce_iter; /* SETLOOPER macro only */
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Base *base;
ListBase pidlist;
Fix T47971: rigid body baking crash due to thread race condition.
2016-04-02 04:24:24 +02:00
PTCacheID *pid = &baker->pid;
2.5 - Warning cleanups (for mingw+scons) Also, made the Outliner's horizontal scrollbar work better for keymaps view. It's still using an approximation of the width, but at least you can scroll now.
2009-08-26 00:38:43 +00:00
PointCache *cache = NULL;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
float frameleno = scene->r.framelen;
int cfrao = CFRA;
Point Cache bake: make sure endframe is initialize with a decent value. This is how the old code used to initialize it. Current value was failing big time when baking all caches (always set to MAXFRAME), and it was added right before commit to quiet a warning... (I know, I know...)
2016-01-17 03:13:15 +01:00
int startframe = MAXFRAME, endframe = baker->anim_init ? scene->r.sfra : CFRA;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
int bake = baker->bake;
int render = baker->render;
Smoke: Patch #20955 / fix by Damien Plisson (damien78) * Smoke got it's own thread loop now and can be esc'ed now * Apple OpenMP bug workaround until they fix the GCC (couldn't test this one) * removing some "static" keyword to prevent crashing on frame display * should also fix problems with wavelet.noise crashing (Bug #20710) Fluid: * sleep delay has been reduced to be more responsive * Fluid progress is displayed in percent using the mouse cursor (like smoke)
2010-02-07 14:25:09 +00:00
Code cleanup: remove TRUE/FALSE & WITH_BOOL_COMPAT define
2014-04-01 11:34:00 +11:00
G.is_break = false;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* set caches to baking mode and figure out start frame */
Fix T47971: rigid body baking crash due to thread race condition.
2016-04-02 04:24:24 +02:00
if (pid->ob) {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* cache/bake a single object */
cache = pid->cache;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if ((cache->flag & PTCACHE_BAKED)==0) {
if (pid->type==PTCACHE_TYPE_PARTICLES) {
only override start and end frames with particle's if the partices are emitters. Was very confusing for hair baking.
2010-04-22 16:23:44 +00:00
ParticleSystem *psys= pid->calldata;
/* a bit confusing, could make this work better in the UI */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (psys->part->type == PART_EMITTER)
only override start and end frames with particle's if the partices are emitters. Was very confusing for hair baking.
2010-04-22 16:23:44 +00:00
psys_get_pointcache_start_end(scene, pid->calldata, &cache->startframe, &cache->endframe);
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (pid->type == PTCACHE_TYPE_SMOKE_HIGHRES) {
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
/* get all pids from the object and search for smoke low res */
ListBase pidlist2;
PTCacheID *pid2;
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
BKE_ptcache_ids_from_object(bmain, &pidlist2, pid->ob, scene, MAX_DUPLI_RECUR);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (pid2=pidlist2.first; pid2; pid2=pid2->next) {
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (pid2->type == PTCACHE_TYPE_SMOKE_DOMAIN) {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid2->cache && !(pid2->cache->flag & PTCACHE_BAKED)) {
if (bake || pid2->cache->flag & PTCACHE_REDO_NEEDED)
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
BKE_ptcache_id_clear(pid2, PTCACHE_CLEAR_ALL, 0);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (bake) {
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
pid2->cache->flag |= PTCACHE_BAKING;
pid2->cache->flag &= ~PTCACHE_BAKED;
}
}
}
}
Unified effector functionality for particles, cloth and softbody * Unified scene wide gravity (currently in scene buttons) instead of each simulation having it's own gravity. * Weight parameters for all effectors and an effector group setting. * Every effector can use noise. * Most effectors have "shapes" point, plane, surface, every point. - "Point" is most like the old effectors and uses the effector location as the effector point. - "Plane" uses the closest point on effectors local xy-plane as the effector point. - "Surface" uses the closest point on an effector object's surface as the effector point. - "Every Point" uses every point in a mesh effector object as an effector point. - The falloff is calculated from this point, so for example with "surface" shape and "use only negative z axis" it's possible to apply force only "inside" the effector object. * Spherical effector is now renamed as "force" as it's no longer just spherical. * New effector parameter "flow", which makes the effector act as surrounding air velocity, so the resulting force is proportional to the velocity difference of the point and "air velocity". For example a wind field with flow=1.0 results in proper non-accelerating wind. * New effector fields "turbulence", which creates nice random flow paths, and "drag", which slows the points down. * Much improved vortex field. * Effectors can now effect particle rotation as well as location. * Use full, or only positive/negative z-axis to apply force (note. the z-axis is the surface normal in the case of effector shape "surface") * New "force field" submenu in add menu, which adds an empty with the chosen effector (curve object for corve guides). * Other dynamics should be quite easy to add to the effector system too if wanted. * "Unified" doesn't mean that force fields give the exact same results for particles, softbody & cloth, since their final effect depends on many external factors, like for example the surface area of the effected faces. Code changes * Subversion bump for correct handling of global gravity. * Separate ui py file for common dynamics stuff. * Particle settings updating is flushed with it's id through DAG_id_flush_update(..). Known issues * Curve guides don't yet have all ui buttons in place, but they should work none the less. * Hair dynamics don't yet respect force fields. Other changes * Particle emission defaults now to frames 1-200 with life of 50 frames to fill the whole default timeline. * Many particles drawing related crashes fixed. * Sometimes particles didn't update on first frame properly. * Hair with object/group visualization didn't work properly. * Memory leaks with PointCacheID lists (Genscher, remember to free pidlists after use :).
