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This repository has been archived on 2023-10-09.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
2e9982f420cab3b2ecf4648c0a91403f757ef9fe
blender-archive/source/blender/blenkernel/intern/pointcache.c

3120 lines
83 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
/*
Clean up headers a bit more.
2011-02-23 18:03:40 +00:00
* $Id$
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
* ***** 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"
#include "DNA_cloth_types.h"
#include "DNA_modifier_types.h"
#include "DNA_object_types.h"
#include "DNA_object_force.h"
#include "DNA_particle_types.h"
#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"
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
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"
#include "WM_api.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"
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 "BKE_blender.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"
#include "BKE_depsgraph.h"
#include "BKE_global.h"
#include "BKE_library.h"
#include "BKE_main.h"
#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"
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 "BKE_utildefines.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
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 */
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"
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
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 "minilzo.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
#else
/* used for non-lzo cases */
#define LZO_OUT_LEN(size) ((size) + (size) / 16 + 64 + 3)
#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
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 */
* cleaning up warnings (mostly windows). A collection of other warning fixes too (undefined function, assuming int, etc.) This compiled fine with scons/msvc and scons/mingw (gcc 4.4.0). Please test and report any problems.
2009-09-06 13:20:05 +00:00
/* untitled blend's need getpid for a unique name */
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
#ifndef WIN32
#include <dirent.h>
* cleaning up warnings (mostly windows). A collection of other warning fixes too (undefined function, assuming int, etc.) This compiled fine with scons/msvc and scons/mingw (gcc 4.4.0). Please test and report any problems.
2009-09-06 13:20:05 +00:00
#include <unistd.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
* 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
#include <process.h>
* cleaning up warnings (mostly windows). A collection of other warning fixes too (undefined function, assuming int, etc.) This compiled fine with scons/msvc and scons/mingw (gcc 4.4.0). Please test and report any problems.
2009-09-06 13:20:05 +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
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
#define PTCACHE_DATA_FROM(data, type, from) if(data[type]) { memcpy(data[type], from, ptcache_data_size[type]); }
#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]); }
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);
static int ptcache_file_write(PTCacheFile *pf, void *f, unsigned int tot, unsigned int size);
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! */
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
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;
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
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! */
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
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;
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
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;
if(old_data) {
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;
if(cfra1 == cfra2)
return;
VECCOPY(keys[1].co, bp->pos);
VECCOPY(keys[1].vel, bp->vec);
if(old_data) {
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
VECCOPY(bp->pos, keys->co);
VECCOPY(bp->vec, keys->vel);
}
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;
}
/* 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 */
if(data[BPHYS_DATA_ROTATION]==NULL) {
vec_to_quat( key->rot, key->vel, OB_NEGX, OB_POSZ);
}
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 */
if(data[BPHYS_DATA_INDEX] && (cfra < pa->time - step || cfra > pa->dietime + step))
return 0;
use c90 compatible static initializers.
2010-11-03 06:31:53 +00:00
times[0]= pa->time;
times[1]= pa->dietime;
times[2]= pa->lifetime;
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
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
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);
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;
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
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
if(index >= psys->totpart)
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
if(cfra > pa->state.time)
memcpy(&pa->prev_state, &pa->state, sizeof(ParticleKey));
if(old_data){
/* 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 */
if(cfra < pa->time)
pa->state.time = pa->time;
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
else if(cfra > pa->dietime)
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
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(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);
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(data[BPHYS_DATA_TIMES]) {
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];
}
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
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);
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 */
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
}
}
/* determine rotation from velocity */
if(data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_ROTATION]) {
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
vec_to_quat( pa->state.rot,pa->state.vel, OB_NEGX, OB_POSZ);
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];
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
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
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
if(index >= psys->totpart)
return;
pa = psys->particles + index;
/* particle wasn't read from first cache so can't interpolate */
if((int)cfra1 < pa->time - psys->pointcache->step || (int)cfra1 > pa->dietime + psys->pointcache->step)
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);
if(cfra1 == cfra2)
return;
memcpy(keys+1, &pa->state, sizeof(ParticleKey));
if(old_data)
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 */
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
}
}
/* determine rotation from velocity */
if(data[BPHYS_DATA_LOCATION] && !data[BPHYS_DATA_ROTATION]) {
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
vec_to_quat( keys[2].rot,keys[2].vel, OB_NEGX, OB_POSZ);
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
}
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
if(cfra > pa->time)
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;
}
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;
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(cfra == 0)
return 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
for(p=0; p<psys->totpart; p++,pa++)
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;
if(psys->part->phystype == PART_PHYS_FLUID &&
psys->part->fluid && psys->part->fluid->flag & SPH_VISCOELASTIC_SPRINGS &&
psys->tot_fluidsprings && psys->fluid_springs) {
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;
for(; extra; extra=extra->next) {
switch(extra->type) {
case BPHYS_EXTRA_FLUID_SPRINGS:
{
if(psys->fluid_springs)
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;
if(old_data) {
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;
if(cfra1 == cfra2)
return;
VECCOPY(keys[1].co, vert->x);
VECCOPY(keys[1].vel, vert->v);
if(old_data) {
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
VECCOPY(vert->x, keys->co);
VECCOPY(vert->v, keys->vel);
/* 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;
Set Scenes and Physics - objects in a set scene now are evaluated with the frame from the current scene. - pointcache now loops over all set scene objects.
