Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 11 Jul 2012 13:42:59 +0200
From: Frank Dittrich <>
Subject: Re: our own training pseudo contest before CMIYC 2012

On 07/10/2012 06:19 PM, Rich Rumble wrote:
> We may also want to "inventory" the "horsepower" that's
> available so that someone with a really fast GPU may take on a harder
> challenge than another person without a GPU. I think we can have a
> better showing if we use our resources a bit better.

You can even have different john builds on the same machine.
E.g., compiled with different compilers, because a few formats which are
not yet optimizes for speed are reasonably faster using a clang build,
compared to a gcc build.
Another example would be one build with OMP support, one without.
(For OMP builds, even the OMP_MAX_THREADS setting can make a difference.)

To get an overview, we could either collect "john --test" output files,
or we could collect information from "john --list=build-info", "john
--list=formats", "john --list=cuda-devices", "john --list=opencl-devices".

If we get huge input files during the contest, even available memory is
an important information to have.


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.