2009-09-30 22:10:14 +00:00
BLI_freelistN(&pidlist2);
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (bake || cache->flag & PTCACHE_REDO_NEEDED)
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
startframe = MAX2(cache->last_exact, cache->startframe);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (bake) {
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
endframe = cache->endframe;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
cache->flag |= PTCACHE_BAKING;
}
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
else {
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
endframe = MIN2(endframe, cache->endframe);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
cache->flag &= ~PTCACHE_BAKED;
}
}
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
else {
for (SETLOOPER(scene, sce_iter, base)) {
/* cache/bake everything in the scene */
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
BKE_ptcache_ids_from_object(bmain, &pidlist, base->object, scene, MAX_DUPLI_RECUR);
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
for (pid=pidlist.first; pid; pid=pid->next) {
cache = pid->cache;
if ((cache->flag & PTCACHE_BAKED)==0) {
if (pid->type==PTCACHE_TYPE_PARTICLES) {
ParticleSystem *psys = (ParticleSystem*)pid->calldata;
/* skip hair & keyed particles */
if (psys->part->type == PART_HAIR || psys->part->phystype == PART_PHYS_KEYED)
continue;
psys_get_pointcache_start_end(scene, pid->calldata, &cache->startframe, &cache->endframe);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Rigidbody: Fix regression introduced in ee3fadd Baking rigid body cache was broken if some cached frames already existed. This is just a band aid for release, the logic need to be looked into further.
2017-09-05 19:53:32 +02:00
// XXX workaround for regression inroduced in ee3fadd, needs looking into
if (pid->type == PTCACHE_TYPE_RIGIDBODY) {
if ((cache->flag & PTCACHE_REDO_NEEDED || (cache->flag & PTCACHE_SIMULATION_VALID)==0) && (render || bake)) {
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
}
}
else if (((cache->flag & PTCACHE_BAKED) == 0) && (render || bake)) {
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
startframe = MIN2(startframe, cache->startframe);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
if (bake || render) {
cache->flag |= PTCACHE_BAKING;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
if (bake)
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
endframe = MAX2(endframe, cache->endframe);
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
cache->flag &= ~PTCACHE_BAKED;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
BLI_freelistN(&pidlist);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
}
Smoke: Patch #20955 / fix by Damien Plisson (damien78) * Smoke got it's own thread loop now and can be esc'ed now * Apple OpenMP bug workaround until they fix the GCC (couldn't test this one) * removing some "static" keyword to prevent crashing on frame display * should also fix problems with wavelet.noise crashing (Bug #20710) Fluid: * sleep delay has been reduced to be more responsive * Fluid progress is displayed in percent using the mouse cursor (like smoke)
2010-02-07 14:25:09 +00:00
CFRA = startframe;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
scene->r.framelen = 1.0;
Temporary fix for "freezing when pointcache is baked": * WM_timecursor is broken somehow, so pointcache baking makes the cursor disappear. * For user this seems like blender has frozen although it's just a matter of no progress indication. * This fix just sets the default "busy" cursor for the whole duration of baking and reports progress in the console. * If there's and easy fix for this then it should probably be done straight away, but I want to re-implement baking using the job system soon anyways, so a proper fix for this is not strictly necessary.
2010-11-04 20:17:38 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
/* bake */
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
bool use_timer = false;
double stime, ptime, ctime, fetd;
char run[32], cur[32], etd[32];
int cancel = 0;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
stime = ptime = PIL_check_seconds_timer();
dont use a thread for baking in background mode, its not really any advantage since it starts a single thread that runs a loop.
2010-05-18 14:38:25 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
for (int fr = CFRA; fr <= endframe; fr += baker->quick_step, CFRA = fr) {
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
BKE_scene_update_for_newframe(bmain->eval_ctx, bmain, scene, scene->lay);
dont use a thread for baking in background mode, its not really any advantage since it starts a single thread that runs a loop.
2010-05-18 14:38:25 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
if (baker->update_progress) {
float progress = ((float)(CFRA - startframe)/(float)(endframe - startframe));
baker->update_progress(baker->bake_job, progress, &cancel);
}
dont use a thread for baking in background mode, its not really any advantage since it starts a single thread that runs a loop.
2010-05-18 14:38:25 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
if (G.background) {
printf("bake: frame %d :: %d\n", CFRA, endframe);
}
else {
ctime = PIL_check_seconds_timer();
fetd = (ctime - ptime) * (endframe - CFRA) / baker->quick_step;
if (use_timer || fetd > 60.0) {
use_timer = true;
ptcache_dt_to_str(cur, ctime - ptime);
ptcache_dt_to_str(run, ctime - stime);
ptcache_dt_to_str(etd, fetd);
printf("Baked for %s, current frame: %i/%i (%.3fs), ETC: %s\r",
run, CFRA - startframe + 1, endframe - startframe + 1, ctime - ptime, etd);
dont use a thread for baking in background mode, its not really any advantage since it starts a single thread that runs a loop.
2010-05-18 14:38:25 +00:00
}
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
ptime = ctime;
}
/* NOTE: breaking baking should leave calculated frames in cache, not clear it */
if ((cancel || G.is_break)) {
break;
Smoke: Patch #20955 / fix by Damien Plisson (damien78) * Smoke got it's own thread loop now and can be esc'ed now * Apple OpenMP bug workaround until they fix the GCC (couldn't test this one) * removing some "static" keyword to prevent crashing on frame display * should also fix problems with wavelet.noise crashing (Bug #20710) Fluid: * sleep delay has been reduced to be more responsive * Fluid progress is displayed in percent using the mouse cursor (like smoke)
2010-02-07 14:25:09 +00:00
}
Null check for baking progressend function Simplify end of line for console progress.
2009-12-04 19:08:07 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
CFRA += 1;
dont use a thread for baking in background mode, its not really any advantage since it starts a single thread that runs a loop.
