[Bf-committers] Re: physics test suite

[Ed Halley]

On Jan 2, 2007, at 7:37 PM, trip somewhere wrote:
 > Now mix this with an old expiriment I tried before to fragment a wall
 > into pre broken shapes, at start they would jiggle just a little up a
 > little to much to ever use sinse it was always random and always
 > removed the pure clean flat ness that is a wall with windows.

This is a bugbear of ALL rigid-body physics solvers, and not likely to
find a solution soon.  The behavior of an object at rest with another
object resting on top of it is very hard to model.

The best solution is simply to disconnect physics simulation from the
"wall" of static objects until the moment that they need to be dynamic.

--
[ e d @ h a l l e y . c c ]