Real-Time Physics Simulation Forum

[Erwin Coumans] So you make a difference between colliding contact and contact points? In Bullet I don't make any distinguisment. As soon as objects 'collide', the collision detection just adds contact points. Those are dealt with in the solver. Yes. You use the same algorithm for solving collisions...

When do you perform collision detection? In between step 1 and 2? Collision detection and resolution I perform before step 1. I made the assumption that collisions occur in an infinitesimal time intervall. So collisions can be resolved before the time step. Just contacts and joints have constraints...

My approach works for joint and contact constraints as follows. A simulation step consists of these parts: 1) Determine the position of the joint or contact points after the next simulation step (in fact a preview of the future is computed). 2) Make an approximation of the velocity change that is ne...

At the moment I am writing my PhD thesis about impulse-based dynamic simulation. In contrast to Mirtich's work joint constraints are simulated by impulses as well as collisions and contacts. So if you are interested in my work, take a look at:

http://www.ibds.de.vu

[raigan2] In fact it is not important, if you use a Taylor series, a Runge-Kutta or just Euler for integration, also for the predictor. The only difference is of course the accuracy. The simulation step works as follows: http://i31www.ira.uka.de/projekte/mechanik/img/Pic8.gif - iterate to determine ...

[raigan2] Maybe you want to try with the approach I presented in this paper: http://i31www.ira.uka.de/docs/Paper_VC05.pdf In fact it is nearly the same method as R. Weinstein has presented. At my institute we use this method since many years. And I think this should converge very fast. I also use a ...