How to improve performance? #307
patrikpatrik
started this conversation in
General
Replies: 1 comment 5 replies
-
First, make sure that you are compiling the library in Release mode and not in Debug mode. If you want to obtain more performance, you can try to decrease the number of iterations of the velocity and position solver (as described here). Note that it will decrease the quality of the physics simulation. In the future, I will make possible to use multi-threading to run the physics simulation in parallel on several CPUs but this is not possible at the moment. |
Beta Was this translation helpful? Give feedback.
5 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello again Daniel,
Just curious, is there any way to increase performance ? For instance is there a way to send all this to the compute shader for the collision detection & resolution to run parallel?
I've been doing some benchmark testing and came to the conclusion that I can't have any more than 200-300 objects without slowing down considerably. This was tested with the default cube, default sphere, default capsule, a custom convex cube, custom convex 3-sided equilateral triangle, and a 4-sided pyramid. Also debug lines and triangles set to
off
and sleepenabled
.My hypothesis was that the 3-sided equilateral triangle with 4 vertices and only 4 sides would be less calculation intensive, but that wasn't the case at all.
Beta Was this translation helpful? Give feedback.
All reactions