This repository has been archived on 2023-10-09.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

1894 lines
51 KiB
C
Raw Normal View History

/* effect.c
2002-10-12 11:37:38 +00:00
*
*
* $Id$
*
* ***** BEGIN GPL/BL DUAL LICENSE BLOCK *****
*
* 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
* of the License, or (at your option) any later version. The Blender
* Foundation also sells licenses for use in proprietary software under
* the Blender License. See http://www.blender.org/BL/ for information
* about this.
*
* 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,
* Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
* The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
* All rights reserved.
*
* The Original Code is: all of this file.
*
* Contributor(s): none yet.
*
* ***** END GPL/BL DUAL LICENSE BLOCK *****
*/
#include <math.h>
#include <stdlib.h>
#include "MEM_guardedalloc.h"
#include "DNA_listBase.h"
#include "DNA_effect_types.h"
#include "DNA_object_types.h"
#include "DNA_object_force.h"
2002-10-12 11:37:38 +00:00
#include "DNA_mesh_types.h"
#include "DNA_meshdata_types.h"
2002-10-12 11:37:38 +00:00
#include "DNA_material_types.h"
#include "DNA_curve_types.h"
#include "DNA_key_types.h"
#include "DNA_texture_types.h"
#include "DNA_scene_types.h"
#include "DNA_lattice_types.h"
#include "DNA_ipo_types.h"
2002-10-12 11:37:38 +00:00
#include "BLI_blenlib.h"
#include "BLI_arithb.h"
#include "BLI_rand.h"
#include "BKE_action.h"
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
#include "BKE_anim.h" /* needed for where_on_path */
#include "BKE_armature.h"
2002-10-12 11:37:38 +00:00
#include "BKE_bad_level_calls.h"
#include "BKE_blender.h"
#include "BKE_constraint.h"
#include "BKE_deform.h"
2002-10-12 11:37:38 +00:00
#include "BKE_displist.h"
#include "BKE_DerivedMesh.h"
#include "BKE_effect.h"
#include "BKE_global.h"
#include "BKE_ipo.h"
#include "BKE_key.h"
2002-10-12 11:37:38 +00:00
#include "BKE_lattice.h"
#include "BKE_mesh.h"
#include "BKE_material.h"
#include "BKE_main.h"
#include "BKE_object.h"
#include "BKE_screen.h"
#include "BKE_utildefines.h"
Phew, a lot of work, and no new features... Main target was to make the inner rendering loop using no globals anymore. This is essential for proper usage while raytracing, it caused a lot of hacks in the raycode as well, which even didn't work correctly for all situations (textures especially). Done this by creating a new local struct RenderInput, which replaces usage of the global struct Render R. The latter now only is used to denote image size, viewmatrix, and the like. Making the inner render loops using no globals caused 1000s of vars to be changed... but the result definitely is much nicer code, which enables making 'real' shaders in a next stage. It also enabled me to remove the hacks from ray.c Then i went to the task of removing redundant code. Especially the calculus of texture coords took place (identical) in three locations. Most obvious is the change in the unified render part, which is much less code now; it uses the same rendering routines as normal render now. (Note; not for halos yet!) I also removed 6 files called 'shadowbuffer' something. This was experimen- tal stuff from NaN days. And again saved a lot of double used code. Finally I went over the blenkernel and blender/src calls to render stuff. Here the same local data is used now, resulting in less dependency. I also moved render-texture to the render module, this was still in Kernel. (new file: texture.c) So! After this commit I will check on the autofiles, to try to fix that. MSVC people have to do it themselves. This commit will need quite some testing help, but I'm around!
2003-12-21 21:52:51 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
#include "render.h" // externtex, bad level call (ton)
#include "PIL_time.h"
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
2002-10-12 11:37:38 +00:00
Effect *add_effect(int type)
{
Effect *eff=0;
PartEff *paf;
int a;
switch(type) {
case EFF_PARTICLE:
paf= MEM_callocN(sizeof(PartEff), "neweff");
eff= (Effect *)paf;
paf->sta= 1.0;
paf->end= 100.0;
paf->lifetime= 50.0;
for(a=0; a<PAF_MAXMULT; a++) {
paf->life[a]= 50.0;
paf->child[a]= 4;
paf->mat[a]= 1;
}
paf->totpart= 1000;
paf->totkey= 8;
paf->staticstep= 5;
paf->defvec[2]= 1.0f;
paf->nabla= 0.05f;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
paf->disp = 100;
paf->speedtex = 8;
paf->omat = 1;
2002-10-12 11:37:38 +00:00
break;
}
eff->type= eff->buttype= type;
eff->flag |= SELECT;
return eff;
}
void free_effect(Effect *eff)
{
PartEff *paf;
if(eff->type==EFF_PARTICLE) {
paf= (PartEff *)eff;
if(paf->keys) MEM_freeN(paf->keys);
}
MEM_freeN(eff);
}
void free_effects(ListBase *lb)
{
Effect *eff;
eff= lb->first;
while(eff) {
BLI_remlink(lb, eff);
free_effect(eff);
eff= lb->first;
}
}
Effect *copy_effect(Effect *eff)
{
Effect *effn;
2004-10-14 22:39:13 +00:00
effn= MEM_dupallocN(eff);
2002-10-12 11:37:38 +00:00
if(effn->type==EFF_PARTICLE) ((PartEff *)effn)->keys= 0;
return effn;
}
void copy_act_effect(Object *ob)
{
/* return a copy of the active effect */
2002-10-12 11:37:38 +00:00
Effect *effn, *eff;
eff= ob->effect.first;
while(eff) {
if(eff->flag & SELECT) {
effn= copy_effect(eff);
BLI_addtail(&ob->effect, effn);
eff->flag &= ~SELECT;
return;
}
eff= eff->next;
}
/* when it comes here: add new effect */
eff= add_effect(EFF_PARTICLE);
2002-10-12 11:37:38 +00:00
BLI_addtail(&ob->effect, eff);
}
void copy_effects(ListBase *lbn, ListBase *lb)
{
Effect *eff, *effn;
lbn->first= lbn->last= 0;
eff= lb->first;
while(eff) {
effn= copy_effect(eff);
BLI_addtail(lbn, effn);
eff= eff->next;
}
}
void deselectall_eff(Object *ob)
{
Effect *eff= ob->effect.first;
while(eff) {
eff->flag &= ~SELECT;
eff= eff->next;
}
}
/* ***************** PARTICLES ***************** */
static Particle *new_particle(PartEff *paf)
2002-10-12 11:37:38 +00:00
{
static Particle *pa;
static int cur;
/* we agree: when paf->keys==0: alloc */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(paf->keys==NULL) {
2002-10-12 11:37:38 +00:00
pa= paf->keys= MEM_callocN( paf->totkey*paf->totpart*sizeof(Particle), "particlekeys" );
cur= 0;
}
else {
if(cur && cur<paf->totpart) pa+=paf->totkey;
cur++;
}
return pa;
}
PartEff *give_parteff(Object *ob)
{
PartEff *paf;
paf= ob->effect.first;
while(paf) {
if(paf->type==EFF_PARTICLE) return paf;
paf= paf->next;
}
return 0;
}
void where_is_particle(PartEff *paf, Particle *pa, float ctime, float *vec)
{
Particle *p[4];
float dt, t[4];
int a;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(paf->totkey==1 || ctime < pa->time) {
2002-10-12 11:37:38 +00:00
VECCOPY(vec, pa->co);
return;
}
/* first find the first particlekey */
2002-10-12 11:37:38 +00:00
a= (int)((paf->totkey-1)*(ctime-pa->time)/pa->lifetime);
if(a>=paf->totkey) a= paf->totkey-1;
else if(a<0) a= 0;
2002-10-12 11:37:38 +00:00
pa+= a;
if(a>0) p[0]= pa-1; else p[0]= pa;
p[1]= pa;
if(a+1<paf->totkey) p[2]= pa+1; else p[2]= pa;
if(a+2<paf->totkey) p[3]= pa+2; else p[3]= p[2];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(p[1]==p[2] || p[2]->time == p[1]->time) dt= 0.0;
2002-10-12 11:37:38 +00:00
else dt= (ctime-p[1]->time)/(p[2]->time - p[1]->time);
if(paf->flag & PAF_BSPLINE) set_four_ipo(dt, t, KEY_BSPLINE);
else set_four_ipo(dt, t, KEY_CARDINAL);
vec[0]= t[0]*p[0]->co[0] + t[1]*p[1]->co[0] + t[2]*p[2]->co[0] + t[3]*p[3]->co[0];
vec[1]= t[0]*p[0]->co[1] + t[1]*p[1]->co[1] + t[2]*p[2]->co[1] + t[3]*p[3]->co[1];
vec[2]= t[0]*p[0]->co[2] + t[1]*p[1]->co[2] + t[2]*p[2]->co[2] + t[3]*p[3]->co[2];
}
static void particle_tex(MTex *mtex, PartEff *paf, float *co, float *no)
2002-10-12 11:37:38 +00:00
{
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
float tin, tr, tg, tb, ta;
2002-10-12 11:37:38 +00:00
float old;
