Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 18 Nov 2012 12:54:15 +0100
From: buawig <>
Subject: Re: cracking passwords with a kerberos traffic dump
 [implementation confirmed to work]

>> To add some figures for how big the slowdown actually is on my side:
>> - 390 c/s (faster implementation with pw length limitation)
>> - 220 c/s (>16 password length support PBKDF2 implementation)
> Don't forget that you can use all cores by enabling OMP in the Makefile.

Yes the above numbers are for a single core. With OMP (8x) I get
actually even lower performance 166 c/s for the faster PBKDF2 impl.,
but with MPI (8x) I reach 1390 c/s.

> Just wait for magnum to do his OpenCL magic ;)

Yes looking forward to it and I'm having some more ideas to reduce the
effort to crack etype 18 which I will send in a separate email (new thread).

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.