2010-04-02 14:57:12 +00:00
return clmd->clothObject ? clmd->clothObject->numverts : 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
/* 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;
if(sds->fluid) {
return sds->res[0]*sds->res[1]*sds->res[2];
}
else
return 0;
}
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;
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
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];
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
float dt, dx, *dens, *densold, *heat, *heatold, *vx, *vy, *vz, *vxold, *vyold, *vzold;
unsigned char *obstacles;
unsigned int in_len = sizeof(float)*(unsigned int)res;
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
smoke_export(sds->fluid, &dt, &dx, &dens, &densold, &heat, &heatold, &vx, &vy, &vz, &vxold, &vyold, &vzold, &obstacles);
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);
ptcache_file_compressed_write(pf, (unsigned char *)densold, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)heat, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)heatold, in_len, out, mode);
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 *)vxold, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)vyold, in_len, out, mode);
ptcache_file_compressed_write(pf, (unsigned char *)vzold, 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));
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
}
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
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];
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
float *dens, *densold, *tcu, *tcv, *tcw;
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
smoke_turbulence_export(sds->wt, &dens, &densold, &tcu, &tcv, &tcw);
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);
ptcache_file_compressed_write(pf, (unsigned char *)densold, 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
}
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_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;
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
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];
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
float dt, dx, *dens, *densold, *heat, *heatold, *vx, *vy, *vz, *vxold, *vyold, *vzold;
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
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_export(sds->fluid, &dt, &dx, &dens, &densold, &heat, &heatold, &vx, &vy, &vz, &vxold, &vyold, &vzold, &obstacles);
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);
ptcache_file_compressed_read(pf, (unsigned char*)dens, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)densold, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)heat, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)heatold, 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*)vxold, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)vyold, out_len);
ptcache_file_compressed_read(pf, (unsigned char*)vzold, 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));
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
if(pf->data_types & (1<<BPHYS_DATA_SMOKE_HIGH) && sds->wt) {
int res = sds->res[0]*sds->res[1]*sds->res[2];
int res_big, res_big_array[3];
float *dens, *densold, *tcu, *tcv, *tcw;
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
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_export(sds->wt, &dens, &densold, &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
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*)dens, out_len_big);
ptcache_file_compressed_read(pf, (unsigned char*)densold, out_len_big);
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 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*)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
}
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
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;
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
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;
}
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;
if(psys->part->type != PART_HAIR)
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;
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
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
if(psys->part->phystype == PART_PHYS_BOIDS)
pid->data_types|= (1<<BPHYS_DATA_AVELOCITY) | (1<<BPHYS_DATA_ROTATION) | (1<<BPHYS_DATA_BOIDS);
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
else if(psys->part->phystype == PART_PHYS_FLUID && psys->part->fluid && psys->part->fluid->flag & SPH_VISCOELASTIC_SPRINGS) {
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
if(psys->part->rotmode!=PART_ROT_VEL
|| psys->part->avemode!=PART_AVE_SPIN || psys->part->avefac!=0.0f)
pid->data_types|= (1<<BPHYS_DATA_AVELOCITY) | (1<<BPHYS_DATA_ROTATION);
if(psys->part->flag & PART_ROT_DYN)
pid->data_types|= (1<<BPHYS_DATA_ROTATION);
pid->info_types= (1<<BPHYS_DATA_TIMES);
}
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;
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;
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;
}
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;
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;
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;
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
if(sds->fluid)
pid->data_types |= (1<<BPHYS_DATA_SMOKE_LOW);
if(sds->wt)
pid->data_types |= (1<<BPHYS_DATA_SMOKE_HIGH);
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +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
void BKE_ptcache_ids_from_object(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;
if(ob->soft) {
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_softbody(pid, ob, ob->soft);
BLI_addtail(lb, pid);
}
for(psys=ob->particlesystem.first; psys; psys=psys->next) {
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
if(psys->part==NULL)
continue;
/* check to make sure point cache is actually used by the particles */
if(ELEM(psys->part->phystype, PART_PHYS_NO, PART_PHYS_KEYED))
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 */
/* if(psys->part->type == PART_HAIR && (psys->flag & PSYS_HAIR_DYNAMICS)==0) */
/* 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
if(psys->part->type == PART_FLUID)
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
}
for(md=ob->modifiers.first; md; md=md->next) {
if(md->type == eModifierType_Cloth) {
pid= MEM_callocN(sizeof(PTCacheID), "PTCacheID");
BKE_ptcache_id_from_cloth(pid, ob, (ClothModifierData*)md);
BLI_addtail(lb, pid);
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
if(md->type == eModifierType_Smoke) {
SmokeModifierData *smd = (SmokeModifierData *)md;
if(smd->type & MOD_SMOKE_TYPE_DOMAIN)
{
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);
}
}
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
}
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
if(scene && (duplis-- > 0) && (ob->transflag & OB_DUPLI)) {
ListBase *lb_dupli_ob;
if((lb_dupli_ob=object_duplilist(scene, ob))) {
DupliObject *dob;
for(dob= lb_dupli_ob->first; dob; dob= dob->next) {
bugfix [#22988] Duplicating Lights by Frame Causes Hangup
2010-07-24 02:51:51 +00:00
if(dob->ob != ob) { /* avoids recursive loops with dupliframes: bug 22988 */
ListBase lb_dupli_pid;
BKE_ptcache_ids_from_object(&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);
bugfix [#22988] Duplicating Lights by Frame Causes Hangup
2010-07-24 02:51:51 +00:00
if(lb_dupli_pid.first)
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 */
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
/* Takes an Object ID and returns a unique name
- id: object id
- cfra: frame for the cache, can be negative
- stack_index: index in the modifier stack. we can have cache for more then one stack_index
*/
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
#define MAX_PTCACHE_FILE ((FILE_MAXDIR+FILE_MAXFILE)*2)
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
{
misc fixes found with clang's static checker.
2010-10-07 10:04:07 +00:00
Library *lib= (pid->ob)? pid->ob->id.lib: NULL;
remove G.sce, use G.main->name instead. Both stored the filename of the blend file, but G.sce stored the last opened file. This will make blender act differently in some cases since a relative path to the last opened file will no longer resolve (which is correct IMHO since that file isnt open and the path might not even be valid anymore). Tested linking with durian files and rendering to relative paths when no files is loaded however we may need to have some operators give an error if they are used on the default startup.blend.
2010-10-18 06:41:16 +00:00
const char *blendfilename= (lib && (pid->cache->flag & PTCACHE_IGNORE_LIBPATH)==0) ? lib->filepath: G.main->name;
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
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(pid->cache->flag & PTCACHE_EXTERNAL) {
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
if(strncmp(filename, "//", 2)==0)
BLI_path_abs(filename, blendfilename);
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) {
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 file[MAX_PTCACHE_PATH]; /* we dont 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
image re-project now uses offscreen render function and has input for render size. unrelated changes that ended up being more trouble to commit separate... - removed BLI_split_dirfile(), was nasty, occasionaly modifying the source string, it could create directories and used the $CWD in some cases. was only used in 2 places in filesel.c, if this gives problems can address without bringing back this function. renamed BLI_split_dirfile_basic --> BLI_split_dirfile - view3d_operator_needs_opengl was being called for offscreen render when it wasnt needed.
