> It helps a lot if the forces are differentiable against object movement. If there are infinitesimal movements which result in discontinuous changes in forces, physics models won't work right.
There are models of non-smooth interactions that get used in mechanics, graphics, and robotics - I'm thinking Baraff and Moreau and Stewart and Trinkle type stuff. You end up with dynamics formulated as differential inclusions, but it all still works out.
Impulse based collisions look great, I've seen some amazing cloth and hair simulations using impulse based methods. Impulse based methods are used all over the place in geomechanics, where they can pretty accurately reproduce stress fields in avalanches, etc. I think there is something else going on making game objects look too light.
There are models of non-smooth interactions that get used in mechanics, graphics, and robotics - I'm thinking Baraff and Moreau and Stewart and Trinkle type stuff. You end up with dynamics formulated as differential inclusions, but it all still works out.