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
externtex(mtex, co, &tin, &tr, &tg, &tb, &ta);
2002-10-12 11:37:38 +00:00
if(paf->texmap==PAF_TEXINT) {
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
tin*= paf->texfac;
no[0]+= tin*paf->defvec[0];
no[1]+= tin*paf->defvec[1];
no[2]+= tin*paf->defvec[2];
2002-10-12 11:37:38 +00:00
}
else if(paf->texmap==PAF_TEXRGB) {
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
no[0]+= (tr-0.5f)*paf->texfac;
no[1]+= (tg-0.5f)*paf->texfac;
no[2]+= (tb-0.5f)*paf->texfac;
2002-10-12 11:37:38 +00:00
}
else { /* PAF_TEXGRAD */
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
old= tin;
2002-10-12 11:37:38 +00:00
co[0]+= paf->nabla;
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
externtex(mtex, co, &tin, &tr, &tg, &tb, &ta);
no[0]+= (old-tin)*paf->texfac;
2002-10-12 11:37:38 +00:00
co[0]-= paf->nabla;
co[1]+= paf->nabla;
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
externtex(mtex, co, &tin, &tr, &tg, &tb, &ta);
no[1]+= (old-tin)*paf->texfac;
2002-10-12 11:37:38 +00:00
co[1]-= paf->nabla;
co[2]+= paf->nabla;
Biiig commit! Thanks to 2-3 weeks of cvs freeze... Render: - New; support for dual CPU render (SDL thread) Currently only works with alternating scanlines, but gives excellent performance. For both normal render as unified implemented. Note the "mutex" locks on z-transp buffer render and imbuf loads. - This has been made possible by major cleanups in render code, especially getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct OSA or using Materials or Texture data to write to. - Made normal render fully 4x32 floats too, and removed all old optimizes with chars or shorts. - Made normal render and unified render use same code for sky and halo render, giving equal (and better) results for halo render. Old render now also uses PostProcess options (brightness, mul, gamma) - Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer after render. Using PostProcess menu you will note an immediate re- display of image too (32 bits RGBA) - Added "Hue" and "Saturation" sliders to PostProcess options - Render module is still not having a "nice" API, but amount of dependencies went down a lot. Next todo: remove abusive "previewrender" code. The last main global in Render (struct Render) now can be re-used for fully controlling a render, to allow multiple "instances" of render to open. - Renderwindow now displays a smal bar on top with the stats, and keeps the stats after render too. Including "spare" page support. Not only easier visible that way, but also to remove the awkward code that was drawing stats in the Info header (extreme slow on some ATIs too) - Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping defines. - I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
externtex(mtex, co, &tin, &tr, &tg, &tb, &ta);
no[2]+= (old-tin)*paf->texfac;
2002-10-12 11:37:38 +00:00
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* -------------------------- Effectors ------------------ */
typedef struct pEffectorCache {
struct pEffectorCache *next, *prev;
struct Object *ob;
/* precalculated variables */
float oldloc[3], oldspeed[3];
float scale, time_scale;
float guide_dist;
} pEffectorCache;
/* returns ListBase handle with objects taking part in the effecting */
ListBase *pdInitEffectors(unsigned int layer)
{
static ListBase listb={NULL, NULL};
Base *base;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
for(base = G.scene->base.first; base; base= base->next) {
if( (base->lay & layer) && base->object->pd) {
Object *ob= base->object;
PartDeflect *pd= ob->pd;
if(pd->forcefield == PFIELD_GUIDE) {
if(ob->type==OB_CURVE) {
Curve *cu= ob->data;
if(cu->flag & CU_PATH) {
pEffectorCache *ec= MEM_callocN(sizeof(pEffectorCache), "effector cache");
if(cu->path==NULL || cu->path->data==NULL)
makeDispListCurveTypes(ob, 0);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
ec->ob= ob;
BLI_addtail(&listb, ec);
}
}
}
else if(pd->forcefield) {
pEffectorCache *ec= MEM_callocN(sizeof(pEffectorCache), "effector cache");
ec->ob= ob;
BLI_addtail(&listb, ec);
}
}
}
if(listb.first)
return &listb;
return NULL;
}
void pdEndEffectors(ListBase *lb)
{
if(lb)
BLI_freelistN(lb);
}
/* local for this c file, only for guides now */
static void precalc_effectors(Object *ob, PartEff *paf, Particle *pa, ListBase *lb)
{
pEffectorCache *ec;
for(ec= lb->first; ec; ec= ec->next) {
if(ec->ob->type==OB_CURVE) {
float vec[4], dir[3];
/* scale corrects speed vector to curve size */
if(paf->totkey>1) ec->scale= (paf->totkey-1)/pa->lifetime;
else ec->scale= 1.0f;
/* time_scale is for random life */
if(pa->lifetime>paf->lifetime)
ec->time_scale= paf->lifetime/pa->lifetime;
else
ec->time_scale= pa->lifetime/paf->lifetime;
/* distance of first path point to particle origin */
where_on_path(ec->ob, 0.0f, vec, dir);
VECCOPY(ec->oldloc, vec); /* store local coord for differences */
Mat4MulVecfl(ec->ob->obmat, vec);
/* for static we need to move to global space */
if(paf->flag & PAF_STATIC) {
VECCOPY(dir, pa->co);
Mat4MulVecfl(ob->obmat, dir);
ec->guide_dist= VecLenf(vec, dir);
}
else
ec->guide_dist= VecLenf(vec, pa->co);
}
}
}
/* -------- pdDoEffectors() --------
generic force/speed system, now used for particles and softbodies
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
lb = listbase with objects that take part in effecting
opco = global coord, as input
force = force accumulator
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
speed = actual current speed which can be altered
cur_time = "external" time in frames, is constant for static particles
loc_time = "local" time in frames, range <0-1> for the lifetime of particle
par_layer = layer the caller is in
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
flags = only used for softbody wind now
guide = old speed of particle
*/
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
void pdDoEffectors(ListBase *lb, float *opco, float *force, float *speed, float cur_time, float loc_time, unsigned int flags)
{
/*
Modifies the force on a particle according to its
relation with the effector object
Different kind of effectors include:
Forcefields: Gravity-like attractor
(force power is related to the inverse of distance to the power of a falloff value)
Vortex fields: swirling effectors
(particles rotate around Z-axis of the object. otherwise, same relation as)
(Forcefields, but this is not done through a force/acceleration)
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
Guide: particles on a path
(particles are guided along a curve bezier or old nurbs)
(is independent of other effectors)
*/
Object *ob;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
pEffectorCache *ec;
PartDeflect *pd;
float vect_to_vert[3];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
float f_force, force_vec[3];
float *obloc;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
float distance, force_val, ffall_val;
float guidecollect[3], guidedist= 0.0f;
short cur_frame;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
guidecollect[0]= guidecollect[1]= guidecollect[2]=0.0f;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Cycle through collected objects, get total of (1/(gravity_strength * dist^gravity_power)) */
/* Check for min distance here? (yes would be cool to add that, ton) */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
for(ec = lb->first; ec; ec= ec->next) {
/* object effectors were fully checked to be OK to evaluate! */
ob= ec->ob;
pd= ob->pd;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Get IPO force strength and fall off values here */
if (has_ipo_code(ob->ipo, OB_PD_FSTR))
force_val = IPO_GetFloatValue(ob->ipo, OB_PD_FSTR, cur_time);
else
force_val = pd->f_strength;
if (has_ipo_code(ob->ipo, OB_PD_FFALL))
ffall_val = IPO_GetFloatValue(ob->ipo, OB_PD_FFALL, cur_time);
