Closed GoogleCodeExporter closed 9 years ago
Yes, this is a known limitations of gimpact, it simply finds overlapping
triangles and performs a local penetration check.
We'll have to come up with a better solution.
Alternatively, it would be great to add convex decomposition in Blender, so you
can use a btCompoundShape with btConvexHull child shapes. HACD can do this.
Do you know someone who wants to help integrating HACD into Blender?
Original comment by erwin.coumans
on 7 Dec 2012 at 7:23
Adding convex decomposition in blender is on my todo list so I'll get around to
integrating HACD eventually.
Original comment by sergej.r...@googlemail.com
on 7 Dec 2012 at 9:26
I added a converter from gimpact to btCompoundShape.
Could you try that in Blender?
You basically need to update those 3 files from this commit:
https://code.google.com/p/bullet/source/detail?r=2631
Then
#include "BulletCollision/Gimpact/btCompoundFromGimpact.h
btCompoundShape* btCreateCompoundFromGimpactShape(const btGImpactMeshShape*
gimpactMesh, btScalar depth);
Thanks,
Erwin
Original comment by erwin.coumans
on 14 Dec 2012 at 11:16
I committed the fix to Blender trunk, see
http://www.miikahweb.com/en/blender/svn-logs/commit/53019
Original comment by erwin.coumans
on 15 Dec 2012 at 1:12
Seems to work well, have to test if there are cases where gimpact works better.
Might be worth it to have an option to choose between the two.
Original comment by sergej.r...@googlemail.com
on 21 Dec 2012 at 2:46
The performance of the btCompoundShape can be improved a bit. There are also
other ways to improve collisions between concave triangle meshes, using
continuous collision detection. Most that is already in Bullet, a few things
needs to be hooked up.
Another nice feature to add to Blender is .bullet file export and import.
Export is only 5 lines of code, once you have a bt*DynamicsWorld. That way, you
can use Blender to create collision and ragdoll/constraint setup and import it
into a game. It is also useful to report bugs here.
Original comment by erwin.coumans
on 21 Dec 2012 at 5:28
See https://github.com/bulletphysics/bullet3/issues/106
Original comment by erwin.coumans
on 30 Mar 2014 at 6:24
Original issue reported on code.google.com by
sergej.r...@googlemail.com
on 20 Oct 2012 at 5:44