2010-05-18 14:38:25 +00:00
}
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
if (use_timer) {
/* start with newline because of \r above */
ptcache_dt_to_str(run, PIL_check_seconds_timer()-stime);
printf("\nBake %s %s (%i frames simulated).\n", (cancel ? "canceled after" : "finished in"), run, CFRA - startframe);
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
/* clear baking flag */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid) {
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
cache->flag &= ~(PTCACHE_BAKING|PTCACHE_REDO_NEEDED);
cache->flag |= PTCACHE_SIMULATION_VALID;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (bake) {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
cache->flag |= PTCACHE_BAKED;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* write info file */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_DISK_CACHE)
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
BKE_ptcache_write(pid, 0);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
else {
for (SETLOOPER(scene, sce_iter, base)) {
Cleanup: moar ugly G.main removal... Note that due to RNA get/setters issue, that one may actually add some G.main usages to the total... But at least it's not hidden anymore in a very low-level, dark corner of BKE pointcache code!
2018-06-13 14:26:26 +02:00
BKE_ptcache_ids_from_object(bmain, &pidlist, base->object, scene, MAX_DUPLI_RECUR);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
for (pid=pidlist.first; pid; pid=pid->next) {
/* skip hair particles */
if (pid->type==PTCACHE_TYPE_PARTICLES && ((ParticleSystem*)pid->calldata)->part->type == PART_HAIR)
continue;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
cache = pid->cache;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
Point Cache: use job system for bake operators. Reviewers: brecht Differential Revision: https://developer.blender.org/D1731
2016-01-17 02:01:41 +01:00
if (baker->quick_step > 1)
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
cache->flag &= ~(PTCACHE_BAKING|PTCACHE_OUTDATED);
else
cache->flag &= ~(PTCACHE_BAKING|PTCACHE_REDO_NEEDED);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
cache->flag |= PTCACHE_SIMULATION_VALID;
if (bake) {
cache->flag |= PTCACHE_BAKED;
if (cache->flag & PTCACHE_DISK_CACHE)
BKE_ptcache_write(pid, 0);
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
style cleanup: dont use 'else for' for extended blocks of code (quite confusing)
2012-03-25 23:12:42 +00:00
BLI_freelistN(&pidlist);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
}
scene->r.framelen = frameleno;
CFRA = cfrao;
* Fix for incorrect disabling after baking cloth sim * Fix for time cursor getting 'stuck' after baking point caches
2009-12-04 04:28:50 +00:00
Threaded object update and EvaluationContext Summary: Made objects update happening from multiple threads. It is a task-based scheduling system which uses current dependency graph for spawning new tasks. This means threading happens on object level, but the system is flexible enough for higher granularity. Technical details: - Uses task scheduler which was recently committed to trunk (that one which Brecht ported from Cycles). - Added two utility functions to dependency graph: * DAG_threaded_update_begin, which is called to initialize threaded objects update. It will also schedule root DAG node to the queue, hence starting evaluation process. Initialization will calculate how much parents are to be evaluation before current DAG node can be scheduled. This value is used by task threads for faster detecting which nodes might be scheduled. * DAG_threaded_update_handle_node_updated which is called from task thread function when node was fully handled. This function decreases num_pending_parents of node children and schedules children with zero valency. As it might have become clear, task thread receives DAG nodes and decides which callback to call for it. Currently only BKE_object_handle_update is called for object nodes. In the future it'll call node->callback() from Ali's new DAG. - This required adding some workarounds to the render pipeline. Mainly to stop using get_object_dm() from modifiers' apply callback. Such a call was only a workaround for dependency graph glitch when rendering scene with, say, boolean modifiers before displaying this scene. Such change moves workaround from one place to another, so overall hackentropy remains the same. - Added paradigm of EvaluaitonContext. Currently it's more like just a more reliable replacement for G.is_rendering which fails in some circumstances. Future idea of this context is to also store all the local data needed for objects evaluation such as local time, Copy-on-Write data and so. There're two types of EvaluationContext: * Context used for viewport updated and owned by Main. In the future this context might be easily moved to Window or Screen to allo per-window/per-screen local time. * Context used by render engines to evaluate objects for render purposes. Render engine is an owner of this context. This context is passed to all object update routines. Reviewers: brecht, campbellbarton Reviewed By: brecht CC: lukastoenne Differential Revision: https://developer.blender.org/D94
2013-12-26 17:24:42 +06:00
if (bake) { /* already on cfra unless baking */
BKE_scene_update_for_newframe(bmain->eval_ctx, bmain, scene, scene->lay);
}
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
/* TODO: call redraw all windows somehow */
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* Helpers */
void BKE_ptcache_disk_to_mem(PTCacheID *pid)
{
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
PointCache *cache = pid->cache;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheMem *pm = NULL;
int baked = cache->flag & PTCACHE_BAKED;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int cfra, sfra = cache->startframe, efra = cache->endframe;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* Remove possible bake flag to allow clear */
cache->flag &= ~PTCACHE_BAKED;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* PTCACHE_DISK_CACHE flag was cleared already */
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* restore possible bake flag */
cache->flag |= baked;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (cfra=sfra; cfra <= efra; cfra++) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
pm = ptcache_disk_frame_to_mem(pid, cfra);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pm)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
BLI_addtail(&pid->cache->mem_cache, pm);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
void BKE_ptcache_mem_to_disk(PTCacheID *pid)
{
PointCache *cache = pid->cache;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheMem *pm = cache->mem_cache.first;
int baked = cache->flag & PTCACHE_BAKED;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* Remove possible bake flag to allow clear */
cache->flag &= ~PTCACHE_BAKED;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* PTCACHE_DISK_CACHE flag was set already */
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* restore possible bake flag */
cache->flag |= baked;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; pm; pm=pm->next) {
if (ptcache_mem_frame_to_disk(pid, pm)==0) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
cache->flag &= ~PTCACHE_DISK_CACHE;
break;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
/* write info file */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_BAKED)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
BKE_ptcache_write(pid, 0);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
void BKE_ptcache_toggle_disk_cache(PTCacheID *pid)
{
PointCache *cache = pid->cache;
int last_exact = cache->last_exact;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
style cleanup (mostly whitespace)
2012-02-27 10:35:39 +00:00
if (!G.relbase_valid) {
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
cache->flag &= ~PTCACHE_DISK_CACHE;
move debug flag into its own global var (G.debug), split up debug options. --debug --debug-ffmpeg --debug-python --debug-events --debug-wm This makes debug output easier to read - event debug prints would flood output too much before. For convenience: --debug-all turns all debug flags on (works as --debug did before). also removed some redundant whitespace in debug prints and prefix some prints with __func__ to give some context.