2010-03-08 20:08:04 +00:00
BLI_split_dirfile(blendfilename, NULL, 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';
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
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 */
/* btempdir is assumed to exist and ALWAYS has a trailing slash */
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
snprintf(filename, MAX_PTCACHE_PATH, "%s"PTCACHE_PATH"%d", btempdir, abs(getpid()));
- 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;
}
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(strcmp(pid->cache->name, "")==0 && (pid->cache->flag & PTCACHE_EXTERNAL)==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
idname = (pid->ob->id.name+2);
/* convert chars to hex so they are always a valid filename */
while('\0' != *idname) {
snprintf(newname, MAX_PTCACHE_FILE, "%02X", (char)(*idname++));
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) {
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
if(pid->cache->index < 0)
Pointcache: *forgot 1 rename
2009-08-25 19:58:42 +00:00
pid->cache->index = pid->stack_index = 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
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(pid->cache->flag & PTCACHE_EXTERNAL) {
if(pid->cache->index >= 0)
snprintf(newname, MAX_PTCACHE_FILE, "_%06d_%02d"PTCACHE_EXT, cfra, pid->stack_index); /* always 6 chars */
else
snprintf(newname, MAX_PTCACHE_FILE, "_%06d"PTCACHE_EXT, cfra); /* always 6 chars */
}
else {
snprintf(newname, MAX_PTCACHE_FILE, "_%06d_%02d"PTCACHE_EXT, cfra, pid->stack_index); /* always 6 chars */
}
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;
char filename[(FILE_MAXDIR+FILE_MAXFILE)*2];
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 */
if(pid->ob->id.lib && mode == PTCACHE_FILE_WRITE)
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) {
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 (!BLI_exists(filename)) {
return NULL;
}
spaces -> tabs, (4 spaces == 1 tab, only for white space preceding text)
2010-03-22 09:30:00 +00:00
fp = fopen(filename, "rb");
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
} 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 */
fp = fopen(filename, "wb");
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
} else if (mode==PTCACHE_FILE_UPDATE) {
BLI_make_existing_file(filename);
fp = 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
}
spaces -> tabs, (4 spaces == 1 tab, only for white space preceding text)
2010-03-22 09:30:00 +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;
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
spaces -> tabs, (4 spaces == 1 tab, only for white space preceding text)
2010-03-22 09:30:00 +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
{
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
if(pf) {
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;
unsigned char *props = MEM_callocN(16*sizeof(char), "tmp");
ptcache_file_read(pf, &compressed, 1, sizeof(unsigned char));
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;
if(in_len==0) {
/* 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
if(compressed == 1)
r = lzo1x_decompress_safe(in, (lzo_uint)in_len, result, (lzo_uint *)&out_len, NULL);
#endif
#ifdef WITH_LZMA
if(compressed == 2)
{
quiet warning & minor pep8 edits.
2011-04-04 22:42:43 +00:00
size_t sizeOfIt;
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 leni = in_len, leno = out_len;
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;
unsigned char *props = MEM_callocN(16*sizeof(char), "tmp");
size_t sizeOfIt = 5;
(void)mode; /* unused when building w/o compression */
#ifdef WITH_LZO
out_len= LZO_OUT_LEN(in_len);
if(mode == 1) {
LZO_HEAP_ALLOC(wrkmem, LZO1X_MEM_COMPRESS);
r = lzo1x_1_compress(in, (lzo_uint)in_len, out, (lzo_uint *)&out_len, wrkmem);
if (!(r == LZO_E_OK) || (out_len >= in_len))
compressed = 0;
else
compressed = 1;
}
#endif
#ifdef WITH_LZMA
if(mode == 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
r = LzmaCompress(out, &out_len, in, in_len,//assume sizeof(char)==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
props, &sizeOfIt, 5, 1 << 24, 3, 0, 2, 32, 2);
if(!(r == SZ_OK) || (out_len >= in_len))
compressed = 0;
else
compressed = 2;
}
#endif
ptcache_file_write(pf, &compressed, 1, sizeof(unsigned char));
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));
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
}
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_write(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 (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
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
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
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;
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
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;
if(fread(bphysics, sizeof(char), 8, pf->fp) != 8)
error = 1;
if(!error && strncmp(bphysics, "BPHYSICS", 8))
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
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
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 */
if(error)
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
if(fwrite(bphysics, sizeof(char), 8, pf->fp) != 8)
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
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
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
{
if(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
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
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 */
Bug fix: particles dying in collisions could lead to crash in some cases * Forgot to check for index range in own recent pointcache commit.
2011-01-10 02:20:30 +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
while(low <= high) {
mid= (low + high)/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
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;
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
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
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
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
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
for(i=0; i<BPHYS_TOT_DATA; i++) {
if(pm->cur[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->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
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
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
for(i=0; i<BPHYS_TOT_DATA; i++)
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
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
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
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
for(i=0; i<BPHYS_TOT_DATA; i++) {
if(data[i])
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;
for(i=0; i<BPHYS_TOT_DATA; i++) {
NULL check for pointcache. not sure this should be fixed elsewhere, commented this needs further checking.
2010-04-29 17:57:17 +00:00
/* 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 */
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
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
if(extra) {
for(; extra; extra=extra->next) {
if(extra->data)
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
{
if(pid->type==PTCACHE_TYPE_SOFTBODY)
return 6 * sizeof(float);
else if(pid->type==PTCACHE_TYPE_PARTICLES)
return sizeof(ParticleKey);
else if(pid->type==PTCACHE_TYPE_CLOTH)
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
{
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
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
int cfra1=frame-1, cfra2=frame+1;
while(cfra1 >= pid->cache->startframe && !BKE_ptcache_id_exist(pid, cfra1))
cfra1--;
if(cfra1 < pid->cache->startframe)
cfra1 = 0;
while(cfra2 <= pid->cache->endframe && !BKE_ptcache_id_exist(pid, cfra2))
cfra2++;
if(cfra2 > pid->cache->endframe)
cfra2 = 0;
if(cfra1 && !cfra2) {
*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
}
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 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;
while(pm->next && pm->next->frame < frame)
pm= pm->next;
if(pm2 && pm2->frame < frame)
pm2 = NULL;
else {
while(pm2->prev && pm2->prev->frame > frame)
pm2= pm2->prev;
}
if(pm && !pm2) {
*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
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(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
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(!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
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
if(!error && (pf->type != pid->type || !pid->read_header(pf)))
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
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
if(!error) {
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
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
if(pf->flag & PTCACHE_TYPEFLAG_COMPRESS) {
for(i=0; i<BPHYS_TOT_DATA; i++) {
unsigned int out_len = pm->totpoint*ptcache_data_size[i];
if(pf->data_types & (1<<i))
ptcache_file_compressed_read(pf, (unsigned char*)(pm->data[i]), out_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
}
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
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
for(i=0; i<pm->totpoint; i++) {
if(!ptcache_file_data_read(pf)) {
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
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
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
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
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
if(pf->flag & PTCACHE_TYPEFLAG_COMPRESS)
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_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
}
if(error && pm) {
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);
if (error && G.f & G_DEBUG)
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);
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
if(pf==NULL) {
if (G.f & G_DEBUG)
