[ODE] Trimesh

[Jon Watte (ODE)]

No, this is not a problem. If the points have identical values, you 
probably will get identical results. However, if the points are 
"slightly different" then you may get hairline cracks that ray-casts can 
slip through, etc.

Proper merging of contacts depend on coincidence of position, not 
adjacency data.
Cheers,

          / h+


LR wrote:
> Declaration 1 (no duplicated points):
>     Points= 0 1 2 3
>     Faces indices = (0 1 2)  + (1 3 2)
>  
> Declaration 2 (duplicated points):
>     Points= 0 1 2 1 3 2
>     Faces indices = (0 1 2)  + (3 4 5)
>  
> Does the declaration 2 can cause problems to solver/colision detection ?
>