Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 29 Feb 2012 21:42:46 +0100
CC: Lukas Odzioba <>
Subject: Re: rule: >NN , format=cuda-mscash, filters, --make-charset

> First of all you should not use mpi with cuda. For now it will not give you
> any speedup. If you have got 8 gpus try -gpu option (you can select only
> one for now) and for example different type of attacks for each gpu, or
> generate candidates with --stdout option, divide it for 8 parts and then
> try each part for one gpu + rules.

Actually my intention was to use 8 CPU cores + GPU, but after reading
your reply I assume that --format=cuda-* uses GPU power only.

> There was bug in cuda-1 version I fixed it already, and now I am trying to
> upload it to wiki (I am having problems with ISP), so it will be available
> on wiki ASAP.

Great, I'll try it.


Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.