else
ffall_val = pd->f_power;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Need to set r.cfra for paths (investigate, ton) (uses ob->ctime now, ton) */
if(ob->ctime!=cur_time) {
cur_frame = G.scene->r.cfra;
G.scene->r.cfra = (short)cur_time;
where_is_object_time(ob, cur_time);
G.scene->r.cfra = cur_frame;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* use center of object for distance calculus */
obloc= ob->obmat[3];
VECSUB(vect_to_vert, obloc, opco);
distance = VecLength(vect_to_vert);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if((pd->flag & PFIELD_USEMAX) && distance>pd->maxdist && pd->forcefield != PFIELD_GUIDE)
; /* don't do anything */
else if(pd->forcefield == PFIELD_WIND) {
VECCOPY(force_vec, ob->obmat[2]);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* wind works harder perpendicular to normal, would be nice for softbody later (ton) */
/* Limit minimum distance to vertex so that */
/* the force is not too big */
if (distance < 0.001) distance = 0.001f;
f_force = (force_val)*(1/(1000 * (float)pow((double)distance, (double)ffall_val)));
/* this option for softbody only */
if(flags && PE_WIND_AS_SPEED){
speed[0] -= (force_vec[0] * f_force );
speed[1] -= (force_vec[1] * f_force );
speed[2] -= (force_vec[2] * f_force );
}
else{
force[0] += force_vec[0]*f_force;
force[1] += force_vec[1]*f_force;
force[2] += force_vec[2]*f_force;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
}
else if(pd->forcefield == PFIELD_FORCE) {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* only use center of object */
obloc= ob->obmat[3];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Now calculate the gravitational force */
VECSUB(vect_to_vert, obloc, opco);
distance = VecLength(vect_to_vert);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Limit minimum distance to vertex so that */
/* the force is not too big */
if (distance < 0.001) distance = 0.001f;
f_force = (force_val)*(1.0/(1000.0 * (float)pow((double)distance, (double)ffall_val)));
force[0] += (vect_to_vert[0] * f_force );
force[1] += (vect_to_vert[1] * f_force );
force[2] += (vect_to_vert[2] * f_force );
}
else if(pd->forcefield == PFIELD_VORTEX) {
float vortexvec[3];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* only use center of object */
obloc= ob->obmat[3];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Now calculate the vortex force */
VECSUB(vect_to_vert, obloc, opco);
distance = VecLength(vect_to_vert);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
Crossf(force_vec, ob->obmat[2], vect_to_vert);
Normalise(force_vec);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Limit minimum distance to vertex so that */
/* the force is not too big */
if (distance < 0.001) distance = 0.001f;
f_force = (force_val)*(1.0/(100.0 * (float)pow((double)distance, (double)ffall_val)));
vortexvec[0]= -(force_vec[0] * f_force );
vortexvec[1]= -(force_vec[1] * f_force );
vortexvec[2]= -(force_vec[2] * f_force );
/* this option for softbody only */
if(flags &&PE_WIND_AS_SPEED) {
speed[0]+= vortexvec[0];
speed[1]+= vortexvec[1];
speed[2]+= vortexvec[2];
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
else {
/* since vortex alters the speed, we have to correct for the previous vortex result */
speed[0]+= vortexvec[0] - ec->oldspeed[0];
speed[1]+= vortexvec[1] - ec->oldspeed[1];
speed[2]+= vortexvec[2] - ec->oldspeed[2];
VECCOPY(ec->oldspeed, vortexvec);
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
else if(pd->forcefield == PFIELD_GUIDE) {
float guidevec[4], guidedir[3];
float mindist= force_val; /* force_val is actually mindist in the UI */
distance= ec->guide_dist;
/* WARNING: bails out with continue here */
if((pd->flag & PFIELD_USEMAX) && distance>pd->maxdist) continue;
/* derive path point from loc_time */
where_on_path(ob, loc_time*ec->time_scale, guidevec, guidedir);
VECSUB(guidedir, guidevec, ec->oldloc);
VECCOPY(ec->oldloc, guidevec);
Mat4Mul3Vecfl(ob->obmat, guidedir);
VecMulf(guidedir, ec->scale); /* correction for lifetime and speed */
/* we subtract the speed we gave it previous step */
VECCOPY(guidevec, guidedir);
VECSUB(guidedir, guidedir, ec->oldspeed);
VECCOPY(ec->oldspeed, guidevec);
/* calculate contribution factor for this guide */
if(distance<=mindist) f_force= 1.0f;
else if(pd->flag & PFIELD_USEMAX) {
if(distance>pd->maxdist || mindist>=pd->maxdist) f_force= 0.0f;
else {
f_force= 1.0f - (distance-mindist)/(pd->maxdist - mindist);
if(ffall_val!=0.0f)
f_force = (float)pow(f_force, ffall_val+1.0);
}
}
else {
f_force= 1.0f/(1.0f + distance-mindist);
if(ffall_val!=0.0f)
f_force = (float)pow(f_force, ffall_val+1.0);
}
/* if it fully contributes, we stop */
if(f_force==1.0) {
VECCOPY(guidecollect, guidedir);
guidedist= 1.0f;
break;
}
else if(guidedist<1.0f) {
VecMulf(guidedir, f_force);
VECADD(guidecollect, guidecollect, guidedir);
guidedist += f_force;
}
}
}
/* all guides are accumulated here */
if(guidedist!=0.0f) {
if(guidedist!=1.0f) VecMulf(guidecollect, 1.0f/guidedist);
VECADD(speed, speed, guidecollect);
}
}
static void cache_object_vertices(Object *ob)
{
Mesh *me;
MVert *mvert;
float *fp;
int a;
me= ob->data;
if(me->totvert==0) return;
fp= ob->sumohandle= MEM_mallocN(3*sizeof(float)*me->totvert, "cache particles");
mvert= me->mvert;
a= me->totvert;
while(a--) {
VECCOPY(fp, mvert->co);
Mat4MulVecfl(ob->obmat, fp);
mvert++;
fp+= 3;
}
}
static int pdDoDeflection(RNG *rng, float opco[3], float npco[3], float opno[3],
float npno[3], float life, float force[3], int def_depth,
float cur_time, unsigned int par_layer, int *last_object,
int *last_face, int *same_face)
{
/* Particle deflection code */
/* The code is in two sections: the first part checks whether a particle has */
/* intersected a face of a deflector mesh, given its old and new co-ords, opco and npco */
/* and which face it hit first */
/* The second part calculates the new co-ordinates given that collision and updates */
/* the new co-ordinates accordingly */
Base *base;
Object *ob, *deflection_object = NULL;
Mesh *def_mesh;
MFace *mface, *deflection_face = NULL;
float *v1, *v2, *v3, *v4, *vcache=NULL;
float nv1[3], nv2[3], nv3[3], nv4[3], edge1[3], edge2[3];
float dv1[3], dv2[3], dv3[3];
float vect_to_int[3], refl_vel[3];
float d_intersect_co[3], d_intersect_vect[3], d_nvect[3], d_i_co_above[3];
float forcec[3];
float k_point3, dist_to_plane;
float first_dist, ref_plane_mag;
float dk_plane=0, dk_point1=0;
float icalctop, icalcbot, n_mag;
float mag_iv, x_m,y_m,z_m;
float damping, perm_thresh;
float perm_val, rdamp_val;
int a, deflected=0, deflected_now=0;
float t,t2, min_t;
float mat[3][3], obloc[3];
short cur_frame;
float time_before, time_after;
float force_mag_norm;
int d_object=0, d_face=0, ds_object=0, ds_face=0;
first_dist = 200000;
min_t = 200000;
/* The first part of the code, finding the first intersected face*/
base= G.scene->base.first;
while (base) {
/*Only proceed for mesh object in same layer */
if(base->object->type==OB_MESH && (base->lay & par_layer)) {
ob= base->object;
/* only with deflecting set */
if(ob->pd && ob->pd->deflect) {
def_mesh= ob->data;
d_object = d_object + 1;
d_face = d_face + 1;
mface= def_mesh->mface;
a = def_mesh->totface;
if(ob->parent==NULL && ob->ipo==NULL) { // static
if(ob->sumohandle==NULL) cache_object_vertices(ob);
vcache= ob->sumohandle;
}
else {
/*Find out where the object is at this time*/
cur_frame = G.scene->r.cfra;
G.scene->r.cfra = (short)cur_time;
where_is_object_time(ob, cur_time);
G.scene->r.cfra = cur_frame;
/*Pass the values from ob->obmat to mat*/
/*and the location values to obloc */
Mat3CpyMat4(mat,ob->obmat);
obloc[0] = ob->obmat[3][0];
obloc[1] = ob->obmat[3][1];
obloc[2] = ob->obmat[3][2];
vcache= NULL;
}
while (a--) {
if(vcache) {
v1= vcache+ 3*(mface->v1);
VECCOPY(nv1, v1);
