[ODE] double precision OPCODE

[nearaz]

> But why is it then necessary to compile ODE with single precision if 
> you want to use trimesh then ? I don't understand this as, if what 
> you say is true, then ODE already converts to floats for Opcode.

Because it isn't necessary! :)

In fact, a while ago (back when TriMesh wasn't part of ODE core) it had lots 
of warnings, but otherwise it nearly worked (I've got it working with doubles 
without much trouble back in ODE 0.35 times - and it gave better results).

Now, it seems that most of the warning-producing places are fixed; in fact, 
ODE's VC6 workspaces have options for both single/double+TriMesh 
configurations...


Aras Pranckevicius aka NeARAZ
http://www.gim.ktu.lt/nesnausk/nearaz/