[ODE] center of mass

[Jon Watte (ODE)]

J. Perkins wrote:

>>If Jon and Jason agree on this change, I'm inclined to commit it to
>>UNSTABLE.  I do not yet have CVS write access, so I'll need that if I am
>>to commit the patch myself.

> I can give you CVS access, if Jon will give a final okay. He clearly
> understands the performance issues better than I do.

I don't think anyone will have a problem with the performance 
difference, given that if you don't use the offset at all, the branch 
will predict perfectly :-) The LCP solver, and trudging through the AABB 
trees inside OPCODE, are usually the heavy parts of using ODE anyway, so 
  I think we should take it. (I haven't read the patch, but from 
discussion, the approach sounds sound)

Cheers,

			/ h+