taco here's a report for you.
First, thanks for the Windoze binaries.
The stock app nets me 1.76k (max) hashes per second (per core).
Your AMD SSE4 binary nets me 2.04k hashes per second (per core).
Tried the Intel SSE3 one for the hell of it, nah it won't run.
But hey, it's an improvement for sure. Thanks Taco.
AMD Athlon II (3.1GHz @ 3.6GHz) 4GB DDR3 @ 1600MHz (9-9-9-24), CPU to NB 2.4GHz.