17:03:23 <m-relay> <x​fedex:matrix.org> Does the branch predictor in Intel/AMD processors slow down RandomX mining? Could it be optimized in a specialized CPU?