Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Tue, 30 Sep 2014 21:01:12 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: NVIDIA GTX 970 (Maxwell 2 / GM204) opencl benchmarks

On 2014-09-30 06:53, Royce Williams wrote:
> On Mon, Sep 29, 2014 at 10:07 AM, magnum <john.magnum@...hmail.com> wrote:
>> I'd much appreciate a benchmark before/after trying this change: (...)

> Looks to have not made much of a difference:
>
> $ ../run/john --test --format=wpapsk-opencl
> Device 0: GeForce GTX 970
> Benchmarking: wpapsk-opencl, WPA/WPA2 PSK [PBKDF2-SHA1 OpenCL]... DONE
> Raw:    134663 c/s real, 135591 c/s virtual

Great, so we don't seem to need any kernel changes for that. But sooner 
or later we'll find Maxwell-specific optimizations.

So using current code, GTX970 is basically on par with HD7970 which 
means GTX980 is on par with R9 290X. Hashcat forums tell the same story.

>  From the tests, ntlmv2-opencl, pwsafe-opencl and rakp-opencel seem a bit
> better.  Others look about the same.

All three of those currently have transfer bottlenecks, so this was 
probably coincidental.

Thank you for testing this!
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.