0.80 multiGPU alpha (Alpha2 is out, do not use this thread)

Discussion and support for the CUDA Multiforcers (Windows and Linux)
  • Ads

Re: 0.80 multiGPU alpha - please review

Postby Bitweasil » Tue Nov 23, 2010 10:16 pm

blazer wrote:any planned speed improvements for later revisions? (would be awesome to have the features of the multiforcer with the speed of ighash) but beggars can' be choosers ;)


There are some in the works, yes. Any chance you could get me some ihasgpu MD5/NTLM numbers?

Right now, I'm still not hash reversing, so I'm not going to be competitive with a tool that does that. However, again, my new architecture makes it /much/ easier to do per-hash tweaks. :)

I also haven't fermi-tuned my code. There are some gains to be had there.
Bitweasil
Site Admin
 
Posts: 912
Joined: Tue Jan 20, 2009 4:26 pm

Re: 0.80 multiGPU alpha - please review

Postby hackajar » Wed Nov 24, 2010 2:57 am

hackajar
 
Posts: 30
Joined: Tue Sep 14, 2010 10:21 pm

Re: 0.80 multiGPU alpha - please review

Postby Bitweasil » Wed Nov 24, 2010 4:11 am

Got the hash framework for "plain" hashes sorted out. It took me about 15 minutes to add NTLM support from my old code. :)

Right around 2500M NTLM/sec on 2 470s.
Bitweasil
Site Admin
 
Posts: 912
Joined: Tue Jan 20, 2009 4:26 pm

Re: 0.80 multiGPU alpha - please review

Postby Bitweasil » Thu Nov 25, 2010 11:36 pm

blazer wrote:any planned speed improvements for later revisions? (would be awesome to have the features of the multiforcer with the speed of ighash) but beggars can' be choosers ;)


At some point, yes. For now, the focus is on getting the framework out and useful. Once that's in place, then I'll start adding hash algorithms and improving speed on existing ones. It'll be much easier to do things like fermi-specific tuning once this is fully fleshed out.
Bitweasil
Site Admin
 
Posts: 912
Joined: Tue Jan 20, 2009 4:26 pm

Previous

Return to CUDA Multiforcers

Who is online

Users browsing this forum: No registered users and 1 guest

cron