v1= vcache+ 3*(mface->v2);
VECCOPY(nv2, v1);
v1= vcache+ 3*(mface->v3);
VECCOPY(nv3, v1);
v1= vcache+ 3*(mface->v4);
VECCOPY(nv4, v1);
}
else {
/* Calculate the global co-ordinates of the vertices*/
v1= (def_mesh->mvert+(mface->v1))->co;
v2= (def_mesh->mvert+(mface->v2))->co;
v3= (def_mesh->mvert+(mface->v3))->co;
v4= (def_mesh->mvert+(mface->v4))->co;
VECCOPY(nv1, v1);
VECCOPY(nv2, v2);
VECCOPY(nv3, v3);
VECCOPY(nv4, v4);
/*Apply the objects deformation matrix*/
Mat3MulVecfl(mat, nv1);
Mat3MulVecfl(mat, nv2);
Mat3MulVecfl(mat, nv3);
Mat3MulVecfl(mat, nv4);
VECADD(nv1, nv1, obloc);
VECADD(nv2, nv2, obloc);
VECADD(nv3, nv3, obloc);
VECADD(nv4, nv4, obloc);
}
deflected_now = 0;
// t= 0.5; // this is labda of line, can use it optimize quad intersection
// sorry but no .. see below (BM)
if( LineIntersectsTriangle(opco, npco, nv1, nv2, nv3, &t) ) {
if (t < min_t) {
deflected = 1;
deflected_now = 1;
}
}
// else if (mface->v4 && (t>=0.0 && t<=1.0)) {
// no, you can't skip testing the other triangle
// it might give a smaller t on (close to) the edge .. this is numerics not esoteric maths :)
// note: the 2 triangles don't need to share a plane ! (BM)
if (mface->v4) {
if( LineIntersectsTriangle(opco, npco, nv1, nv3, nv4, &t2) ) {
if (t2 < min_t) {
deflected = 1;
deflected_now = 2;
}
}
}
if ((deflected_now > 0) && ((t < min_t) ||(t2 < min_t))) {
min_t = t;
ds_object = d_object;
ds_face = d_face;
deflection_object = ob;
deflection_face = mface;
if (deflected_now==1) {
min_t = t;
VECCOPY(dv1, nv1);
VECCOPY(dv2, nv2);
VECCOPY(dv3, nv3);
}
else {
min_t = t2;
VECCOPY(dv1, nv1);
VECCOPY(dv2, nv3);
VECCOPY(dv3, nv4);
}
}
mface++;
}
}
}
base = base->next;
}
/* Here's the point to do the permeability calculation */
/* Set deflected to 0 if a random number is below the value */
/* Get the permeability IPO here*/
if (deflected) {
if (has_ipo_code(deflection_object->ipo, OB_PD_PERM))
perm_val = IPO_GetFloatValue(deflection_object->ipo, OB_PD_PERM, cur_time);
else
perm_val = deflection_object->pd->pdef_perm;
perm_thresh = rng_getFloat(rng) - perm_val;
if (perm_thresh < 0 ) {
deflected = 0;
}
}
/* Now for the second part of the deflection code - work out the new speed */
/* and position of the particle if a collision occurred */
if (deflected) {
VECSUB(edge1, dv1, dv2);
VECSUB(edge2, dv3, dv2);
Crossf(d_nvect, edge2, edge1);
n_mag = Normalise(d_nvect);
dk_plane = INPR(d_nvect, nv1);
dk_point1 = INPR(d_nvect,opco);
VECSUB(d_intersect_vect, npco, opco);
d_intersect_co[0] = opco[0] + (min_t * (npco[0] - opco[0]));
d_intersect_co[1] = opco[1] + (min_t * (npco[1] - opco[1]));
d_intersect_co[2] = opco[2] + (min_t * (npco[2] - opco[2]));
d_i_co_above[0] = (d_intersect_co[0] + (0.001f * d_nvect[0]));
d_i_co_above[1] = (d_intersect_co[1] + (0.001f * d_nvect[1]));
d_i_co_above[2] = (d_intersect_co[2] + (0.001f * d_nvect[2]));
mag_iv = Normalise(d_intersect_vect);
VECCOPY(npco, d_intersect_co);
VECSUB(vect_to_int, opco, d_intersect_co);
first_dist = Normalise(vect_to_int);
/* Work out the lengths of time before and after collision*/
time_before = (life*(first_dist / (mag_iv)));
time_after = (life*((mag_iv - first_dist) / (mag_iv)));
/* We have to recalculate what the speed would have been at the */
/* point of collision, not the key frame time */
npno[0]= opno[0] + time_before*force[0];
npno[1]= opno[1] + time_before*force[1];
npno[2]= opno[2] + time_before*force[2];
/* Reflect the speed vector in the face */
x_m = (2 * npno[0] * d_nvect[0]);
y_m = (2 * npno[1] * d_nvect[1]);
z_m = (2 * npno[2] * d_nvect[2]);
refl_vel[0] = npno[0] - (d_nvect[0] * (x_m + y_m + z_m));
refl_vel[1] = npno[1] - (d_nvect[1] * (x_m + y_m + z_m));
refl_vel[2] = npno[2] - (d_nvect[2] * (x_m + y_m + z_m));
/*A random variation in the damping factor........ */
/*Get the IPO values for damping here*/
if (has_ipo_code(deflection_object->ipo, OB_PD_SDAMP))
damping = IPO_GetFloatValue(deflection_object->ipo, OB_PD_SDAMP, cur_time);
else
damping = deflection_object->pd->pdef_damp;
if (has_ipo_code(deflection_object->ipo, OB_PD_RDAMP))
rdamp_val = IPO_GetFloatValue(deflection_object->ipo, OB_PD_RDAMP, cur_time);
else
rdamp_val = deflection_object->pd->pdef_rdamp;
damping = damping + ((1.0f - damping) * rng_getFloat(rng) *rdamp_val);
damping = damping * damping;
ref_plane_mag = INPR(refl_vel,d_nvect);
if (damping > 0.999) damping = 0.999f;
/* Now add in the damping force - only damp in the direction of */
/* the faces normal vector */
npno[0] = (refl_vel[0] - (d_nvect[0] * ref_plane_mag * damping));
npno[1] = (refl_vel[1] - (d_nvect[1] * ref_plane_mag * damping));
npno[2] = (refl_vel[2] - (d_nvect[2] * ref_plane_mag * damping));
/* Now reset opno */
VECCOPY(opno,npno);
VECCOPY(forcec, force);
/* If the particle has bounced more than four times on the same */
/* face within this cycle (depth > 4, same face > 4 ) */
/* Then set the force to be only that component of the force */
/* in the same direction as the face normal */
/* i.e. subtract the component of the force in the direction */
/* of the face normal from the actual force */
if ((ds_object == *last_object) && (ds_face == *last_face)) {
/* Increment same_face */
*same_face = *same_face + 1;
if ((*same_face > 3) && (def_depth > 3)) {
force_mag_norm = INPR(forcec, d_nvect);
forcec[0] = forcec[0] - (d_nvect[0] * force_mag_norm);
forcec[1] = forcec[1] - (d_nvect[1] * force_mag_norm);
forcec[2] = forcec[2] - (d_nvect[2] * force_mag_norm);
}
}
else *same_face = 1;
*last_object = ds_object;
*last_face = ds_face;
/* We have the particles speed at the point of collision */
/* Now we want the particles speed at the current key frame */
npno[0]= npno[0] + time_after*forcec[0];
npno[1]= npno[1] + time_after*forcec[1];
npno[2]= npno[2] + time_after*forcec[2];
/* Now we have to recalculate pa->co for the remainder*/
/* of the time since the intersect*/
npco[0]= npco[0] + time_after*npno[0];
npco[1]= npco[1] + time_after*npno[1];
npco[2]= npco[2] + time_after*npno[2];
/* And set the old co-ordinates back to the point just above the intersection */
VECCOPY(opco, d_i_co_above);
/* Finally update the time */
life = time_after;
cur_time += time_before;
/* The particle may have fallen through the face again by now!!*/
/* So check if the particle has changed sides of the plane compared*/
/* the co-ordinates at the last keyframe*/
/* But only do this as a last resort, if we've got to the end of the */
/* number of collisions allowed */
if (def_depth==9) {
k_point3 = INPR(d_nvect,npco);
if (((dk_plane > k_point3) && (dk_plane < dk_point1))||((dk_plane < k_point3) && (dk_plane > dk_point1))) {
/* Yup, the pesky particle may have fallen through a hole!!! */
/* So we'll cheat a bit and move the particle along the normal vector */
/* until it's just the other side of the plane */
icalctop = (dk_plane - d_nvect[0]*npco[0] - d_nvect[1]*npco[1] - d_nvect[2]*npco[2]);
icalcbot = (d_nvect[0]*d_nvect[0] + d_nvect[1]*d_nvect[1] + d_nvect[2]*d_nvect[2]);
dist_to_plane = icalctop / icalcbot;
/* Now just increase the distance a little to place */
/* the point the other side of the plane */
dist_to_plane *= 1.1f;
npco[0]= npco[0] + (dist_to_plane * d_nvect[0]);
npco[1]= npco[1] + (dist_to_plane * d_nvect[1]);
npco[2]= npco[2] + (dist_to_plane * d_nvect[2]);
}
}
}
return deflected;
}
/*
rng= random number generator
ob = object that spawns the particles
depth = for fireworks
nr = index nr of current particle
paf = the particle system
part = current particle
force = force vector
deform = flag to indicate lattice deform
*/
static void make_particle_keys(RNG *rng, Object *ob, int depth, int nr, PartEff *paf, Particle *part, float *force, int deform, MTex *mtex, ListBase *effectorbase)
2002-10-12 11:37:38 +00:00
{
Particle *pa, *opa = NULL;
float damp, deltalife, life;
float cur_time;