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;
if(pm->extradata.first)
pf->flag |= PTCACHE_TYPEFLAG_EXTRADATA;
if(pid->cache->compression)
pf->flag |= PTCACHE_TYPEFLAG_COMPRESS;
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
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
if(!error) {
if(pid->cache->compression) {
for(i=0; i<BPHYS_TOT_DATA; i++) {
if(pm->data[i]) {
unsigned int in_len = pm->totpoint*ptcache_data_size[i];
unsigned char *out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len)*4, "pointcache_lzo_buffer");
ptcache_file_compressed_write(pf, (unsigned char*)(pm->data[i]), in_len, out, pid->cache->compression);
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
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
for(i=0; i<pm->totpoint; i++) {
ptcache_data_copy(pm->cur, pf->cur);
if(!ptcache_file_data_write(pf)) {
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
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
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
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
for(; extra; extra=extra->next) {
if(extra->data == NULL || extra->totdata == 0)
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
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];
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
unsigned char *out = (unsigned char *)MEM_callocN(LZO_OUT_LEN(in_len)*4, "pointcache_lzo_buffer");
ptcache_file_compressed_write(pf, (unsigned char*)(extra->data), in_len, out, pid->cache->compression);
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);
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
if (error && G.f & 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
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
if(pid->read_stream == NULL)
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
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
if(pf == NULL) {
if (G.f & G_DEBUG)
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
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
if(!ptcache_file_header_begin_read(pf))
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
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
if(!error && (pf->type != pid->type || !pid->read_header(pf)))
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
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
if(!error && pf->totpoint != pid->totpoint(pid->calldata, cfra))
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
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
if(!error) {
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
pid->read_stream(pf, pid->calldata);
}
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
}
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 */
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
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;
while(pm && pm->frame != cfra)
pm = pm->next;
}
/* read the cache */
if(pm) {
int totpoint = pm->totpoint;
if((pid->data_types & (1<<BPHYS_DATA_INDEX)) == 0)
totpoint = MIN2(totpoint, pid->totpoint(pid->calldata, cfra));
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
for(i=0; i<totpoint; i++) {
if(pm->data_types & (1<<BPHYS_DATA_INDEX))
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
if(pid->read_extra_data && pm->extradata.first)
pid->read_extra_data(pid->calldata, pm, (float)pm->frame);
/* clean up temporary memory cache */
if(pid->cache->flag & PTCACHE_DISK_CACHE) {
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 */
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
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;
while(pm && pm->frame != cfra2)
pm = pm->next;
}
/* read the cache */
if(pm) {
int totpoint = pm->totpoint;
if((pid->data_types & (1<<BPHYS_DATA_INDEX)) == 0)
totpoint = MIN2(totpoint, pid->totpoint(pid->calldata, (int)cfra));
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
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
if(pm->data_types & (1<<BPHYS_DATA_INDEX))
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
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(pid->interpolate_extra_data && pm->extradata.first)
pid->interpolate_extra_data(pid->calldata, pm, cfra, (float)cfra1, (float)cfra2);
/* clean up temporary memory cache */
if(pid->cache->flag & PTCACHE_DISK_CACHE) {
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 */
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 BKE_ptcache_read(PTCacheID *pid, float 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
{
int cfrai = (int)cfra, cfra1=0, cfra2=0;
int ret = 0;
/* nothing to read to */
if(pid->totpoint(pid->calldata, cfrai) == 0)
return 0;
if(pid->cache->flag & PTCACHE_READ_INFO) {
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 */
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(cfra == (float)cfrai && BKE_ptcache_id_exist(pid, cfrai))
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 */
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(cfra1 == 0)
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
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(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 */
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(cfra1 == 0 && cfra2 && cfra2 <= pid->cache->simframe)
return 0;
if(cfra1 && cfra1 == cfra2)
return 0;
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +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
if(cfra1) {
if(pid->read_stream)
ptcache_read_stream(pid, cfra1);
else if(pid->read_point)
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
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(cfra2) {
if(pid->read_stream)
ptcache_read_stream(pid, cfra2);
else if(pid->read_point) {
if(cfra1 && cfra2 && pid->interpolate_point)
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
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(cfra1)
ret = (cfra2 ? PTCACHE_READ_INTERPOLATED : PTCACHE_READ_EXACT);
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
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
if((pid->cache->flag & PTCACHE_QUICK_CACHE)==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
cfrai = (int)cfra;
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
/* clear invalid cache frames so that better stuff can be simulated */
if(pid->cache->flag & PTCACHE_OUTDATED) {
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_AFTER, cfrai);
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 if(pid->cache->flag & PTCACHE_FRAMES_SKIPPED) {
if(cfra <= pid->cache->last_exact)
pid->cache->flag &= ~PTCACHE_FRAMES_SKIPPED;
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_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
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
if(pf==NULL) {
if (G.f & G_DEBUG)
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
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
if(!error && (!ptcache_file_header_begin_write(pf) || !pid->write_header(pf)))
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
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
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);
if (error && G.f & 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
}
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);
if(overwrite) {
if(cache->flag & PTCACHE_DISK_CACHE) {
int fra = cfra-1;
while(fra >= cache->startframe && !BKE_ptcache_id_exist(pid, fra))
fra--;
pm2 = ptcache_disk_frame_to_mem(pid, fra);
}
else
pm2 = cache->mem_cache.last;
}
if(pid->write_point) {
for(i=0; i<totpoint; i++) {
int write = pid->write_point(i, pid->calldata, pm->cur, cfra);
if(write) {
BKE_ptcache_mem_pointers_incr(pm);
/* newly born particles have to be copied to previous cached frame */
if(overwrite && write == 2 && pm2 && BKE_ptcache_mem_pointers_seek(i, pm2))
pid->write_point(i, pid->calldata, pm2->cur, cfra);
}
}
}
if(pid->write_extra_data)
pid->write_extra_data(pid->calldata, pm, cfra);
pm->frame = cfra;
if(cache->flag & PTCACHE_DISK_CACHE) {
error += !ptcache_mem_frame_to_disk(pid, pm);
if(pm) {
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
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(pm2) {
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;
/* allways start from scratch on the first frame */
if(cfra && cfra == cache->startframe) {
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, cfra);
cache->flag &= ~PTCACHE_REDO_NEEDED;
return 1;
}
if(pid->cache->flag & PTCACHE_DISK_CACHE) {
if(cfra==0 && cache->startframe > 0)
return 1;
/* find last cached frame */
while(efra > cache->startframe && !BKE_ptcache_id_exist(pid, efra))
efra--;
/* find second last cached frame */
ofra = efra-1;
while(ofra > cache->startframe && !BKE_ptcache_id_exist(pid, ofra))
ofra--;
}
else {
PTCacheMem *pm = cache->mem_cache.last;
/* don't write info file in memory */
if(cfra == 0)
return 0;
if(pm == NULL)
return 1;
efra = pm->frame;
ofra = (pm->prev ? pm->prev->frame : efra - cache->step);
}
if(efra >= cache->startframe && cfra > efra) {
if(ofra >= cache->startframe && efra - ofra < cache->step) {
/* 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
merge of last commit to trunk
2010-05-11 20:06:20 +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;
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
if(ptcache_write_needed(pid, cfra, &overwrite)==0)
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
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(pid->write_stream) {
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
}
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 if(pid->write_point) {
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. */
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
if(cfra - cache->last_exact == 1 || cfra == cache->startframe) {
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) */
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 */
Fix for [#25781] Bake cache for particles fails at end , ALT + A not * Missing check for info frame (frame 0).
2011-01-25 17:08:43 +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!
