PDA

View Full Version : Real Time Physics

phoelix
11-05-2009, 02:36 AM
Hi there!

I've been doing lately some R&D about this topic. I have got some results and i would like to share them with you, i'll be glad if you can give me some opinions and suggestions :bounce:

All this were made using verlet integrals, and some maxscript inside scripted controllers.

(Click Image)

Experiment #1
http://joleanes.com/technical_stuff/objects/RealTimePhysics_Experiment_1.jpg (http://joleanes.com/technical_stuff/RealTimePhysics_Experiment_1.html)

Experiment #2
http://joleanes.com/technical_stuff/objects/RealTimePhysics_Experiment_2.jpg (http://joleanes.com/technical_stuff/RealTimePhysics_Experiment_2.html)

Thanks :D
eek
11-05-2009, 04:30 AM
I've been and am building a system like this at work - I cant really say much else only that yes its based on verlet and is aimed at animator friendlyness.

We have very similar studies Felix, we should discuss more about these and other r&d projects. I think i'll be attacking CCDIK next - my current IK approach is too analytical for my liking.
phoelix
11-05-2009, 01:36 PM
We have very similar studies Felix, we should discuss more about these and other r&d projects.

yeah sure, it's very interesting to hear what other people have achieved in this kind of things, sometimes its pretty hard to find useful information about this on the net, so... these forums become the right place to do it :D.

By the way what does CCDIK stand for? :blush:
eek
11-05-2009, 02:10 PM
yeah sure, it's very interesting to hear what other people have achieved in this kind of things, sometimes its pretty hard to find useful information about this on the net, so... these forums become the right place to do it :D.

By the way what does CCDIK stand for? :blush:

Cyclic-Coordinate Descent inverse kinematics.
scroll-lock
11-05-2009, 11:45 PM
I was just going to reply that Charles had wrote about verlet integration on his blog some time ago, but I saw you both already met :) The topic is also interesting to me but only as readings. Anyone read this paper btw?
http://www.teknikus.dk/tj/gdc2001.htm
phoelix
11-06-2009, 02:40 AM
that's one of the best articles on internet of verlets that i have read, it have a lot of topics about that subject and are very well explained, :D, thanks for the link Marin, I'll be checking eek's blog soon as well. It seems there is a lot more of information about this than i thought at the beginning of my experiments...
eek
11-06-2009, 03:00 AM
I was just going to reply that Charles had wrote about verlet integration on his blog some time ago, but I saw you both already met :) The topic is also interesting to me but only as readings. Anyone read this paper btw?
http://www.teknikus.dk/tj/gdc2001.htm
Thats the paper i pretty much used, but i also picked up on what joe stam described nucleus as - particles with a radius, constrained by rigid constraints. He also uses space-time math to calculate collision.

For what im doing which is basically a simple rigid framework, verlet is perfect - anything more complex id need better collision handling. The nice this is that basically from this idea you can do soft-bodies, rigid, cloth , particles, etc, etc

Im also interested mocap and did some r&d into rigid body solving for marker data - works pretty well.
CGTalk Moderation
11-06-2009, 03:00 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.