Hey,
thanks for the links, looks very interesting. However, it’s still baffling to me that such approach is even neccessary in this case. This is litreally 500 simple meshes, all using same, debug material, that we’re talking about here. It’s not their logic, since they have no logic at all (attaching ss of the bp as a proof, children of this class don’t expand on the logic at all). I’m also attaching screenshot of the actual placed actor. Also, the bottleneck is clearly on GPU, if it would be Blueprint causing the issue, wouldn’t CPU time go up instead?
Additionally, as for the Instance static mesh, if I understand correctly it’s mostly to help with draw calls taking a lot of time and for me they don’t. On my main scene draw calls take < 0.5ms. If any additional info is needed to try and pinpoint the issue, please let me know :). Thanks for the answers so far.
