# Separating Axis Theorem for moving circles

This topic is 4350 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hi, I'm developing small physics engine based on http://uk.geocities.com/olivier_rebellion/Polycolly.zip . Now I have completed part 3 from this tutorial for polygon-polygon and polygon-circle based on SAT, but when I tried the same method with circle-circle I end up with something like this: http://img206.imageshack.us/img206/179/problemrn8.png I'm testing only two axis - one conecting circle centers and one for velocity. Is there any way to make it to work properly for two circles? Maybe by trying some more axis? I would be greatful for any help.

##### Share on other sites
you can't really do that with circles. Imagine a circle as a polygon wioth lots of faces. Then you would end up doing hundreds of tests. There is an infinity of axes you will have to test. Instead, solve the equations between the two cricles, it's a lot easier and faster.

EDIT : God, I gotta fix that zip... There are much better algos for CD between polygons.

##### Share on other sites
Quote:
 Original post by oliiiEDIT : God, I gotta fix that zip... There are much better algos for CD between polygons.

Could you point me to some of it?

Quote:
 Original post by shurcoolSee if this is helpful.http://www.harveycartel.org/metanet/tutorials/tutorialA.html

It's for poly-circle, not for circle-circle.

1. 1
2. 2
3. 3
Rutin
22
4. 4
frob
16
5. 5

• 9
• 33
• 13
• 13
• 10
• ### Forum Statistics

• Total Topics
632579
• Total Posts
3007183

×