Optimizing RTX 4090 hashcat cracking performance

[OverTheTop]

Thanks - that's good to hear.

But why are the reported benchmark values "Accel:16 Loops:512 Thr:512" way higher than the one I experienced i.e. "Accel:8 Loops:256 Thr:256"?

I'm currently looking into the options:
-n, --kernel-accel            | Num  | Manual workload tuning, set outerloop step size to X | -n 64
-u, --kernel-loops            | Num  | Manual workload tuning, set innerloop step size to X | -u 256
However I think hashcat already does some sort of autotuning to figure out the best workload.
Edit: I found this post about the autotune engine https://hashcat.net/forum/thread-5184.html

Thanks