[ODE] double precision OPCODE

[Aras Pranckevicius]

> is there ever going to be a double-precision OPCODE ? is anybody working
> on one ?

OPCODE doesn't need it. It just detects overlapping primitives - floats
suffice here. Of course, the response code (ODE's TriMesh) does need (or:
can benefit from) double precision - and it can use it.

In fact, if you use quantized trees in OPCODE, then it's not even floats,
but some 16 bit quantities (I don't know, whether some float format or just
integers).


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