Advertisement

Rigid body simulation

Started by May 31, 2002 04:09 PM
0 comments, last by NewDeal 22 years, 8 months ago
Hi all, exams are over and i suddenly have a lot of time on my hands. For the past few weeks ive been playing with the thought of creating a rigid body simulator. My goal is not an engine suitable for games but rather an accurate simulation of physics. For instance: If i place a sphere/ball above a tilted plane i want the ball to drop, bounce some, and the begin rotation/translation according to the torque resulting from gravity/friction. Simple as that. In the perfect simulation i should be able to model a car and have it behave in a realistic manner. Or place some balls on a table and simulate a game of 8-ball. You get the picture. I have a good grasp of the physics/math behind but im wondering if its too much of a challenge implementing. Any thoughts ? Any resources you can recommend ? Am i in over my head ? Any known problems i might run into ? Thanks in advance
A physics engine is not a very easy thing to implement (ofcourse depending on the type of engine you want), you will encounter a ton of things that look simple, but are in fact quite difficult. Things like moving to the next frame suddenly isn''t as easy as x = x+v*t anymore. You get all kinds of instability problems etc if you are implementing a physics engine.

Papers I can recommend are of course the Baraff papers:
http://www-2.cs.cmu.edu/~baraff/sigcourse/index.html

And for a completely different look on physics in games I can recommend
http://www.ioi.dk/Homepages/tj/publications/gdc2001.htm

Cheers!
Nick

This topic is closed to new replies.

Advertisement