Possible fix for all the particles related SIMD SVBVH bugs.
authorJanne Karhu <jhkarh@gmail.com>
Fri, 17 Sep 2010 19:02:19 +0000 (19:02 +0000)
committerJanne Karhu <jhkarh@gmail.com>
Fri, 17 Sep 2010 19:02:19 +0000 (19:02 +0000)
* Velocity for particles that were born at exactly integer frames was calculated wrong when they were born.
Note: If you had a raytrace acceleration related bug, please clear the pointcache for all particles, toggle a particle setting to reset pointcache and rebake to create a valid simulation.

source/blender/blenkernel/intern/particle_system.c

index 598dd3c03fec7dbb9ca5d6651de036f9a11bfb79..5318754951fb28817b50a8e63b2669d43ed0fd8a 100644 (file)
@@ -3441,7 +3441,7 @@ static void dynamics_step(ParticleSimulationData *sim, float cfra)
                else if(part->phystype == PART_PHYS_NO)
                        reset_particle(sim, pa, dtime, cfra);
 
-               if(dfra>0.0 && ELEM(pa->alive,PARS_ALIVE,PARS_DYING)){
+               if(pa_dfra>0.0 && ELEM(pa->alive,PARS_ALIVE,PARS_DYING)){
                        switch(part->phystype){
                                case PART_PHYS_NEWTON:
                                        /* do global forces & effectors */