* 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
* mode - PTCACHE_CLEAR_ALL,
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
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
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;
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 */
if(pid->ob->id.lib)
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 */
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
/* clear all files in the temp dir with the prefix of the ID and the ".bphys" suffix */
switch (mode) {
case PTCACHE_CLEAR_ALL:
case PTCACHE_CLEAR_BEFORE:
case PTCACHE_CLEAR_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
if(pid->cache->flag & PTCACHE_DISK_CACHE) {
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, cfra, 0, 0); /* no path */
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
dir = opendir(path);
if (dir==NULL)
return;
snprintf(ext, sizeof(ext), "_%02d"PTCACHE_EXT, pid->stack_index);
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
if (strncmp(filename, de->d_name, len ) == 0) { /* do we have the right prefix */
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);
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_delete(path_full, 0, 0);
} else {
/* read the number of the file */
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 frame, len2 = (int)strlen(de->d_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
char num[7];
if (len2 > 15) { /* could crash if trying to copy a string out of this range*/
BLI_strncpy(num, de->d_name + (strlen(de->d_name) - 15), sizeof(num));
frame = atoi(num);
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
if((mode==PTCACHE_CLEAR_BEFORE && frame < cfra) ||
(mode==PTCACHE_CLEAR_AFTER && frame > cfra) ) {
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);
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_delete(path_full, 0, 0);
Fix for [#25282] Crash when weting initial mesh pressets with a smoke simulation
2010-12-18 09:46:52 +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
if(mode == PTCACHE_CLEAR_ALL && pid->cache->cached_frames)
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;
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);
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
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
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
if(pid->cache->cached_frames)
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 {
while(pm) {
if((mode==PTCACHE_CLEAR_BEFORE && pm->frame < cfra) ||
(mode==PTCACHE_CLEAR_AFTER && pm->frame > cfra) ) {
link = pm;
Fix for [#24501] Apeend object with cloth sim from another file crash.
2010-11-02 15:21:43 +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:
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(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 */
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_delete(filename, 0, 0);
}
}
else {
PTCacheMem *pm = pid->cache->mem_cache.first;
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;
}
}
}
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
if(pid->cache->cached_frames && cfra>=sta && cfra<=end)
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
{
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(!pid->cache)
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
if(cfra<pid->cache->startframe || cfra > pid->cache->endframe)
return 0;
if(pid->cache->cached_frames && pid->cache->cached_frames[cfra-pid->cache->startframe]==0)
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
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(pid->cache->flag & PTCACHE_DISK_CACHE) {
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;
for(; pm; pm=pm->next) {
if(pm->frame==cfra)
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
{
Object *ob;
PointCache *cache;
float offset, time, nexttime;
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: this has to be sorter out once bsystem_time gets redone, */
/* 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
* varies over time is not simpe to support.
* - field and motion blur offsets are currently ignored, proper solution
* is probably to interpolate results from two frames for that ..
*/
ob= pid->ob;
cache= pid->cache;
if(timescale) {
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
time= bsystem_time(scene, ob, cfra, 0.0f);
nexttime= bsystem_time(scene, ob, cfra+1.0f, 0.0f);
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);
}
if(startframe && endframe) {
*startframe= cache->startframe;
*endframe= cache->endframe;
2.5: Blender "Animato" - New Animation System Finally, here is the basic (functional) prototype of the new animation system which will allow for the infamous "everything is animatable", and which also addresses several of the more serious shortcomings of the old system. Unfortunately, this will break old animation files (especially right now, as I haven't written the version patching code yet), however, this is for the future. Highlights of the new system: * Scrapped IPO-Curves/IPO/(Action+Constraint-Channels)/Action system, and replaced it with F-Curve/Action. - F-Curves (animators from other packages will feel at home with this name) replace IPO-Curves. - The 'new' Actions, act as the containers for F-Curves, so that they can be reused. They are therefore more akin to the old 'IPO' blocks, except they do not have the blocktype restriction, so you can store materials/texture/geometry F-Curves in the same Action as Object transforms, etc. * F-Curves use RNA-paths for Data Access, hence allowing "every" (where sensible/editable that is) user-accessible setting from RNA to be animated. * Drivers are no longer mixed with Animation Data, so rigs will not be that easily broken and several dependency problems can be eliminated. (NOTE: drivers haven't been hooked up yet, but the code is in place) * F-Curve modifier system allows useful 'large-scale' manipulation of F-Curve values, including (I've only included implemented ones here): envelope deform (similar to lattices to allow broad-scale reshaping of curves), curve generator (polynomial or py-expression), cycles (replacing the old cyclic extrapolation modes, giving more control over this). (NOTE: currently this cannot be tested, as there's not access to them, but the code is all in place) * NLA system with 'tracks' (i.e. layers), and multiple strips per track. (NOTE: NLA system is not yet functional, as it's only partially coded still) There are more nice things that I will be preparing some nice docs for soon, but for now, check for more details: http://lists.blender.org/pipermail/bf-taskforce25/2009-January/000260.html So, what currently works: * I've implemented two basic operators for the 3D-view only to Insert and Delete Keyframes. These are tempolary ones only that will be replaced in due course with 'proper' code. * Object Loc/Rot/Scale can be keyframed. Also, the colour of the 'active' material (Note: this should really be for nth material instead, but that doesn't work yet in RNA) can also be keyframed into the same datablock. * Standard animation refresh (i.e. animation resulting from NLA and Action evaluation) is now done completely separate from drivers before anything else is done after a frame change. Drivers are handled after this in a separate pass, as dictated by depsgraph flags, etc. Notes: * Drivers haven't been hooked up yet * Only objects and data directly linked to objects can be animated. * Depsgraph will need further tweaks. Currently, I've only made sure that it will update some things in the most basic cases (i.e. frame change). * Animation Editors are currently broken (in terms of editing stuff). This will be my next target (priority to get Dopesheet working first, then F-Curve editor - i.e. old IPO Editor) * I've had to put in large chunks of XXX sandboxing for old animation system code all around the place. This will be cleaned up in due course, as some places need special review. In particular, the particles and sequencer code have far too many manual calls to calculate + flush animation info, which is really bad (this is a 'please explain yourselves' call to Physics coders!).
