[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ffae08e2c2d5bc7b304a8d65b19bef7c@smtp.hushmail.com>
Date: Sun, 30 Sep 2012 03:22:58 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Any secret to use a salt_hash function?
On 30 Sep, 2012, at 2:55 , Claudio André <claudioandre.br@...il.com> wrote:
> Em 29-09-2012 21:37, magnum escreveu:
>>
>> Cool. Did the split kernel tests end up as expected?
>>
>> magnum
> On AMD only (for now).
>
> Tests end up better than expected, even with performance gain.
> Responsiveness is not that good if you are doing a real crack (at least on 6770).
>
> I stayed at bull for more than 2 hours (i suppose), and nothing bad happened. I even run AMD and NVIDIA together.
>
> The bad part is that TS is going to take longer. It used to take 4 minutes to finish, now i expect 10 minutes (default GWS is bigger now). Well, If you do an "export GWS=1024" and than run TS, you will take 1 minute. So, it might be considered "solvable".
That might be considerable. We could export GWS=1024 (or even lower) for all of Test Suite, for better speed. I'll try this at some point in spacetime.
magnum
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.
