Christoph Hormann wrote:

> Do you think recalculating a bounding tree after every simulation step
> will be faster in the end?
> 
> It would be possible to speed up the triangle function a bit since we
> don't actually need its value if it exceeds the mass radius of course.
> 
> To speed up collision calculations the mechsim patch has the possibility
> to group masses and only enable collisions between masses of different
> groups.
> 
> Christoph
> 

Yes, I think bounding would speed up the hole thing. Building a bounding box
or sphere tree is too slow, but using a hash function is fast enough.

Daniel

Post a reply to this message