float opco[3], opno[3], npco[3], npno[3], new_force[3], new_speed[3];
int b, rt1, rt2, deflected, deflection, finish_defs, def_count;
int last_ob, last_fc, same_fc;
2002-10-12 11:37:38 +00:00
damp= 1.0f-paf->damp;
pa= part;
/* start speed: random */
2002-10-12 11:37:38 +00:00
if(paf->randfac!=0.0) {
pa->no[0]+= paf->randfac*(rng_getFloat(rng) - 0.5f);
pa->no[1]+= paf->randfac*(rng_getFloat(rng) - 0.5f);
pa->no[2]+= paf->randfac*(rng_getFloat(rng) - 0.5f);
2002-10-12 11:37:38 +00:00
}
/* start speed: texture */
2002-10-12 11:37:38 +00:00
if(mtex && paf->texfac!=0.0) {
particle_tex(mtex, paf, pa->co, pa->no);
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* effectors here? */
if(effectorbase)
precalc_effectors(ob, paf, pa, effectorbase);
if(paf->totkey>1) deltalife= pa->lifetime/(paf->totkey-1);
else deltalife= pa->lifetime;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* longer lifetime results in longer distance covered */
VecMulf(pa->no, deltalife);
opa= pa;
pa++;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
for(b=1; b<paf->totkey; b++) {
/* new time */
pa->time= opa->time+deltalife;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
cur_time = pa->time;
/* set initial variables */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
VECCOPY(opco, opa->co);
VECCOPY(new_force, force);
VECCOPY(new_speed, opa->no);
VecMulf(new_speed, 1.0f/deltalife);
//new_speed[0] = new_speed[1] = new_speed[2] = 0.0f;
/* handle differences between static (local coords, fixed frame) and dynamic */
if(effectorbase) {
float loc_time= ((float)b)/(float)(paf->totkey-1);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(paf->flag & PAF_STATIC) {
float opco1[3], new_force1[3];
/* move co and force to global coords */
VECCOPY(opco1, opco);
Mat4MulVecfl(ob->obmat, opco1);
VECCOPY(new_force1, new_force);
Mat4Mul3Vecfl(ob->obmat, new_force1);
Mat4Mul3Vecfl(ob->obmat, new_speed);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
cur_time = G.scene->r.cfra;
/* force fields */
pdDoEffectors(effectorbase, opco1, new_force1, new_speed, cur_time, loc_time, 0);
/* move co, force and newspeed back to local */
VECCOPY(opco, opco1);
Mat4MulVecfl(ob->imat, opco);
VECCOPY(new_force, new_force1);
Mat4Mul3Vecfl(ob->imat, new_force);
Mat4Mul3Vecfl(ob->imat, new_speed);
}
else {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* force fields */
pdDoEffectors(effectorbase, opco, new_force, new_speed, cur_time, loc_time, 0);
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* new speed */
pa->no[0]= deltalife * (new_speed[0] + new_force[0]);
pa->no[1]= deltalife * (new_speed[1] + new_force[1]);
pa->no[2]= deltalife * (new_speed[2] + new_force[2]);
/* new location */
pa->co[0]= opa->co[0] + pa->no[0];
pa->co[1]= opa->co[1] + pa->no[1];
pa->co[2]= opa->co[2] + pa->no[2];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Particle deflection code */
if((paf->flag & PAF_STATIC)==0) {
deflection = 0;
finish_defs = 1;
def_count = 0;
VECCOPY(opno, opa->no);
VECCOPY(npco, pa->co);
VECCOPY(npno, pa->no);
life = deltalife;
cur_time -= deltalife;
last_ob = -1;
last_fc = -1;
same_fc = 0;
/* First call the particle deflection check for the particle moving */
/* between the old co-ordinates and the new co-ordinates */
/* If a deflection occurs, call the code again, this time between the */
/* intersection point and the updated new co-ordinates */
/* Bail out if we've done the calculation 10 times - this seems ok */
/* for most scenes I've tested */
while (finish_defs) {
deflected = pdDoDeflection(rng, opco, npco, opno, npno, life, new_force,
def_count, cur_time, ob->lay,
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
&last_ob, &last_fc, &same_fc);
if (deflected) {
def_count = def_count + 1;
deflection = 1;
if (def_count==10) finish_defs = 0;
}
else {
finish_defs = 0;
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Only update the particle positions and speed if we had a deflection */
if (deflection) {
pa->co[0] = npco[0];
pa->co[1] = npco[1];
pa->co[2] = npco[2];
pa->no[0] = npno[0];
pa->no[1] = npno[1];
pa->no[2] = npno[2];
}
}
/* speed: texture */
if(mtex && paf->texfac!=0.0) {
particle_tex(mtex, paf, pa->co, pa->no);
}
if(damp!=1.0) {
pa->no[0]*= damp;
pa->no[1]*= damp;
pa->no[2]*= damp;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
2002-10-12 11:37:38 +00:00
opa= pa;
pa++;
/* opa is used later on too! */
2002-10-12 11:37:38 +00:00
}
if(deform) {
/* deform all keys */
2002-10-12 11:37:38 +00:00
pa= part;
b= paf->totkey;
while(b--) {
calc_latt_deform(pa->co, 1.0f);
2002-10-12 11:37:38 +00:00
pa++;
}
}
/* the big multiplication */
2002-10-12 11:37:38 +00:00
if(depth<PAF_MAXMULT && paf->mult[depth]!=0.0) {
/* new 'child' emerges from an average 'mult' part from
the particles */
2002-10-12 11:37:38 +00:00
damp = (float)nr;
rt1= (int)(damp*paf->mult[depth]);
rt2= (int)((damp+1.0)*paf->mult[depth]);
if(rt1!=rt2) {
for(b=0; b<paf->child[depth]; b++) {
pa= new_particle(paf);
*pa= *opa;
pa->lifetime= paf->life[depth];
if(paf->randlife!=0.0) {
pa->lifetime*= 1.0f + paf->randlife*(rng_getFloat(rng) - 0.5f);
2002-10-12 11:37:38 +00:00
}
pa->mat_nr= paf->mat[depth];
make_particle_keys(rng, ob, depth+1, b, paf, pa, force, deform, mtex, effectorbase);
2002-10-12 11:37:38 +00:00
}
}
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
static void init_mv_jit(float *jit, int num, int seed2)
2002-10-12 11:37:38 +00:00
{
RNG *rng;
2002-10-12 11:37:38 +00:00
float *jit2, x, rad1, rad2, rad3;
2002-10-30 00:37:19 +00:00
int i, num2;
2002-10-12 11:37:38 +00:00
if(num==0) return;
rad1= (float)(1.0/sqrt((float)num));
rad2= (float)(1.0/((float)num));
rad3= (float)sqrt((float)num)/((float)num);
rng = rng_new(31415926 + num + seed2);
2002-10-12 11:37:38 +00:00
x= 0;
2002-10-30 00:37:19 +00:00
num2 = 2 * num;
for(i=0; i<num2; i+=2) {
2002-10-12 11:37:38 +00:00
jit[i]= x + rad1*(0.5f - rng_getFloat(rng));
jit[i+1]= i/(2.0f*num) + rad1*(0.5f - rng_getFloat(rng));
2002-10-12 11:37:38 +00:00
jit[i]-= (float)floor(jit[i]);
jit[i+1]-= (float)floor(jit[i+1]);
x+= rad3;
x -= (float)floor(x);
}
jit2= MEM_mallocN(12 + 2*sizeof(float)*num, "initjit");
for (i=0 ; i<4 ; i++) {
RE_jitterate1(jit, jit2, num, rad1);
RE_jitterate1(jit, jit2, num, rad1);
RE_jitterate2(jit, jit2, num, rad2);
}
MEM_freeN(jit2);
rng_free(rng);
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
#define JIT_RAND 32
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* for a position within a face, tot is total amount of faces */
static void give_mesh_particle_coord(PartEff *paf, MVert *mvert, MFace *mface, int partnr, int subnr, float *co, short *no)
2002-10-12 11:37:38 +00:00
{
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
static float *jit= NULL;
static float *trands= NULL;
static int jitlevel= 1;
float *v1, *v2, *v3, *v4;
float u, v;
2002-10-12 11:37:38 +00:00
short *n1, *n2, *n3, *n4;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* free signal */
if(paf==NULL) {
2002-10-12 11:37:38 +00:00
if(jit) MEM_freeN(jit);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
jit= NULL;
if(trands) MEM_freeN(trands);
trands= NULL;
2002-10-12 11:37:38 +00:00
return;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* first time initialize jitter or trand, partnr then is total amount of particles, subnr total amount of faces */
if(trands==NULL && jit==NULL) {
RNG *rng = rng_new(31415926 + paf->seed);
int i, tot;
if(paf->flag & PAF_TRAND)
tot= partnr;
else
tot= JIT_RAND; /* arbitrary... allows JIT_RAND times more particles in a face for jittered distro */
trands= MEM_callocN(2+2*tot*sizeof(float), "trands");
for(i=0; i<tot; i++) {
trands[2*i]= rng_getFloat(rng);
trands[2*i+1]= rng_getFloat(rng);
}
rng_free(rng);
if((paf->flag & PAF_TRAND)==0) {
jitlevel= paf->userjit;
if(jitlevel == 0) {
jitlevel= partnr/subnr;
