LM CUDA 22.33M/s Somethings not right!

[middle-siggy]

Does this make sense...

I was doing some CUDA bruteforce cracking on NTLM hashes the other day, and was getting 100-110M/s hashes when I was doing that,
But for sake of argument I wanted to try LM, and see what it would do, and it only seems to be soing 22-23M/s....

That does not make sense... Thoughts?

I'm running on a NVIDIA GeForce GT 520

Siggy

[Bitweasil]

If you're seeing 100M NTLM per second, 22-23M LM/sec is perfectly sane. LM uses DES, which is a much more complex algorithm than the MD4 that NTLM uses.

For len6, I get 38.7 NTLM/sec on my Ion2 Netbook, and 4.88M LM/sec.

There's nothing wrong with the code - it's just algorithm complexity differences.

[middle-siggy]

Good to know, Thanks...