InternalEdgeDemo does not appear to work

Post Reply
MaxDZ8
Posts: 149
Joined: Fri Jun 24, 2011 8:53 am

InternalEdgeDemo does not appear to work

Post by MaxDZ8 » Tue Feb 28, 2012 8:52 am

This is a follow up to my previous thread here.
I have started looking at a solution by using btBvhTriangleMeshShape for the world. I searched the forum for more information and stumbled on a google code bug track explaining what it should solve and what not.
It appears I must somehow use btInternalEdgeUtility or at least provide some connectivity information. I am considering the latter as I could very likely pull it out of my asset pipeline somehow. I have not looked at the details yet, but I wanted to try InternalEdgeDemo.

Problem is, it does not appear to work.

InternalEdgeDemo presents itself as a box and a hull sliding down an inclined quadrilateral. I'm not sure I see any difference disabling InternalEdgeUtility.

The demo also allows to shoot boxes at the plane. This is very useful. Again, I see no difference.

I'm a bit scared about what I should do at this point. If InternalEdgeDemo does not work as expected... what am I supposed to do? Perhaps I've built it incorrectly? Help would be appreciated.

openeye
Posts: 10
Joined: Sun Apr 29, 2012 3:22 pm

Re: InternalEdgeDemo does not appear to work

Post by openeye » Tue May 01, 2012 1:47 pm

Yes, I found it, as well.

When I shoot boxes in the InternalEdgeDemo, I suppose what I have to see is the boxes not to jump up at the cross triangle edge of the base btBvhTriangleMeshShape.

But those fired boxes jumping up at that cross edge. With or without InternalEdgeutility enabled or disabled.

So I do not understand this.

I tried to set different values in the triangleInfoMap, like smaller m_maxEdgeAngleThreshold. But the fired boxes still jump at the cross edge.

openeye
Posts: 10
Joined: Sun Apr 29, 2012 3:22 pm

Re: InternalEdgeDemo does not appear to work

Post by openeye » Tue May 01, 2012 6:04 pm

I think that I found the "solution".

It is in my own thread :

http://bulletphysics.org/Bullet/phpBB3/ ... f=9&t=8082

Post Reply