2009-01-17 03:12:50 +00:00
// XXX ipoflag is depreceated - old animation system stuff
if (/*(ob->ipoflag & OB_OFFS_PARENT) &&*/ (ob->partype & PARSLOW)==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
offset= give_timeoffset(ob);
*startframe += (int)(offset+0.5f);
*endframe += (int)(offset+0.5f);
}
}
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 */
if(cache->cached_frames) {
if(MEM_allocN_len(cache->cached_frames) != sizeof(char) * (cache->endframe-cache->startframe+1)) {
MEM_freeN(cache->cached_frames);
cache->cached_frames = NULL;
}
}
Fix for own recent commit 32147 * Pointcache wasn't checked to be valid before allocating the cached frames array
2010-09-28 09:27:35 +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");
if(pid->cache->flag & PTCACHE_DISK_CACHE) {
/* 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;
snprintf(ext, sizeof(ext), "_%02d"PTCACHE_EXT, pid->stack_index);
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
if (strncmp(filename, de->d_name, len ) == 0) { /* do we have the right prefix */
/* read the number of the file */
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 frame, len2 = (int)strlen(de->d_name);
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
char num[7];
if (len2 > 15) { /* could crash if trying to copy a string out of this range*/
BLI_strncpy(num, de->d_name + (strlen(de->d_name) - 15), sizeof(num));
frame = atoi(num);
if(frame >= sta && frame <= end)
cache->cached_frames[frame-sta] = 1;
}
}
}
}
closedir(dir);
}
else {
PTCacheMem *pm= pid->cache->mem_cache.first;
while(pm) {
if(pm->frame >= sta && pm->frame <= end)
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
if(!pid->cache)
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
if(mode == PTCACHE_RESET_DEPSGRAPH) {
if(!(cache->flag & PTCACHE_BAKED) && !BKE_ptcache_get_continue_physics()) {
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
if(cache->flag & PTCACHE_QUICK_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
clear= 1;
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
}
else if(mode == PTCACHE_RESET_BAKED) {
if(!BKE_ptcache_get_continue_physics()) {
reset= 1;
clear= 1;
}
else
cache->flag |= PTCACHE_OUTDATED;
}
else if(mode == PTCACHE_RESET_OUTDATED) {
reset = 1;
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
if(cache->flag & PTCACHE_OUTDATED && !(cache->flag & PTCACHE_BAKED)) {
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
}
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
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);
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
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);
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
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);
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
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);
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
else if(pid->type == PTCACHE_TYPE_SMOKE_HIGHRES)
Smoke: * Bugfix for non-working high res display * Bugfix for smoke disappearing on 2nd frame
2009-09-09 23:52:48 +00:00
smokeModifier_reset_turbulence(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
}
if(clear)
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 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
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
if(ob->soft) {
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
}
for(psys=ob->particlesystem.first; psys; psys=psys->next) {
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
/* children or just redo can be calculated without reseting anything */
if(psys->recalc & PSYS_RECALC_REDO || psys->recalc & PSYS_RECALC_CHILD)
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 */
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
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);
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
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;
}
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
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
}
}
for(md=ob->modifiers.first; md; md=md->next) {
if(md->type == eModifierType_Cloth) {
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
}
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
if(md->type == eModifierType_Smoke) {
SmokeModifierData *smd = (SmokeModifierData *)md;
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);
}
}
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
}
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
Fix for bug #8930: issue removing temporary pointcache directory on windows.
2008-04-12 12:56:49 +00:00
if (BLI_exist(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) {
if( strcmp(de->d_name, ".")==0 || strcmp(de->d_name, "..")==0) {
/* do nothing */
} 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);
* 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
BLI_delete(path_full, 0, 0);
} else {
rmdir = 0; /* unknown file, dont remove the dir */
}
}
Fix for bug #8930: issue removing temporary pointcache directory on windows.
2008-04-12 12:56:49 +00:00
closedir(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
} else {
rmdir = 0; /* path dosnt exist */
}
if (rmdir) {
BLI_delete(path, 1, 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
/* Continuous Interaction */
static int CONTINUE_PHYSICS = 0;
2.5: more removal of G.main.
2010-08-13 14:23:44 +00:00
void BKE_ptcache_set_continue_physics(Main *bmain, Scene *scene, int enable)
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
{
Object *ob;
if(CONTINUE_PHYSICS != enable) {
CONTINUE_PHYSICS = enable;
if(CONTINUE_PHYSICS == 0) {
2.5: more removal of G.main.
2010-08-13 14:23:44 +00:00
for(ob=bmain->object.first; ob; ob=ob->id.next)
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(BKE_ptcache_object_reset(scene, ob, PTCACHE_RESET_OUTDATED))
Dependency graph: changed DAG_id_flush_update to DAG_id_tag_update. Now it only tags the ID and does the actual flush/update delayed, before the next redraw. For objects the update was already delayed, just flushing wasn't yet. This should help performance in python and animation editors, by making calls to RNA property update quicker. Still need to add calls in a few places where this was previously avoided due to bad performance.
2010-12-05 18:59:23 +00:00
DAG_id_tag_update(&ob->id, OB_RECALC_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
}
}
}
warning cleanup. - fix mistake with grease pencil UI (&& was intended but & used). - use (void) rather then () across _all_ blenders code. - a few minor edits, don't shadow stack variables in roll calculation & avoid running memset() for VBO vertex map.
2011-02-13 15:02:21 +00:00
int BKE_ptcache_get_continue_physics(void)
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 CONTINUE_PHYSICS;
}
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;
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->step= 10;
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
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(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
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);
if(cache->edit && cache->free_edit)
cache->free_edit(cache->edit);
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
if(cache->cached_frames)
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)
{
PointCache *cache = ptcaches->first;
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
while(cache) {
BLI_remlink(ptcaches, cache);
BKE_ptcache_free(cache);
cache = ptcaches->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
}
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
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
static PointCache *ptcache_copy(PointCache *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
{
PointCache *ncache;
ncache= MEM_dupallocN(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
/* hmm, should these be copied over instead? */
ncache->mem_cache.first = NULL;
ncache->mem_cache.last = NULL;
Fix for [#25079] Duplicating object with particles system on it make Blender crashs * pointcache->cached_frames wasn't set to NULL when copying pointcaches * also set pointcache->edit to null just in case
2010-12-07 22:17:58 +00:00
ncache->cached_frames = NULL;
ncache->edit = NULL;
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
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
ncache->flag= 0;
ncache->simframe= 0;
return ncache;
}
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
/* returns first point cache */
PointCache *BKE_ptcache_copy_list(ListBase *ptcaches_new, ListBase *ptcaches_old)
{
PointCache *cache = ptcaches_old->first;
ptcaches_new->first = ptcaches_new->last = 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
Proper copy code for multiple point caches.
2009-08-14 16:25:59 +00:00
for(; cache; cache=cache->next)
BLI_addtail(ptcaches_new, ptcache_copy(cache));
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
/* 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;
baker.bake=0;
baker.break_data=NULL;
baker.break_test=NULL;
baker.pid=NULL;
baker.progressbar=NULL;
* 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
baker.progressend=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
baker.progresscontext=NULL;
baker.render=0;
Rendering animations with particle trails cached.
2009-07-04 12:09:21 +00:00
baker.anim_init = 0;
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
baker.main=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
baker.scene=scene;
Testing for the need to quick cache was causing slowdowns on files with many duplis. * The test is now only done when some object that uses cache has actually changed. * The added scene->physics_settings->quick_cache_step is only an internal counter, not a user changeable value.
