Jump to content
  • Advertisement
Sign in to follow this  
TheC00L1

Triangle and AABB Collision Detection and Response

This topic is 4518 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I know this seems very specific, but could you tell how to do this?, i searched the net but i got nothing that good, it was always skipping the Response part. I was going to do triangle-triangle CD & CR but it seemed to hard to implement. The triangle is usually stationary. Oh yeah this is all in 3D, incase it matters that much. Would it be better to use a OBB, instead of a AABB? Thanx in advance!

Share this post


Link to post
Share on other sites
Advertisement
I'm not sure quite what you're getting at but I'll try and help anyway. The collision detection and response can usualy be treated as completely seperate systems which is why they're often not together in one paper (at least for rigidbodies). The collision detection is concerned only with objects geometry, in this case I get the impression you've got a list of tri's for each object with an AABB or OBB hierarchy over them for tri collision culling? The collision response however is only concerned with the physical properties of the two colliding bodies (usualy described in a rigidbody structure) and the collision point and normal, and needs to know nothing of the geometry. Because of this any collision response system will do the job for you and you dont need one specificaly tailored for tri-tri collision. Like ozmic66 says, check out:

http://www.d6.com/users/checker/dynamics.htm

As far as whether OBB's or AABB's are better (at least if you're talking about a collision hierarchy like I mention above anyway) you'll find OBB's to be best most likely. This is because they give a much closer representation of the underlying shape which will save a lot of un-needed tri-tri collision tests which are expensive, and also they dont need to be recalculated each frame. The OBB collision test is more expensive than an AABB test however but its possible to create a highly optimised version of seperating axis theorem for OBB's which makes it not so bad.

Hope I managed to get to the point.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!