if(paf->flag & PAF_EDISTR) jitlevel*= 2; /* looks better in general, not very scietific */
if(jitlevel<3) jitlevel= 3;
if(jitlevel>100) jitlevel= 100;
}
jit= MEM_callocN(2+ jitlevel*2*sizeof(float), "jit");
init_mv_jit(jit, jitlevel, paf->seed);
BLI_array_randomize(jit, 2*sizeof(float), jitlevel, paf->seed); /* for custom jit or even distribution */
}
return;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(paf->flag & PAF_TRAND) {
u= trands[2*partnr];
v= trands[2*partnr+1];
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
else {
/* jittered distribution gets fixed random offset */
if(subnr>=jitlevel) {
int jitrand= (subnr/jitlevel) % JIT_RAND;
subnr %= jitlevel;
u= jit[2*subnr] + trands[2*jitrand];
v= jit[2*subnr+1] + trands[2*jitrand+1];
if(u > 1.0f) u-= 1.0f;
if(v > 1.0f) v-= 1.0f;
}
else {
u= jit[2*subnr];
v= jit[2*subnr+1];
}
}
v1= (mvert+(mface->v1))->co;
v2= (mvert+(mface->v2))->co;
v3= (mvert+(mface->v3))->co;
n1= (mvert+(mface->v1))->no;
n2= (mvert+(mface->v2))->no;
n3= (mvert+(mface->v3))->no;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(mface->v4) {
float uv= u*v;
float muv= (1.0f-u)*(v);
float umv= (u)*(1.0f-v);
float mumv= (1.0f-u)*(1.0f-v);
v4= (mvert+(mface->v4))->co;
n4= (mvert+(mface->v4))->no;
co[0]= mumv*v1[0] + muv*v2[0] + uv*v3[0] + umv*v4[0];
co[1]= mumv*v1[1] + muv*v2[1] + uv*v3[1] + umv*v4[1];
co[2]= mumv*v1[2] + muv*v2[2] + uv*v3[2] + umv*v4[2];
no[0]= (short)(mumv*n1[0] + muv*n2[0] + uv*n3[0] + umv*n4[0]);
no[1]= (short)(mumv*n1[1] + muv*n2[1] + uv*n3[1] + umv*n4[1]);
no[2]= (short)(mumv*n1[2] + muv*n2[2] + uv*n3[2] + umv*n4[2]);
2002-10-12 11:37:38 +00:00
}
else {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* mirror triangle uv coordinates when on other side */
if(u + v > 1.0f) {
u= 1.0f-u;
v= 1.0f-v;
}
co[0]= v1[0] + u*(v3[0]-v1[0]) + v*(v2[0]-v1[0]);
co[1]= v1[1] + u*(v3[1]-v1[1]) + v*(v2[1]-v1[1]);
co[2]= v1[2] + u*(v3[2]-v1[2]) + v*(v2[2]-v1[2]);
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
no[0]= (short)(n1[0] + u*(n3[0]-n1[0]) + v*(n2[0]-n1[0]));
no[1]= (short)(n1[1] + u*(n3[1]-n1[1]) + v*(n2[1]-n1[1]));
no[2]= (short)(n1[2] + u*(n3[2]-n1[2]) + v*(n2[2]-n1[2]));
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Gets a MDeformVert's weight in group (0 if not in group) */
/* note; this call could be in mesh.c or deform.c, but OK... it's in armature.c too! (ton) */
static float vert_weight(MDeformVert *dvert, int group)
{
MDeformWeight *dw;
int i;
if(dvert) {
dw= dvert->dw;
for(i= dvert->totweight; i>0; i--, dw++) {
if(dw->def_nr == group) return dw->weight;
if(i==1) break; /*otherwise dw will point to somewhere it shouldn't*/
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
}
return 0.0;
}
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* Gets a faces average weight in a group, helper for below, face and weights are always set */
static float face_weight(MFace *face, float *weights)
{
float tweight;
tweight = weights[face->v1] + weights[face->v2] + weights[face->v3];
if(face->v4) {
tweight += weights[face->v4];
tweight /= 4.0;
}
else {
tweight /= 3.0;
}
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
return tweight;
}
/* helper function for build_particle_system() */
static void make_weight_tables(PartEff *paf, Mesh *me, int totpart, MVert *vertlist, int totvert, MFace *facelist, int totface, float **vweights, float **fweights)
{
MFace *mface;
float *foweights=NULL, *voweights=NULL;
float totvweight=0.0f, totfweight=0.0f;
int a;
if((paf->flag & PAF_FACE)==0) totface= 0;
/* collect emitting vertices & faces if vert groups used */
if(paf->vertgroup && me->dvert) {
/* allocate weights array for all vertices, also for lookup of faces later on. note it's a malloc */
*vweights= voweights= MEM_mallocN( totvert*sizeof(float), "pafvoweights" );
totvweight= 0.0f;
for(a=0; a<totvert; a++) {
voweights[a]= vert_weight(me->dvert+a, paf->vertgroup-1);
totvweight+= voweights[a];
}
if(totface) {
/* allocate weights array for faces, note it's a malloc */
*fweights= foweights= MEM_mallocN(totface*sizeof(float), "paffoweights" );
for(a=0, mface=facelist; a<totface; a++, mface++) {
foweights[a] = face_weight(mface, voweights);
}
}
}
/* make weights for faces or for even area distribution */
if(totface && (paf->flag & PAF_EDISTR)) {
float maxfarea= 0.0f, curfarea;
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* two cases for area distro, second case we already have group weights */
if(foweights==NULL) {
/* allocate weights array for faces, note it's a malloc */
*fweights= foweights= MEM_mallocN(totface*sizeof(float), "paffoweights" );
for(a=0, mface=facelist; a<totface; a++, mface++) {
if (mface->v4)
curfarea= AreaQ3Dfl(vertlist[mface->v1].co, vertlist[mface->v2].co, vertlist[mface->v3].co, vertlist[mface->v4].co);
else
curfarea= AreaT3Dfl(vertlist[mface->v1].co, vertlist[mface->v2].co, vertlist[mface->v3].co);
if(curfarea>maxfarea)
maxfarea = curfarea;
foweights[a]= curfarea;
}
2002-10-12 11:37:38 +00:00
}
else {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
for(a=0, mface=facelist; a<totface; a++, mface++) {
if(foweights[a]!=0.0f) {
if (mface->v4)
curfarea= AreaQ3Dfl(vertlist[mface->v1].co, vertlist[mface->v2].co, vertlist[mface->v3].co, vertlist[mface->v4].co);
else
curfarea= AreaT3Dfl(vertlist[mface->v1].co, vertlist[mface->v2].co, vertlist[mface->v3].co);
if(curfarea>maxfarea)
maxfarea = curfarea;
foweights[a]*= curfarea;
}
}
}
/* normalize weights for max face area, calculate tot */
if(maxfarea!=0.0f) {
maxfarea= 1.0f/maxfarea;
for(a=0; a< totface; a++) {
if(foweights[a]!=0.0) {
foweights[a] *= maxfarea;
totfweight+= foweights[a];
}
}
}
}
else if(foweights) {
/* only add totfweight value */
for(a=0; a< totface; a++) {
if(foweights[a]!=0.0) {
totfweight+= foweights[a];
}
}
}
/* if weight arrays, we turn these arrays into the amount of particles */
if(totvert && voweights) {
float mult= (float)totpart/totvweight;
for(a=0; a< totvert; a++) {
if(voweights[a]!=0.0)
voweights[a] *= mult;
}
}
if(totface && foweights) {
float mult= (float)totpart/totfweight;
for(a=0; a< totface; a++) {
if(foweights[a]!=0.0)
foweights[a] *= mult;
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
}
}
2002-10-12 11:37:38 +00:00
/* helper function for build_particle_system() */
static void make_length_tables(PartEff *paf, Mesh *me, int totvert, MFace *facelist, int totface, float **vlengths, float **flengths)
{
MFace *mface;
float *folengths=NULL, *volengths=NULL;
int a;
if((paf->flag & PAF_FACE)==0) totface= 0;
/* collect emitting vertices & faces if vert groups used */
if(paf->vertgroup_v && me->dvert) {
/* allocate lengths array for all vertices, also for lookup of faces later on. note it's a malloc */
*vlengths= volengths= MEM_mallocN( totvert*sizeof(float), "pafvolengths" );
for(a=0; a<totvert; a++) {
volengths[a]= vert_weight(me->dvert+a, paf->vertgroup_v-1);
}
if(totface) {
/* allocate lengths array for faces, note it's a malloc */
*flengths= folengths= MEM_mallocN(totface*sizeof(float), "paffolengths" );
for(a=0, mface=facelist; a<totface; a++, mface++) {
folengths[a] = face_weight(mface, volengths);
}
}
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* for paf start to end, store all matrices for objects */
typedef struct pMatrixCache {
float obmat[4][4];
float imat[3][3];
} pMatrixCache;
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
static pMatrixCache *cache_object_matrices(Object *ob, int start, int end)
{
pMatrixCache *mcache, *mc;
Object *par;
float framelenold, sfrao;
int cfrao;
mcache= mc= MEM_mallocN( (end-start+1)*sizeof(pMatrixCache), "ob matrix cache");
framelenold= G.scene->r.framelen;
G.scene->r.framelen= 1.0f;
cfrao= G.scene->r.cfra;
sfrao= ob->sf;
ob->sf= 0.0f;
for(G.scene->r.cfra= start; G.scene->r.cfra<=end; G.scene->r.cfra++, mc++) {
