particle system volume grid particles out of volume #52732
OS: Ubuntu 16.04.2 LTS
GPU: TITAN X (Pascal)
CPU: Intel(R) Core(TM) i7-6700K
Broken: 2.78 and 2.79
Short description of error
Lines of particles appear from -x to x from the mesh with a high-poly (69451 faces) mesh. After decimating the mesh the problem disappears.
Exact steps for others to reproduce the error
1 - Download a high poly STL (e.g. the stanford bunny (69451 faces) )
2 - Import mesh
3 - Add a particle system and emit from volume grid.
4 - Set emitter start and end to 0, resolution to 15, physics to NO and size to 0.005
5 - Duplicate
6 - Add a decimation modifier and decimate to 0.01 (693 faces) and apply
Related task: https://developer.blender.org/T46481
The issue here is caused by failed
isect_axial_line_segment_tri_v3. This patch will allow to catch that case: P529. What i'm not sure about, is why to use axial here? At least, for volume distribution. Would think something like this will make it: P530. This patch solves this issue.
@mont29, mind having a second pair of eyes here? :)
There are still issues on other grid resolutions, but those are caused by non-manifold nature of the model.
@Sergey no idea really why that intersection algo was chosen, would bet it was best available in our code at the time. ;)
So if watertight works better, P530 looks totally fine to me.
@mont29, ok, will steal the report back to self, run additional tests and commit fix.
This issue was referenced by
OS: Ubuntu 16.04
Is this truly fixed, or am I overlooking something obvious?
It seems that now no particles appear, unless you press 'invert grid'
I double checked to be sure. The following blend file works "sort of" in 2.79 official - particles appear, but some are outside the "watermelon" shape object.
No due date set.
No dependencies set.
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?