2012-03-31 00:59:17 +00:00
if (G.debug & G_DEBUG)
Fix [#22199] Cloth Cache Panel > Disk Cache doesn't work Condition for this to work (.blend file must be saved) was poorly communicated in the UI (printfs are no good for this - ideally should use reports). Tweaked this a bit.
2010-05-03 03:02:27 +00:00
printf("File must be saved before using disk cache!\n");
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return;
Pointcache refresh part 1: * Particles support larger than 1 frame changes, bigger frame changes can result in inaccurate results, but it's super fast and you get a nice feeling of how the particles behave! * "Cache to current frame" button calculates the exact result of particles at current frame. * Current state of cache can be protected by making it a bake. * Cache is now in memory by default, disk cache is an option. * Only "viewport %" number of particles are calculated and cached in viewport, baking and rendering calculate all particles. * Info on cached frames and memory usage given in ui. * Support for exact "autocaching" of changes and large frame changes(disabled for now until exact place in event system is decided) * "Continue physics" is probably deprecated after this and should be removed once sb & cloth use the new cache code. Todo: * Make softbody & cloth use the new cache things. Other changes: * Some cleanup of particle buttons.
2009-06-21 10:16:52 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->cached_frames) {
Small changes to new pointcache code: * Removed the unused fuctions spotted by Campbel. * Removed the gotos to make Ton happy. * Added better debug prints to disk cache operations. * Fixed a memory unmap error that seemed to happen on a test file.
2011-01-02 18:10:33 +00:00
MEM_freeN(cache->cached_frames);
cache->cached_frames=NULL;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_DISK_CACHE)
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
BKE_ptcache_mem_to_disk(pid);
else
BKE_ptcache_disk_to_mem(pid);
cache->flag ^= PTCACHE_DISK_CACHE;
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
cache->flag ^= PTCACHE_DISK_CACHE;
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
cache->last_exact = last_exact;
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
BKE_ptcache_id_time(pid, NULL, 0.0f, NULL, NULL, NULL);
Pointcache refresh part 2 * Based on what happens during simulation the cache is marked (also in cache panel, this could possibly be extended to 3d view as well) as: - exact (not marked) - outdated (simulation is not done completely with current settings) - non-exact (frames were skipped during simulation) * The parameter "cache step" effects the number of frames between saved cache frames. - This can save a lot of memory (or disk space) if absolutely frame accurate simulation is not required. - Speeds up the "quick caching" very much. - Frames between cached frames are interpolated from the cached frames. - Current default value of 10 frames works nicely with up/down-arrows (skip 10 frames forwards/backwards on timeline), but can be changed if wanted. * The caching can work in normal or "quick" mode: [Normal cache] - Basic: Calculate what even happens (settings change, big frame steps etc.) and cache results, if possible try to use "cache step" when saving cache frames. - Becomes non-exact: After larger than 1 frame steps. - Becomes outdated: After any change effecting the simulation other than frame steps. - Pros/cons: Freedom of doing anything and playing with particles, but exact results have to calculated from the beginning. [Quick cache] - Basic: Calculate simulation up to current frame automatically on changes with cache step sized jumps in simulation. With multiple "quick cached" simulations the smallest cache step is used. - Becomes non-exact: Always from frame 1 (unless cache step = 1). - Becomes outdated: Never. - Pros/cons: Not very accurate, but super fast! - Todo: Transform of any animated (non-autokeyed) object is locked! Probably needs some tinkering with anim sys overrides. * The simulation can be run forwards or backwards even if it's cache is outdated or non-exact, the following rules apply in these situations: - step forwards (to unknown) -> simulate from last exact frame, store result - step backwards (to known) -> result is interpolated from existing frames, store result, clear cache forwards if current frame is after last exact frame * "Calculate to current frame" runs the simulation from start to current frame with a frame steps of 1. - Baking does the same, but runs the simulation all the way to the end of simulation. - Rendering does this automatically if the simulation is outdated of non-exact, so all rendered simulations will always be updated and exact. * Every cache panel also holds buttons to "Bake all dynamics", "Free all dynamics" and "Update all dynamics to current frame". * Cloth simulation supports the new cache too.
