qubit has slowed from 3609 kH/s on the previous version to 3571 kH/s on the current one - GTX 850M Linux, using "git pull" to update, build.sh has -O3 passed to configure
The fresh algo uses 2 rounds of shavite, 2 of simd and 1 echo.
shavite512
simd512
shavite512
simd512
echo512
The shavite and echo is affected in my AES change, that is faster on the compute 5.2 cards. I am rewriting the AES now, so the next commit will hopefully increase the performance of fresh.