Hey everyone,
I’m currently struggling to optimize a few RenderMeshIndirect calls.
These instanced draw calls already reduce performance impact but I saw that we’re currently struggling with ResourceBinds per draw call.
While I was able to reduce most of them I wanted to see if I can get rid of the unnecessary resource binds for the ComputeBuffer. I combined them into one bigger one and applied it to the material directly.
So every one of these RenderMeshIndirect calls share the same shader, material, and buffer but they still call two resource bind API calls for every draw call even though the buffer doesn’t differ at all between them:
Is there a way to get rid of this?
Many thanks in advance