#### Archived

This topic is now archived and is closed to further replies.

# OBB - OBB Tree Stumps

This topic is 5549 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hey, Pardon the pun about the stumps. But I am stumped. How does one go about finding the point of collision of two (moving) OBB (oriented bounding box)s? Also, finding the normal of the collision (if it exists... I probably need to find the normal relative to each box, right?) is a problem. Walt

*bump*

##### Share on other sites
From what I''ve seen in my searches for this is that brute force is the way to go (all points, edges, planes from one box against all from the other). You can quickly reduce the number of comparisons to only select point/eges/planes.

There is the seperating axis algorithm but I think this will just add extra overhead because you would still need to compute a lot of the brute force version to get collision information.

more information can be found here http://www.realtimerendering.com/int/. Check out the Magic Software algorithm. I have seen their box-box algorithm used elsewhere with very good results and it has some early exits as well.

##### Share on other sites
Yes, I''m using magic software''s box->box col detection routine (slighty modified by yours truly though). But the problem is it doesn''t provide a point of contact, which is a large problem when you''re trying to get proper collision response .

*sigh* I guess I''ll have to bust out the ''ol pen and paper and try to crank out an accurate algo for finding a point of collision.

Thanks
Walt

##### Share on other sites
hmmm didn''t know that. There is a nice thread on flipcode with a lot of info I just found though, I guess you can get all the info you need from seperating axes.

Hope this helps (I need to implement this soon too *sigh*).

##### Share on other sites
quote:
Original post by Kudaba
hmmm didn''t know that. There is a nice thread on flipcode with a lot of info I just found though, I guess you can get all the info you need from seperating axes.

Hope this helps (I need to implement this soon too *sigh*).

lol... look at the poster of that thread .

1. 1
Rutin
47
2. 2
3. 3
4. 4
5. 5

• 13
• 10
• 12
• 10
• 13
• ### Forum Statistics

• Total Topics
632993
• Total Posts
3009758
• ### Who's Online (See full list)

There are no registered users currently online

×