CUDA update broke my throughput

[cobaltspark]

worth checking the driver first.

If you post your exact settings, people here can probably narrow it down pretty fast.

[thetapath]

worth checking the driver first.

[oxidemixer]

That matches what I was seeing. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[kernelrunner]

Good point. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

If you post your exact settings, people here can probably narrow it down pretty fast.