2009-06-27 15:28:58 +00:00
BKE_ptcache_update_info(pid);
Fix T35328: Disk caches of multiple particle systems on a single object overwrite each other It was intended to work actually using particle cache's stack index but this index might have been calculated incorrect in special case: * With default cube scene, add particle system to the cube * Add disk cache to the particle system * Save file and reload it * Add another particle system and enable disk cache This would lead to two point caches with the same stack index of zero. This happened because point cache indices list wasn't stored in the .blend file so once you've reload your file blender doesn't know anything about number or point caches used. And what was even more confusing is that point cache indices list was trying to be load from the file, but this failed because it wasn't in the file. This commit solves the root of the issue which is ability of producing .blend file with two point caches using the same disk cache. This is done by making it sure that point cache indices list is stored in the .blend file. And also made it so disabling disk cache will tag it to recalculate stack index. Old broken files wouldn't magically start working, but fixing them is rather simple manually by toggling Disk Cache option. Reviewers: lukastoenne, brecht CC: sergof Differential Revision: https://developer.blender.org/D286
2014-02-05 23:46:10 +06:00
if ((cache->flag & PTCACHE_DISK_CACHE) == 0) {
if (cache->index) {
BKE_object_delete_ptcache(pid->ob, cache->index);
cache->index = -1;
}
}
2.5: warning fixes Mostly harmless ones, except for one about "gzopen64" being undeclared. This needs some defines in BLI_storage.h to be set before <unistd.h> is included. Might fix a crash in compressed file reading, though I'm not sure since it's hard to repeat the crash consistently.
2009-06-22 18:19:18 +00:00
}
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
use (const char*) rather than (char*) where possible. also removed some unused function definitons.
2011-11-05 11:04:28 +00:00
void BKE_ptcache_disk_cache_rename(PTCacheID *pid, const char *name_src, const char *name_dst)
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
{
char old_name[80];
int len; /* store the length of the string */
/* mode is same as fopen's modes */
DIR *dir;
struct dirent *de;
char path[MAX_PTCACHE_PATH];
char old_filename[MAX_PTCACHE_FILE];
char new_path_full[MAX_PTCACHE_FILE];
char old_path_full[MAX_PTCACHE_FILE];
char ext[MAX_PTCACHE_PATH];
/* save old name */
strcpy() --> BLI_strncpy(), where source strings are not fixed and target size is known.
2011-10-19 23:10:54 +00:00
BLI_strncpy(old_name, pid->cache->name, sizeof(old_name));
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
/* get "from" filename */
use (const char*) rather than (char*) where possible. also removed some unused function definitons.
2011-11-05 11:04:28 +00:00
BLI_strncpy(pid->cache->name, name_src, sizeof(pid->cache->name));
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
len = ptcache_filename(pid, old_filename, 0, 0, 0); /* no path */
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
ptcache_path(pid, path);
dir = opendir(path);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (dir==NULL) {
strcpy() --> BLI_strncpy(), where source strings are not fixed and target size is known.
2011-10-19 23:10:54 +00:00
BLI_strncpy(pid->cache->name, old_name, sizeof(pid->cache->name));
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
return;
}
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const char *fext = ptcache_file_extension(pid);
BLI_snprintf(ext, sizeof(ext), "_%02u%s", pid->stack_index, fext);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
/* put new name into cache */
use (const char*) rather than (char*) where possible. also removed some unused function definitons.
2011-11-05 11:04:28 +00:00
BLI_strncpy(pid->cache->name, name_dst, sizeof(pid->cache->name));
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (STREQLEN(old_filename, de->d_name, len)) { /* do we have the right prefix */
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
/* read the number of the file */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const int frame = ptcache_frame_from_filename(de->d_name, ext);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (frame != -1) {
enforce string limits (reported by pedantic checking tools & some developers). mostly replace strcpy with BLI_strncpy and multiple strcat's with a BLI_snprintf(). also fix possible crash if CWD isnt available.
2011-02-13 03:21:27 +00:00
BLI_join_dirfile(old_path_full, sizeof(old_path_full), path, de->d_name);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
ptcache_filename(pid, new_path_full, frame, 1, 1);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
BLI_rename(old_path_full, new_path_full);
}
}
}
}
- resource leak in pointcache, opendir without closedir() - reading pointcache was using an incorrect, always NULL variable. - commented NDof code, was giving warnings and isnt used now.
2011-02-12 09:09:52 +00:00
closedir(dir);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
use BLI_strncpy and BLI_snprintf when the size of the string is known. fix for sequencer unique naming which was missed with string length update.
2012-01-11 12:33:51 +00:00
BLI_strncpy(pid->cache->name, old_name, sizeof(pid->cache->name));
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
}
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
void BKE_ptcache_load_external(PTCacheID *pid)
{
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/*todo*/
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
PointCache *cache = pid->cache;
int len; /* store the length of the string */
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int info = 0;
Fix for [#24597] Option External in Smoke cache affects settings of start and end frame of simulation * Don't change anything in the pointcache unless a valid external cache is found.
2010-11-08 11:19:53 +00:00
int start = MAXFRAME;
int end = -1;
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
/* mode is same as fopen's modes */
DIR *dir;
struct dirent *de;
char path[MAX_PTCACHE_PATH];
char filename[MAX_PTCACHE_FILE];
char ext[MAX_PTCACHE_PATH];
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (!cache)
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
return;
ptcache_path(pid, path);
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
len = ptcache_filename(pid, filename, 1, 0, 0); /* no path */
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
dir = opendir(path);
if (dir==NULL)
return;
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const char *fext = ptcache_file_extension(pid);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->index >= 0)
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_snprintf(ext, sizeof(ext), "_%02d%s", cache->index, fext);
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
else
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
BLI_strncpy(ext, fext, sizeof(ext));
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
Cleanup: strcmp/strncmp -> STREQ/STREQLEN (in boolean usage). Makes usage of those funcs much more clear, we even had mixed '!strcmp(foo, bar)' and 'strcmp(foo, bar) == 0' in several places...