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
par= ob;
while(par) {
par->ctime= -1234567.0; /* hrms? */
do_ob_key(par);
if(par->type==OB_ARMATURE) {
do_all_pose_actions(par); // only does this object actions
where_is_pose(par);
}
par= par->parent;
}
where_is_object(ob);
Mat4CpyMat4(mc->obmat, ob->obmat);
Mat4Invert(ob->imat, ob->obmat);
Mat3CpyMat4(mc->imat, ob->imat);
Mat3Transp(mc->imat);
}
/* restore */
G.scene->r.cfra= cfrao;
G.scene->r.framelen= framelenold;
ob->sf= sfrao;
/* restore hierarchy */
par= ob;
while(par) {
/* do not do ob->ipo: keep insertkey */
par->ctime= -1234567.0; /* hrms? */
do_ob_key(par);
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(par->type==OB_ARMATURE) {
do_all_pose_actions(par); // only does this object actions
where_is_pose(par);
}
par= par->parent;
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
where_is_object(ob);
return mcache;
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* main particle building function
one day particles should become dynamic (realtime) with the current method as a 'bake' (ton) */
2002-10-12 11:37:38 +00:00
void build_particle_system(Object *ob)
{
RNG *rng;
2002-10-12 11:37:38 +00:00
PartEff *paf;
Particle *pa;
Mesh *me;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
Base *base;
MTex *mtexmove=0, *mtextime=0;
2002-10-12 11:37:38 +00:00
Material *ma;
DispListMesh *dlm;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
MFace *facelist= NULL;
MVert *vertlist= NULL;
DerivedMesh *dm;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
pMatrixCache *mcache=NULL, *mcnow, *mcprev;
ListBase *effectorbase;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
double startseconds= PIL_check_seconds_timer();
float ftime, dtime, force[3], vec[3];
float fac, co[3];
float *voweights= NULL, *foweights= NULL, maxw=1.0f;
float *volengths= NULL, *folengths= NULL;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
int deform=0, a, totpart, paf_sta, paf_end;
int dmNeedsFree, waitcursor_set= 0, totvert, totface, curface, curvert;
2002-10-12 11:37:38 +00:00
short no[3];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* return conditions */
2002-10-12 11:37:38 +00:00
if(ob->type!=OB_MESH) return;
me= ob->data;
2002-10-12 11:37:38 +00:00
paf= give_parteff(ob);
if(paf==NULL) return;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(paf->keys) MEM_freeN(paf->keys); /* free as early as possible, for returns */
paf->keys= NULL;
if(paf->end < paf->sta) return;
if( (paf->flag & PAF_OFACE) && (paf->flag & PAF_FACE)==0) return;
if(me->totvert==0) return;
/* this call returns NULL during editmode, just ignore it and
* particles should be recalc'd on exit. */
dm = mesh_get_derived_deform(ob, &dmNeedsFree);
if (dm==NULL)
return;
totpart= (R.flag & R_RENDERING)?paf->totpart:(paf->disp*paf->totpart)/100;
if(totpart==0) return;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* No returns after this line! */
/* material */
ma= give_current_material(ob, paf->omat);
if(ma) {
if(paf->speedtex)
mtexmove= ma->mtex[paf->speedtex-1];
mtextime= ma->mtex[paf->timetex-1];
}
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
disable_speed_curve(1); /* check this... */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* initialize particles */
new_particle(paf);
2002-10-12 11:37:38 +00:00
/* reset deflector cache, sumohandle is free, but its still sorta abuse... (ton) */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
for(base= G.scene->base.first; base; base= base->next)
base->object->sumohandle= NULL;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* all object positions from start to end */
paf_sta= (int)floor(paf->sta);
paf_end= (int)ceil(paf->end);
if((paf->flag & PAF_STATIC)==0)
mcache= cache_object_matrices(ob, paf_sta, paf_end);
/* mult generations? */
2002-10-12 11:37:38 +00:00
for(a=0; a<PAF_MAXMULT; a++) {
if(paf->mult[a]!=0.0) {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* interesting formula! this way after 'x' generations the total is paf->totpart */
2002-10-12 11:37:38 +00:00
totpart= (int)(totpart / (1.0+paf->mult[a]*paf->child[a]));
}
else break;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* for static particles, calculate system on current frame (? ton) */
if(ma) do_mat_ipo(ma);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* matrix invert for static too */
Mat4Invert(ob->imat, ob->obmat);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* new random generator */
rng = rng_new(paf->seed);
2002-10-12 11:37:38 +00:00
/* otherwise it goes way too fast */
2002-10-12 11:37:38 +00:00
force[0]= paf->force[0]*0.05f;
force[1]= paf->force[1]*0.05f;
force[2]= paf->force[2]*0.05f;
if( paf->flag & PAF_STATIC ) deform= 0;
else {
deform= (ob->parent && ob->parent->type==OB_LATTICE && ob->partype==PARSKEL);
if(deform) init_latt_deform(ob->parent, 0);
}
2002-10-12 11:37:38 +00:00
/* get the effectors */
effectorbase= pdInitEffectors(ob->lay);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* init geometry */
dlm = dm->convertToDispListMesh(dm, 1);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* subsurfs don't have vertexgroups, so we need to use me->mvert in that case */
if(paf->vertgroup && me->dvert && (dlm->totvert != me->totvert)) {
vertlist= me->mvert;
facelist= me->mface;
totvert= me->totvert;
totface= me->totface;
}
else {
vertlist= dlm->mvert;
facelist= dlm->mface;
totvert= dlm->totvert;
totface= dlm->totface;
}
/* if vertexweights or even distribution, it makes weight tables, also checks where it emits from */
make_weight_tables(paf, me, totpart, vertlist, totvert, facelist, totface, &voweights, &foweights);
/* vertexweights can define lengths too */
make_length_tables(paf, me, totvert, facelist, totface, &volengths, &folengths);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* now define where to emit from, if there are face weights we skip vertices */
if(paf->flag & PAF_OFACE) totvert= 0;
if((paf->flag & PAF_FACE)==0) totface= 0;
if(foweights) totvert= 0;
/* initialize give_mesh_particle_coord */
if(totface)
give_mesh_particle_coord(paf, vertlist, facelist, totpart, totface, NULL, NULL);
/* correction for face timing when using weighted average */
if(totface && foweights) {
maxw= (paf->end-paf->sta)/foweights[0];
}
else if(totvert && voweights) {
maxw= (paf->end-paf->sta)/voweights[0];
}
/* for loop below */
if (paf->flag & PAF_STATIC) {
ftime = G.scene->r.cfra;
dtime= 0.0f;
} else {
ftime= paf->sta;
dtime= (paf->end - paf->sta)/(float)totpart;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
curface= curvert= 0;
2002-10-12 11:37:38 +00:00
for(a=0; a<totpart; a++, ftime+=dtime) {
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* we set waitcursor only when a half second expired, particles now are realtime updated */
if(waitcursor_set==0 && (a % 256)==255) {
double seconds= PIL_check_seconds_timer();
if(seconds - startseconds > 0.5) {
waitcursor(1);
waitcursor_set= 1;
}
}
2002-10-12 11:37:38 +00:00
pa= new_particle(paf);
pa->time= ftime;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* get coordinates from faces, only when vertices set to zero */
if(totvert==0 && totface) {
int curjit;
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* use weight table, we have to do faces in order to be able to use jitter table... */
if(foweights) {
if(foweights[curface] < 1.0f) {
float remainder= 0.0f;
while(remainder + foweights[curface] < 1.0f && curface<totface-1) {
remainder += foweights[curface];
curface++;
}
foweights[curface] += remainder;
maxw= (paf->end-paf->sta)/foweights[curface];
}
if(foweights[curface]==0.0f)
break; /* WARN skips here out of particle generating */
else {
if(foweights[curface] > 1.0f)
foweights[curface] -= 1.0f;
curjit= (int) foweights[curface];
give_mesh_particle_coord(paf, vertlist, facelist+curface, a, curjit, co, no);
