[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20121212234609.GA10650@openwall.com>
Date: Thu, 13 Dec 2012 03:46:09 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: office2013-opencl
On Wed, Dec 12, 2012 at 10:08:55PM +0100, magnum wrote:
> Cool! I just took Claudio's SHA-512 as base and optimised it a little for iterated use. Actually I think he has a separate kernel for nvidia that I do not yet use. The format is still just first PoC. So much to do, so little time...
Curious. Claudio's sha512crypt-opencl code actually performs very well
on NVIDIA (yes, separate kernel), but not so well (compared to other
tools) on ATI/AMD, including on the 7970. In this context, it's a bit
puzzling that it outperformed hashkill on the 7970 at Office 2013.
Perhaps there's something unoptimal in Claudio's code specific to
sha512crypt, which thus did not make it into office2013-opencl.
> On 12 Dec, 2012, at 20:59 , Solar Designer <solar@...nwall.com> wrote:
> > Also tried 1225 MHz (which previously worked fine for bf-opencl) - got
> > ASIC hang after a few minutes, at 82C (although the card can be stable
> > at higher temperatures - previously observed up to 88C with other
> > formats). Perhaps some parts of the chip, which bf-opencl did not use
> > actively enough, become unstable at 1225 MHz.
>
> Was that with iterations/kernel increased?
Yes.
> Maybe you went over that 200ms "death threshold" for each kernel call.
Actually, probably yes. There was no ASIC hang at lower clock rates,
but this could be a mere coincidence.
Alexander
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.