2015-01-26 16:03:11 +01:00
if (STREQLEN(filename, de->d_name, len)) { /* do we have the right prefix */
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
/* read the number of the file */
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
const int frame = ptcache_frame_from_filename(de->d_name, ext);
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
Implementation of OpenVDB as a possible cache format for smoke simulations. This commits implements OpenVDB as an extra cache format in the Point Cache system for smoke simulations. Compilation with the library is turned off by default for now, and shall be enabled when the library is present. A documentation of its doings is available here: http:// wiki.blender.org/index.php/User:Kevindietrich/OpenVDBSmokeExport. A guide to compile OpenVDB can be found here (Linux): http:// wiki.blender.org/index.php?title=Dev:Doc/Building_Blender/Linux/ Dependencies_From_Source#OpenVDB Reviewers: sergey, lukastoenne, brecht, campbellbarton Reviewed By: brecht, campbellbarton Subscribers: galenb, Blendify, robocyte, Lapineige, bliblubli, jtheninja, lukasstockner97, dingto, brecht Differential Revision: https://developer.blender.org/D1721
2016-01-23 08:39:29 +01:00
if (frame != -1) {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (frame) {
Fix for [#24597] Option External in Smoke cache affects settings of start and end frame of simulation * Don't change anything in the pointcache unless a valid external cache is found.
2010-11-08 11:19:53 +00:00
start = MIN2(start, frame);
end = MAX2(end, frame);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
else
info = 1;
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
}
}
}
}
closedir(dir);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (start != MAXFRAME) {
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
PTCacheFile *pf;
Fix for [#24597] Option External in Smoke cache affects settings of start and end frame of simulation * Don't change anything in the pointcache unless a valid external cache is found.
2010-11-08 11:19:53 +00:00
cache->startframe = start;
cache->endframe = end;
cache->totpoint = 0;
style cleanup
2012-10-14 13:08:19 +00:00
if (pid->type == PTCACHE_TYPE_SMOKE_DOMAIN) {
/* necessary info in every file */
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
/* read totpoint from info file (frame 0) */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (info) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
pf= ptcache_file_open(pid, PTCACHE_FILE_READ, 0);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf) {
if (ptcache_file_header_begin_read(pf)) {
if (pf->type == pid->type && pid->read_header(pf)) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
cache->totpoint = pf->totpoint;
cache->flag |= PTCACHE_READ_INFO;
}
else {
cache->totpoint = 0;
}
}
ptcache_file_close(pf);
}
}
/* or from any old format cache file */
else {
float old_data[14];
Extreme makeover of pointcache code: * Pointcache code was quite ugly looking and complicated, so here are mostly just cosmetic adjustments, but some improved logic also. * Slight cleanup of pointcache ui too. * Shouldn't have any functional changes what so ever, so poke me right away if something seems off.
2010-12-18 15:03:31 +00:00
int elemsize = ptcache_old_elemsize(pid);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
pf= ptcache_file_open(pid, PTCACHE_FILE_READ, cache->startframe);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pf) {
while (ptcache_file_read(pf, old_data, 1, elemsize))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
cache->totpoint++;
ptcache_file_close(pf);
}
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
}
Fix for [#24597] Option External in Smoke cache affects settings of start and end frame of simulation * Don't change anything in the pointcache unless a valid external cache is found.
2010-11-08 11:19:53 +00:00
cache->flag |= (PTCACHE_BAKED|PTCACHE_DISK_CACHE|PTCACHE_SIMULATION_VALID);
cache->flag &= ~(PTCACHE_OUTDATED|PTCACHE_FRAMES_SKIPPED);
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
}
Fix [#35541]: Point cache doesn't load existing frames from disk if cache file name is changed.
2013-08-02 16:39:50 +00:00
/* make sure all new frames are loaded */
if (cache->cached_frames) {
MEM_freeN(cache->cached_frames);
cache->cached_frames=NULL;
}
External cache Particle point cache can now be loaded from external files. - Activated by "external" checkbox in cache panel and giving proper folder path and file name identifier. - External cache panel has controls for particle emission start, end, lifetime and random lifetime. These should be set according to the external data for correct playback. - External files should be named "identifier_frame_index.bphys" or "identifier_frame.bphys" where: * "identifier" is a freely choseable name. * "frame" is the cached frame number. ** Six digits padded with zeros!, for example "000024". * "index" can be used to tell caches with the same identifier apart. ** Two digits starting from zero. ** The index and the underscore before are optional. If no index is present the index number in ui should be set to -1. - Cache file format is pure floating point numbers (in binary, not text!) with each particle's data one after the other with the following data members: * 3 floats: particle's location vector * 3 floats: particle's velocity vector (per second) * 4 floats: particle's rotation quaternion * 3 floats: particle's angular velocity vector (per second) * 1 float: frame of the actual data (this can be non-integer for particles that are born or die between two integer frames, but otherwise should be the same as the "frame" in the file name) - Cache files don't have to exist for each frame. * Frames without actual data are interpolated from surrounding frames that have data (extrapolation is not supported). - Cache file formats with extended (or reduced even) data members are in future plans for easier usage. - Current code only does particles, don't yet know if it's applicable to cloth or sb. - Known issue: endianness can't yet be handled in any way. Other changes: New hard limits for many particle parameters. Some examples: - Maximum amount of particles: 10M particles :) And before you all go and crash your Blender trying this out remember that this limit is only for those freaks who really have the machine power to handle it. 10M particles alone take around 2.2 Gb of memory / disk space in saved file and each cached frame takes around 0.5 Gb of memory / disk space depending on cache mode. * Known issue: To actually use this many particles they most likely need to be allocated in parts as taking hold of a 2.2Gb chunk of memory at once is probably not ok with any operating system. - Maximum amount of children: 100k children/particle (1T childparticles here we come :D) - Kink frequency: -100k to 100k half-rotations (really strange the previous limit was only from zero upwards) - Path draw steps: 10 (power of 2 remember) - Path render steps: 20 (power of 2 also!! If over 1M segments doesn't get you smooth paths then I think nothing will!)