/* time correction to make particles appear evenly, maxw does interframe (0-1) */
pa->time= paf->sta + maxw*foweights[curface];
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
else {
curface= a % totface;
curjit= a/totface;
give_mesh_particle_coord(paf, vertlist, facelist+curface, a, curjit, co, no);
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* get coordinates from vertices */
if(totvert) {
/* use weight table */
if(voweights) {
if(voweights[curvert] < 1.0f) {
float remainder= 0.0f;
while(remainder + voweights[curvert] < 1.0f && curvert<totvert-1) {
remainder += voweights[curvert];
curvert++;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
voweights[curvert] += remainder;
maxw= (paf->end-paf->sta)/voweights[curvert];
}
if(voweights[curvert]==0.0f)
break; /* WARN skips here out of particle generating */
else {
if(voweights[curvert] > 1.0f)
voweights[curvert] -= 1.0f;
/* time correction to make particles appear evenly */
pa->time= paf->sta + maxw*voweights[curvert];
2002-10-12 11:37:38 +00:00
}
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
else {
curvert= a % totvert;
if(a >= totvert && totface)
totvert= 0;
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
VECCOPY(co, vertlist[curvert].co);
VECCOPY(no, vertlist[curvert].no);
2002-10-12 11:37:38 +00:00
}
VECCOPY(pa->co, co);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* dynamic options */
if((paf->flag & PAF_STATIC)==0) {
int cur;
/* particle retiming with texture */
if(mtextime && (paf->flag2 & PAF_TEXTIME)) {
float tin, tr, tg, tb, ta, orco[3];
/* calculate normalized orco */
orco[0] = (co[0]-me->loc[0])/me->size[0];
orco[1] = (co[1]-me->loc[1])/me->size[1];
orco[2] = (co[2]-me->loc[2])/me->size[2];
externtex(mtextime, orco, &tin, &tr, &tg, &tb, &ta);
if(paf->flag2neg & PAF_TEXTIME)
pa->time = paf->sta + (paf->end - paf->sta)*tin;
else
pa->time = paf->sta + (paf->end - paf->sta)*(1.0f-tin);
}
/* set ob at correct time, we use cached matrices */
cur= (int)floor(pa->time) + 1 ; /* + 1 has a reason: (obmat/prevobmat) otherwise comet-tails start too late */
if(cur <= paf_end) mcnow= mcache + cur - paf_sta;
else mcnow= mcache + paf_end - paf_sta + 1;
if(cur > paf_sta) mcprev= mcnow-1;
else mcprev= mcache;
/* move to global space */
Mat4MulVecfl(mcnow->obmat, pa->co);
2002-10-12 11:37:38 +00:00
VECCOPY(vec, co);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
Mat4MulVecfl(mcprev->obmat, vec);
2002-10-12 11:37:38 +00:00
/* first start speed: object */
VECSUB(pa->no, pa->co, vec);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
2002-10-12 11:37:38 +00:00
VecMulf(pa->no, paf->obfac);
/* calculate the correct inter-frame */
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
fac= (pa->time- (float)floor(pa->time));
2002-10-12 11:37:38 +00:00
pa->co[0]= fac*pa->co[0] + (1.0f-fac)*vec[0];
pa->co[1]= fac*pa->co[1] + (1.0f-fac)*vec[1];
pa->co[2]= fac*pa->co[2] + (1.0f-fac)*vec[2];
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* start speed: normal */
if(paf->normfac!=0.0) {
/* imat is transpose ! */
VECCOPY(vec, no);
Mat3MulVecfl(mcnow->imat, vec);
Normalise(vec);
VecMulf(vec, paf->normfac);
VECADD(pa->no, pa->no, vec);
}
}
else {
if(paf->normfac!=0.0) {
VECCOPY(pa->no, no);
Normalise(pa->no);
VecMulf(pa->no, paf->normfac);
}
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
2002-10-12 11:37:38 +00:00
pa->lifetime= paf->lifetime;
if(paf->randlife!=0.0) {
pa->lifetime*= 1.0f + paf->randlife*(rng_getFloat(rng) - 0.5f);
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
pa->mat_nr= paf->omat;
2002-10-12 11:37:38 +00:00
if(folengths)
pa->lifetime*= folengths[curface];
make_particle_keys(rng, ob, 0, a, paf, pa, force, deform, mtexmove, effectorbase);
2002-10-12 11:37:38 +00:00
}
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
/* free stuff */
give_mesh_particle_coord(NULL, NULL, NULL, 0, 0, NULL, NULL);
if(voweights) MEM_freeN(voweights);
if(foweights) MEM_freeN(foweights);
if(volengths) MEM_freeN(volengths);
if(folengths) MEM_freeN(folengths);
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(mcache) MEM_freeN(mcache);
2002-10-12 11:37:38 +00:00
The long awaited Particle patch from Janne Karhu http://www.blender3d.org/cms/New_Particle_options_a.721.0.html There's no doubt this patch had a lot of good ideas for features, and I want to compliment Janne again for getting it all to work even! A more careful review of the features and code did show however quite some flaws and bugs... partially because the current particle code was very much polluted already, but also because of the implementation lacked quality. However, the patch was too good to reject, so I've fixed and recoded the parts that needed it most. :) Here's a list of of most evident changes in the patch; - Guides support recoded. It was implemented as a true 'force field', checking all Curve path points for each particle to find the closest. Was just far too slow, and didn't support looping or bends well. The new implementation is fast (real time) and treats the paths as actual trajectory for the particle. - Guides didn't integrate in the physics/speed system either, was added as exception. Now it's integrated and can be combined with other velocities or forces - Use of Fields was slow code in general, made it use a Cache instead. - The "even" distribution didn't work for Jittered sample patterns. - The "even" or "vertexgroup" code in the main loops were badly constructed, giving too much cpu for a simple task. Instead of going over all faces many times, it now only does it once. Same part of the code used a lot of temporal unneeded mallocs. - Use of DerivedMesh or Mesh was confused, didn't work for Subsurfs in all cases - Support for vertex groups was slow, evaluating vertexgroups too often - When a vertexgroup failed to read, it was wrongly handled (set to zero). VertexGroup support now is with a name. - Split up the too huge build_particle() call in some parts (moving new code) - The "texture re-timing" option failed for moving Objects. The old code used the convention that particles were added with increasing time steps. Solved by creating a object Matrix Cache. Also: the texture coordinates had to be corrected to become "OrCo". - The "Disp" option only was used to draw less particles. Changed it to actually calculate fewer particles for 3D viewing, but render all still. So now it can be used to keep editing realtime. Removed; The "speed threshold" and "Tight" features were not copied over. This resembled too much to feature overkill. Needs re-evaluation. Also the "Deform" option was not added, I prefer to first check if the current particle system really works for the Modifier system. And: - Added integration for particle force fields in the dependency graph - Added TAB completion for vertexgroup names! - Made the 'wait cursor' only appear when particles take more than 0.5 sec - The particle jitter table order now is randomized too, giving much nicer emitting of particles in large faces. - Vortex field didn't correctly use speed/forces, so it didn't work for collisions. - Triangle distribution was wrong - Removed ancient bug that applied in a *very* weird way speed and forces. (location changes got the half force, speed the full...???) So much... might have forgotten some notes! :)
2005-11-10 16:01:56 +00:00
if(deform) end_latt_deform();
if(effectorbase)
pdEndEffectors(effectorbase);
/* reset deflector cache */
for(base= G.scene->base.first; base; base= base->next) {
if(base->object->sumohandle) {
MEM_freeN(base->object->sumohandle);
base->object->sumohandle= NULL;
}
}
2002-10-12 11:37:38 +00:00
disable_speed_curve(0);
2002-10-12 11:37:38 +00:00
waitcursor(0);
displistmesh_free(dlm);
if (dmNeedsFree) dm->release(dm);
rng_free(rng);
2002-10-12 11:37:38 +00:00
}