[Spoam]

Originally Posted by gblake

Developing a generic physics-based simulator that would be usefully accurate, is another tall order (see my other post). I would not be optimistic, except at the level of approximating things like generic high-speed and low-torque, or low-speed and high torque, (or middle-of-the-road on both) options.

In a separate, but related vein, developing a software-interfaces simulator that would help teams develop, test, and integrate their robot code would definitely be feasible without falling too far down too many rabbit holes. Someone should do it.

Blake

See my post, here.