[ODE] Ways to contribute brainpower to ODE? (Higher order integrator?)

[Jon Watte (ODE)]

STenyaK (Bruno Gonzalez) wrote:
> What's the current state of moving trimesh-trimesh colisions? I think this  
> area needed some polishing?

Note that no current physics API does that well, at least not with 
enough CPU to spare. Even Ageia needs a buttload of memory for their 
pre-processed data (which takes lots of time to generate) for concave 
trimesh<->trimesh collision.

I'd rather we focused on a convex primitive, and made that work. Perhaps 
with an additional heightmap primitive (ideally with configurable basis 
vectors) -- but general convex trimesh<->trimesh is just too hard a 
problem, even theoretically. There's lots of lower hanging fruit.

Cheers,

			/ h+