improve performance of raycast versus compound shapes about bullet

[GoogleCodeExporter]

We could start using the btOptimizedBvh (dynamic AABB tree) to accelerate 
queries.

Original comment by erwin.coumans on 31 Mar 2008 at 4:41

[GoogleCodeExporter]

Issue 281 has been merged into this issue.

Original comment by erwin.coumans on 21 Dec 2009 at 11:02

[GoogleCodeExporter]

Patch to use DynamicAabbTree if available to accelerate queries.

Original comment by [email protected] on 26 Feb 2011 at 12:23

Attachments:

• compound.patch

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

I had a further look at the patch, and transforming the rayFrom/rayTo into 
local space of the btCompoundShape makes things work. 

It is committed in latest trunk:
http://code.google.com/p/bullet/source/detail?r=2300

It isn't thoroughly checked/tested, so in case there are strange issues with 
raycast versus compound shape, there is a #define to disable it in 
btCollisionWorld
#define DISABLE_DBVT_COMPOUNDSHAPE_RAYCAST_ACCELERATION

Thanks for the patch!

Original comment by erwin.coumans on 26 Feb 2011 at 10:20

• Changed state: Fixed

[GoogleCodeExporter]

Here is a minimal patch to fix 64-bit compile error: cast from ‘void*’ to 
‘int’ loses precision.

Original comment by [email protected] on 28 Feb 2011 at 9:04

Attachments:

• void_to_int.patch

[GoogleCodeExporter]

Applied in latest trunk:
http://code.google.com/p/bullet/source/detail?r=2319

thanks again for the help!

Original comment by erwin.coumans on 28 Feb 2011 at 5:24