2009-07-23 00:19:01 +00:00
BKE_ptcache_update_info(pid);
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
void BKE_ptcache_update_info(PTCacheID *pid)
{
PointCache *cache = pid->cache;
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
PTCacheExtra *extra = NULL;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int totframes = 0;
char mem_info[64];
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_EXTERNAL) {
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
int cfra = cache->startframe;
style cleanup: also rename bmesh_decimate.c --> bmesh_decimate_collapse.c
2012-10-23 03:38:26 +00:00
for (; cfra <= cache->endframe; cfra++) {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (BKE_ptcache_id_exist(pid, cfra))
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
totframes++;
}
Fix for [#23732] Smoke / load external cache doesn't work
2010-09-28 10:04:41 +00:00
/* smoke doesn't use frame 0 as info frame so can't check based on totpoint */
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (pid->type == PTCACHE_TYPE_SMOKE_DOMAIN && totframes)
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(cache->info, sizeof(cache->info), IFACE_("%i frames found!"), totframes);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (totframes && cache->totpoint)
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(cache->info, sizeof(cache->info), IFACE_("%i points found!"), cache->totpoint);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
else
replace sprintf -> strcpy where its not needed.
2013-03-12 12:47:57 +00:00
BLI_strncpy(cache->info, IFACE_("No valid data to read!"), sizeof(cache->info));
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
return;
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_DISK_CACHE) {
style cleanup: changes to brace placement / newlines - for/while/if/switch
2012-04-28 06:31:57 +00:00
if (pid->type == PTCACHE_TYPE_SMOKE_DOMAIN) {
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
int totpoint = pid->totpoint(pid->calldata, 0);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->totpoint > totpoint)
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(mem_info, sizeof(mem_info), IFACE_("%i cells + High Resolution cached"), totpoint);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
else
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(mem_info, sizeof(mem_info), IFACE_("%i cells cached"), totpoint);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
else {
int cfra = cache->startframe;
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
style cleanup: also rename bmesh_decimate.c --> bmesh_decimate_collapse.c
2012-10-23 03:38:26 +00:00
for (; cfra <= cache->endframe; cfra++) {
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (BKE_ptcache_id_exist(pid, cfra))
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
totframes++;
}
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(mem_info, sizeof(mem_info), IFACE_("%i frames on disk"), totframes);
Smoke now uses only one point cache where both normal and high resolution smoke are stored together: * Separate caches were causing quite a lot of problems both in principle and practice. * For example it doesn't really make sense to have different frame ranges for normal and high resolution smoke, but this was fully possible before. * Also to fully bake the smoke you had to do a "Bake All Dynamics", which completely defeats the whole point of the feature! * As a result of this change the smoke cache usage is much much simpler and less error prone. * This is quite a big change, but hopefully there should be less rather than more problems as a result :) Some other related changes: * Changing the cache name now works for disk caches properly too, it now just renames the cache files so should be faster too! * Smoke is now always forced to disk cache with step 1 on file load as there were some strange cases where smoke was trying to use memory cache. * Disabled smoke debug prints from console. * Disabled changing smoke parameters when smoke is baked. Note to users: The unfortunate side effect of this is that old high resolution simulations have to be baked again, but in return you get much better and more logical functionality. Sorry none the less!
2010-11-30 21:31:18 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
else {
style cleanup: trailing tabs & expand some non prefix tabs into spaces.
2012-10-21 05:46:41 +00:00
PTCacheMem *pm = cache->mem_cache.first;
Add number and memory size formatting throughout the UI This commit adds number formatting (thousands separator) to the baking panel. It also adds a new function to format memory sizes (KB/GB/etc) and applies it to the baking panel and scene stats. The new function is unit tested. Reviewers: Severin Tags: #user_interface Differential Revision: https://developer.blender.org/D1248
2018-05-25 22:17:15 +02:00
char formatted_tot[16];
char formatted_mem[15];
long long int bytes = 0.0f;
int i;
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (; pm; pm=pm->next) {
for (i=0; i<BPHYS_TOT_DATA; i++)
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
bytes += MEM_allocN_len(pm->data[i]);
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
for (extra=pm->extradata.first; extra; extra=extra->next) {
Pointcache code cleanup and disk cache compression options: * Massive reorganization of pointcache code, things are now cleaner than ever. * For all but smoke the cache is first written to memory when using disk cache and after that written to disk in one operation. This allows less disk operations and the possibility to compress the data before writing it to disk. * Previously only smoke cache could be compressed, now the same options exist for other physics types too (when using disk cache). For now the default compression option is still "no compression", but if there aren't any problems this can be set to "light compression" as it's actually faster than no compression in most cases since there's less data to write to the disk. Based on quick tests heavy compression can reduce the file size down to 1/3rd of the original size, but is really slow compared to other options, so it should be used only if file size is critical! * The pointcache code wasn't really 64bit compatible (for disk cache) until now, so this update should fix some crashes on 64bit builds too. Now all integer data that's saved to disk uses 32 bit unsigned integers, so simulations done on 64bit should load fine on 32bit machines and vice versa. (Important disk cache simulations made on 64bit builds should be converted to memory cache in a revision before this commit). * There are also the beginnings of extradata handling code in pointcache in anticipation of adding the dynamic springs for particle fluids (the springs need to be stored as extradata into point cache). * Particles were being read from the cache with a slightly wrong framerate. In most cases this probably wasn't noticeable, but none the less the code is now correct in every way. * Small other fixes here and there & some cosmetic changes to cache panel, but over all there should be no functional changes other than the new disk cache compression options. * This whole re-organization also seems to fix bug #25436 and hopefully shouldn't introduce any new ones!