2010-04-13 20:06:55 +00:00
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
}
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
/* Simulation thread, no need for interlocks as data written in both threads
are only unitary integers (I/O assumed to be atomic for them) */
typedef struct {
int break_operation;
int thread_ended;
int endframe;
int step;
int *cfra_ptr;
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
Main *main;
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
Scene *scene;
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_bake_data;
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
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)
{
if(dtime > 60.0) {
if(dtime > 3600.0)
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);
}
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_bake_thread(void *ptr) {
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
int usetimer = 0, sfra, efra;
double stime, ptime, ctime, fetd;
char run[32], cur[32], etd[32];
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_bake_data *data = (ptcache_bake_data*)ptr;
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
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
stime = ptime = PIL_check_seconds_timer();
sfra = *data->cfra_ptr;
efra = data->endframe;
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
for(; (*data->cfra_ptr <= data->endframe) && !data->break_operation; *data->cfra_ptr+=data->step) {
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
scene_update_for_newframe(data->main, data->scene, data->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
if(G.background) {
printf("bake: frame %d :: %d\n", (int)*data->cfra_ptr, data->endframe);
}
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
else {
ctime = PIL_check_seconds_timer();
fetd = (ctime-ptime)*(efra-*data->cfra_ptr)/data->step;
if(usetimer || fetd > 60.0) {
usetimer = 1;
ptcache_dt_to_str(cur, ctime-ptime);
ptcache_dt_to_str(run, ctime-stime);
ptcache_dt_to_str(etd, fetd);
quiet gcc float -> double promotion warnings.
2011-04-02 02:08:33 +00:00
printf("Baked for %s, current frame: %i/%i (%.3fs), ETC: %s \r", run, *data->cfra_ptr-sfra+1, efra-sfra+1, ctime-ptime, etd);
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
}
ptime = ctime;
}
}
if(usetimer) {
ptcache_dt_to_str(run, PIL_check_seconds_timer()-stime);
printf("Bake %s %s (%i frames simulated). \n", (data->break_operation ? "canceled after" : "finished in"), run, *data->cfra_ptr-sfra);
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
}
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
data->thread_ended = TRUE;
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
/* if bake is not given run simulations to current frame */
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_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
{
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
Main *bmain = baker->main;
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;
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;
int startframe = MAXFRAME;
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
ListBase threads;
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_bake_data thread_data;
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
int progress, old_progress;
thread_data.endframe = baker->anim_init ? scene->r.sfra : CFRA;
thread_data.step = baker->quick_step;
thread_data.cfra_ptr = &CFRA;
thread_data.scene = baker->scene;
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
thread_data.main = baker->main;
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
G.afbreek = 0;
/* set caches to baking mode and figure out start frame */
if(pid) {
/* cache/bake a single object */
cache = pid->cache;
if((cache->flag & PTCACHE_BAKED)==0) {
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
if(pid->type==PTCACHE_TYPE_PARTICLES) {
ParticleSystem *psys= pid->calldata;
/* a bit confusing, could make this work better in the UI */
if(psys->part->type == PART_EMITTER)
psys_get_pointcache_start_end(scene, pid->calldata, &cache->startframe, &cache->endframe);
}
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
else if(pid->type == PTCACHE_TYPE_SMOKE_HIGHRES) {
/* get all pids from the object and search for smoke low res */
ListBase pidlist2;
PTCacheID *pid2;
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
BKE_ptcache_ids_from_object(&pidlist2, pid->ob, scene, MAX_DUPLI_RECUR);
Smoke: * This should make low res bake too if high res gets baked
2009-09-15 16:05:59 +00:00
for(pid2=pidlist2.first; pid2; pid2=pid2->next) {
if(pid2->type == PTCACHE_TYPE_SMOKE_DOMAIN)
{
if(pid2->cache && !(pid2->cache->flag & PTCACHE_BAKED)) {
if(bake || pid2->cache->flag & PTCACHE_REDO_NEEDED)
BKE_ptcache_id_clear(pid2, PTCACHE_CLEAR_ALL, 0);
if(bake) {
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
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
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);
if(bake) {
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
thread_data.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 {
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
thread_data.endframe = MIN2(thread_data.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;
}
}
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
else for(SETLOOPER(scene, sce_iter, base)) {
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 everything in the scene */
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
BKE_ptcache_ids_from_object(&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
for(pid=pidlist.first; pid; pid=pid->next) {
cache = pid->cache;
if((cache->flag & PTCACHE_BAKED)==0) {
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +00:00
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
ParticleSystem *psys = (ParticleSystem*)pid->calldata;
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
/* skip hair & keyed particles */
if(psys->part->type == PART_HAIR || psys->part->phystype == PART_PHYS_KEYED)
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +00:00
continue;
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_get_pointcache_start_end(scene, pid->calldata, &cache->startframe, &cache->endframe);
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +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 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
if((cache->flag & PTCACHE_REDO_NEEDED || (cache->flag & PTCACHE_SIMULATION_VALID)==0)
&& ((cache->flag & PTCACHE_QUICK_CACHE)==0 || render || 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
BKE_ptcache_id_clear(pid, PTCACHE_CLEAR_ALL, 0);
startframe = MIN2(startframe, cache->startframe);
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
if(bake || render) {
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
if(bake)
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
thread_data.endframe = MAX2(thread_data.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_BAKED;
}
}
BLI_freelistN(&pidlist);
}
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;
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
thread_data.break_operation = FALSE;
thread_data.thread_ended = FALSE;
old_progress = -1;
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
WM_cursor_wait(1);
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
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
if(G.background) {
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_bake_thread((void*)&thread_data);
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
}
else {
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
BLI_init_threads(&threads, ptcache_bake_thread, 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
BLI_insert_thread(&threads, (void*)&thread_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
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
while (thread_data.thread_ended == 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
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
if(bake)
progress = (int)(100.0f * (float)(CFRA - startframe)/(float)(thread_data.endframe-startframe));
else
progress = CFRA;
/* NOTE: baking should not redraw whole ui as this slows things down */
if ((baker->progressbar) && (progress != old_progress)) {
baker->progressbar(baker->progresscontext, progress);
old_progress = progress;
}
/* Delay to lessen CPU load from UI thread */
PIL_sleep_ms(200);
/* NOTE: breaking baking should leave calculated frames in cache, not clear it */
if(blender_test_break() && !thread_data.break_operation) {
thread_data.break_operation = TRUE;
if (baker->progressend)
baker->progressend(baker->progresscontext);
WM_cursor_wait(1);
}
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
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
BLI_end_threads(&threads);
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
}
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 */
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;
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(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 */
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
}
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
else for(SETLOOPER(scene, sce_iter, base)) {
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
BKE_ptcache_ids_from_object(&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
for(pid=pidlist.first; pid; pid=pid->next) {
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +00:00
/* skip hair 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
if(pid->type==PTCACHE_TYPE_PARTICLES && ((ParticleSystem*)pid->calldata)->part->type == PART_HAIR)
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +00:00
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
cache = pid->cache;
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
if(thread_data.step > 1)
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_OUTDATED);
else
cache->flag &= ~(PTCACHE_BAKING|PTCACHE_REDO_NEEDED);
cache->flag |= PTCACHE_SIMULATION_VALID;
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(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
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
}
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
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
A bunch of fun stuff now possible because of new pointcache code: * Baked normal particles can now use the "Path" visualization. * Path "max length" & "abs length" are now history: - New option to set path start & end times + random variation to length. - Much more flexible (and calculated better) than previous options. - This works with parents, children, hair & normal particles unlike old length option. - Only known issue for now is that children from faces don't get calculated correctly when using path start time. * New option "trails" for "halo", "line" and "billboard" visualizations: - Draws user controllable number of particle instances along particles path backwards from current position. - Works with children too for cool/weird visualizations that weren't possible before. * Normal particle children's velocities are now approximated better when needed so that "line" visualization trails will look nice. * New particle instance modifier options: - "path"-option works better and has controllable (max)position along path (with random variation possible). - "keep shape"-option for hair, keyed, or baked particles allows to place the instances to a single point (with random variation possible) along particle path. - "axis" option to make rotation handling better (still not perfect, but will have to do for now). Some fixes & cleanup done along the way: * Random path length didn't work for non-child particles. * Cached & unborn particles weren't reset to emit locations. * Particle numbers weren't drawn in the correct place. * Setting proper render & draw visualizations was lost somewhere when initializing new particle settings. * Changing child mode wasn't working correctly. * Some cleanup & modularization of particle child effector code and particle drawing & rendering code. * Object & group visualizations didn't work. * Child simplification didn't work.
