[ODE] Trimesh problems

[gl]

This might help indirectly - the heightmap contribution works correctly with 
spheres, at least last time I worked with it.  I had fairly large terrain 
tris, and spheres rolled nicely on them, didn't jump at edges etc.  In fact, 
I even wrote a little custom sphere/tri collider for it IIRC, but even the 
default worked for me.

If you check the code you might find the problem in the tricollider (if one 
exists).
--
gl

----- Original Message ----- 
From: Flavien Brebion
Cc: ode at q12.org
Sent: Tuesday, January 10, 2006 6:14 PM
Subject: Re: [ODE] Trimesh problems


Bram Stolk wrote:
Anders Olofsson wrote:
  That doesn't sound right. Why would I want to push it towards some edge?.
Me hacking code in ODE is like monkey do brainsurgery. Funny but none will
survive. Well, I'll try.. Did you allways use the triangles normal instead
of normal pointing towards some edge?. I don't immediately see how this
applies to jumpy spheres on large triangles though. Did you modify
something more in the trimesh/sphere code?.

I still think you should switch to UNSTABLE, look at this diff:

http://cvs.sourceforge.net/viewcvs.py/opende/ode/ode/src/collision_trimesh_sphere.cpp?r1=1.7&r2=1.7.2.2

  Bram
  For what it's worth.. i tried to apply this patch to my standard 0.5 
version in my Minas Tirith demo, and this modification:

//Dist= Radius - Dist;
Dist= Radius - side; // (mg) penetration depth is distance along normal not 
shortest distance

Make all my spheres bounce like hell ( and never come at a rest ) on flat 
triangles..

I'll stay with the stable version for now.

F. Brebion




--------------------------------------------------------------------------------


_______________________________________________
ODE mailing list
ODE at q12.org
http://q12.org/mailman/listinfo/ode