2011-01-02 06:52:47 +00:00
bytes += MEM_allocN_len(extra->data);
bytes += sizeof(PTCacheExtra);
}
bytes += sizeof(PTCacheMem);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
totframes++;
Point cache optimization: only cache particles that are alive. This reduces point cache sizes dramatically especially if particle life time is small compared to total simulation length. For example with the settings: particle amount = 10000, start = 1, end = 200, life = 10, cache step = 1, the unoptimized blend file size (compressed) was a little over 22 Mb and with this optimization the file is a little under 2 Mb (again compressed). In addition to saving memory/disk space this also probably speeds up reading from cache, since there's less data to read. As an additional fix the memory cache size (displayed in cache panel) is now calculated correctly.
2010-03-09 03:01:18 +00:00
}
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Add number and memory size formatting throughout the UI This commit adds number formatting (thousands separator) to the baking panel. It also adds a new function to format memory sizes (KB/GB/etc) and applies it to the baking panel and scene stats. The new function is unit tested. Reviewers: Severin Tags: #user_interface Differential Revision: https://developer.blender.org/D1248
2018-05-25 22:17:15 +02:00
BLI_str_format_int_grouped(formatted_tot, totframes);
BLI_str_format_byte_unit(formatted_mem, bytes, true);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
Add number and memory size formatting throughout the UI This commit adds number formatting (thousands separator) to the baking panel. It also adds a new function to format memory sizes (KB/GB/etc) and applies it to the baking panel and scene stats. The new function is unit tested. Reviewers: Severin Tags: #user_interface Differential Revision: https://developer.blender.org/D1248
2018-05-25 22:17:15 +02:00
BLI_snprintf(mem_info, sizeof(mem_info), IFACE_("%s frames in memory (%s)"), formatted_tot, formatted_mem);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache->flag & PTCACHE_OUTDATED) {
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
BLI_snprintf(cache->info, sizeof(cache->info), IFACE_("%s, cache is outdated!"), mem_info);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
else if (cache->flag & PTCACHE_FRAMES_SKIPPED) {
More UI messages fixes...
2013-02-28 15:31:20 +00:00
BLI_snprintf(cache->info, sizeof(cache->info), IFACE_("%s, not exact since frame %i"),
Another huge bunch of UI translation fixes, mostly reported by Leon Cheung, Sv.Lockal, Gabriel Gazzán and Satoshi Yamasaki, thanks!
2013-02-17 13:56:49 +00:00
mem_info, cache->last_exact);
use BLI_strncpy and BLI_snprintf when the size of the string is known. fix for sequencer unique naming which was missed with string length update.
2012-01-11 12:33:51 +00:00
}
else {
BLI_snprintf(cache->info, sizeof(cache->info), "%s.", mem_info);
New point cache file format: - HEADER (beginning of each file) * general header: + 8 char: "BPHYSICS" + 1 int: simulation type (same as PTCacheID->type) * custom header (same for sb, particles and cloth, but can be different for new dynamics) + 1 int: totpoint (number of points) + 1 int: data_types (bit flags for what the stored data is) - DATA (directly after header) *totpoint times the data as specified in data_types flags - simulation type soft body = 0, particles = 1, cloth = 2 - data types (more can be added easily when needed) data flag contains ---------------------------------------- index (1<<0) 1 int (index of current point) location (1<<1) 3 float velocity (1<<2) 3 float rotation (1<<3) 4 float (quaternion) avelocity (1<<4) 3 float (used for particles) xconst (1<<4) 3 float (used for cloth) size (1<<5) 1 float times (1<<6) 3 float (birth, die & lifetime of particle) boids (1<<7) 1 BoidData Notes: - Every frame is not nescessary since data is interpolated for the inbetween frames. - For now every point is needed for every cached frame, the "index" data type is reserved for future usage. - For loading external particle caches only "location" data is necessary, other needed values are determined from the given data. - Non-dynamic data should be written into an info file if external usage is desired. * Info file is named as normal cache files, but with frame number 0; * "Non-dynamic" means data such as particle times. * Written automatically when baking to disk so basically a library of particle simulations should be possible. - Old disk cache format is supported for reading, so pre 2.5 files shouldn't break. However old style memory cache (added during 2.5 development) is not supported. To keep memory cached simulations convert the cache to disk cache before svn update and save the blend. - External sb and cloth caches should be perfectly possible, but due to lack of testing these are not yet enabled in ui. Other changes: - Multiple point caches per dynamics system. * In the future these will hopefully be nla editable etc, but for now things are simple and the current (selected) point cache is used. * Changing the amount of cached points (for example particle count) is allowed, but might not give correct results if multiple caches are present. - Generalization of point cache baking etc operator & rna code. - Comb brushing particle hair didn't work smoothly.
2009-08-12 09:54:29 +00:00
}
Commenting out code from commit 22457 by Genscher for PointCache. There was some missing code or something causing compiling problems.
2009-08-15 00:40:44 +00:00
}
Big cleanup of particle system core, also some minor pointcache cleanup. Shouldn't contain any functional changes.
2010-03-21 20:36:06 +00:00
void BKE_ptcache_validate(PointCache *cache, int framenr)
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache) {
Fix for [#24237] Hair dynamics with zero particles generates a segmentation fault
2010-10-14 09:01:03 +00:00
cache->flag |= PTCACHE_SIMULATION_VALID;
cache->simframe = framenr;
}
Big cleanup of particle system core, also some minor pointcache cleanup. Shouldn't contain any functional changes.
2010-03-21 20:36:06 +00:00
}
void BKE_ptcache_invalidate(PointCache *cache)
{
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
if (cache) {
Fix for [#24237] Hair dynamics with zero particles generates a segmentation fault
2010-10-14 09:01:03 +00:00
cache->flag &= ~PTCACHE_SIMULATION_VALID;
cache->simframe = 0;
cache->last_exact = MIN2(cache->startframe, 0);
}
Purging compiler warnings
2010-03-26 03:10:58 +00:00
}
Copy Permalink