[ODE] Trimesh and sphere collision

[João Rangel]

Hi,

Sorry, but I already removed #define MERGECONTACTS and the result was the
same... :(

I read some threads in the ODE forum where people solve the problem this
way, but does not work with me. I am still thinking that it may be some kind
of bug that depends of some particular mesh configuration....

Thanks for your help anyway,
Joao Rangel



----- Original Message -----
From: "Hannes Norda" <norda at customred.com>
To: "João Rangel" <jpgr at mega.ist.utl.pt>
Cc: <ode at q12.org>
Sent: Sunday, August 22, 2004 7:47 AM
Subject: Re: [ODE] Trimesh and sphere collision


> Hi!
> In collision_trimesh_sphere.cpp there is a #define MERGECONTACTS. Remove
> that define and you'll get more contact points.
>
> /Hannes Norda
>
> João Rangel wrote:
>
> > Hi,
> >
> > Thanks for your replay, but i am affraid that these solutions does not
> > solve my problem. In my implementation I have a dCollision
> > array[8]. And when a sphere collides with the triMesh it returns
> > only one collision point from dCollide.
> >
> > I tried to replace the sphere geom with a box geom. The same code,
> > only with this little difference, and it works.
> >
> > With sphere wheels, it returns 1collision point. With box wheels, it
> > return 8 points. I am forced to conclude that the ode trimesh-sphere
> > collision detection does not work, but I already saw it working in Jon
> > Watte's demo (by the way, nice demo). It must be some parameter or
> > option that changes the dCollide beahavior, but I have no idea what it
> > might be. I am using ode in debug mode, with single precision and
trimesh.
> >
> > Any sugestions?
> >
> > Joao Rangel
> >
> >
> >------------------------------------------------------------------------
> >
> >_______________________________________________
> >ODE mailing list
> >ODE at q12.org
> >http://q12.org/mailman/listinfo/ode
> >
> >
>
>