Moving collision generation to another cpu thread

Has anyone been able to move collision geometry generation to another thread? I’m making a voxel game, and the Unity collision generation is slow enough that I can’t go beyond 10 meters per second or the character will eventually fall off the world map. So I’m trying to figure out how to move the collision geometry to another thread exclusively to speed up the process. Is that even possible without the source code? Any help would be appreciated!

I wouldn’t think so(could be wrong), unless you created your own collision system in Unity that uses threads and you control. Threads get dicey when you don’t control the locking(or life time) of objects in the main thread of the application. If you are passing unity object types around in threads, but unity is modifying the objects(in the main thread) while your thread is trying to have it’s way with them, you get Threading exceptions’o plenty.

Is there anyway to optimize your voxel game?