2009-07-04 03:50:12 +00:00
if(bake) /* already on cfra unless 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
scene_update_for_newframe(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
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
if (thread_data.break_operation)
WM_cursor_wait(0);
else if (baker->progressend)
baker->progressend(baker->progresscontext);
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
WM_cursor_wait(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
/* 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
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
for(cfra=sfra; cfra <= efra; cfra++) {
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
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(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
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
for(; pm; pm=pm->next) {
if(ptcache_mem_frame_to_disk(pid, pm)==0) {
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 */
if(cache->flag & PTCACHE_BAKED)
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
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 (!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;
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
if (G.f & G_DEBUG)
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
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
if(cache->cached_frames) {
MEM_freeN(cache->cached_frames);
cache->cached_frames=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
if(cache->flag & PTCACHE_DISK_CACHE)
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);
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
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
void BKE_ptcache_disk_cache_rename(PTCacheID *pid, char *from, char *to)
{
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(old_name, pid->cache->name);
/* get "from" filename */
strcpy(pid->cache->name, from);
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);
if(dir==NULL) {
strcpy(pid->cache->name, old_name);
return;
}
snprintf(ext, sizeof(ext), "_%02d"PTCACHE_EXT, pid->stack_index);
/* put new name into cache */
strcpy(pid->cache->name, to);
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
if (strncmp(old_filename, de->d_name, len ) == 0) { /* do we have the right prefix */
/* read the number of the file */
int frame, len2 = (int)strlen(de->d_name);
char num[7];
if (len2 > 15) { /* could crash if trying to copy a string out of this range*/
BLI_strncpy(num, de->d_name + (strlen(de->d_name) - 15), sizeof(num));
frame = atoi(num);
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
strcpy(pid->cache->name, old_name);
}
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];
if(!cache)
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;
if(cache->index >= 0)
snprintf(ext, sizeof(ext), "_%02d"PTCACHE_EXT, cache->index);
else
strcpy(ext, PTCACHE_EXT);
while ((de = readdir(dir)) != NULL) {
if (strstr(de->d_name, ext)) { /* do we have the right extension?*/
if (strncmp(filename, de->d_name, len ) == 0) { /* do we have the right prefix */
/* read the number of the file */
Pointcache: * prepare pointcache for smoke (smoke doesn't use it yet, commit follows later)
2009-08-14 17:39:27 +00:00
int frame, len2 = (int)strlen(de->d_name);
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
char num[7];
if (len2 > 15) { /* could crash if trying to copy a string out of this range*/
BLI_strncpy(num, de->d_name + (strlen(de->d_name) - 15), sizeof(num));
frame = atoi(num);
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(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);
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
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;
Bug fix for problem 2 in [#25973] Bake End Frame Not Configurable * External caches didn't load for smoke straight away. Smoke caches store all necessary info in every file, so no need to try to look for an info file.
2011-02-11 14:59:19 +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) */
Bug fix for problem 2 in [#25973] Bake End Frame Not Configurable * External caches didn't load for smoke straight away. Smoke caches store all necessary info in every file, so no need to try to look for an info file.
2011-02-11 14:59:19 +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);
if(pf) {
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
if(ptcache_file_header_begin_read(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
if(pf->type == pid->type && pid->read_header(pf)) {
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);
if(pf) {
while(ptcache_file_read(pf, old_data, 1, elemsize))
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
}
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];
if(cache->flag & PTCACHE_EXTERNAL) {
int cfra = cache->startframe;
for(; cfra<=cache->endframe; cfra++) {
if(BKE_ptcache_id_exist(pid, cfra))
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 */
if(pid->type == PTCACHE_TYPE_SMOKE_DOMAIN && totframes)
sprintf(cache->info, "%i frames found!", totframes);
else if(totframes && 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
sprintf(cache->info, "%i points found!", cache->totpoint);
else
sprintf(cache->info, "No valid data to read!");
return;
}
if(cache->flag & PTCACHE_DISK_CACHE) {
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
if(pid->type == PTCACHE_TYPE_SMOKE_DOMAIN)
{
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
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
if(cache->totpoint > totpoint)
sprintf(mem_info, "%i cells + High Resolution cached", totpoint);
else
fix for some mistakes in recent commit.
2010-12-01 02:40:13 +00:00
sprintf(mem_info, "%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
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
for(; cfra<=cache->endframe; cfra++) {
if(BKE_ptcache_id_exist(pid, cfra))
totframes++;
}
sprintf(mem_info, "%i frames on disk", totframes);
}
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 {
PTCacheMem *pm = cache->mem_cache.first;
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
float bytes = 0.0f;
int i, mb;
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
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]);
for(extra=pm->extradata.first; extra; extra=extra->next) {
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
mb = (bytes > 1024.0f * 1024.0f);
sprintf(mem_info, "%i frames in memory (%.1f %s)",
totframes,
bytes / (mb ? 1024.0f * 1024.0f : 1024.0f),
mb ? "Mb" : "kb");
}
if(cache->flag & PTCACHE_OUTDATED) {
sprintf(cache->info, "%s, cache is outdated!", mem_info);
}
else if(cache->flag & PTCACHE_FRAMES_SKIPPED) {
sprintf(cache->info, "%s, not exact since frame %i.", mem_info, cache->last_exact);
}
else
sprintf(cache->info, "%s.", mem_info);
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)
{
Fix for [#24237] Hair dynamics with zero particles generates a segmentation fault
2010-10-14 09:01:03 +00:00
if(cache) {
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)
{
Fix for [#24237] Hair dynamics with zero particles generates a segmentation fault
2010-10-14 09:01:03 +00:00
if(cache) {
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