Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Wed, 12 Dec 2012 23:59:30 +0400
From: Solar Designer <>
Subject: office2013-opencl

magnum -

More stuff from Twitter:

zn7> <%solardiz> RT @thorsheim: .@...sware updates once again: GPU cracking for office 2013 (700 c/s) files & PGP passwords and more:
zn8> <%solardiz> RT @gat3way: @passware @thorsheim @chilikov tested on @solardiz's bull. Speed (gtx570 oc) is ~680c/s, speed on 7970 is ~710c/s
zn9> <@...ardiz> @gat3way In other words, both @passware's proprietary software and your free software (hashkill) achieves ~700 c/s per GPU at Office 2013

zo2> <%solardiz> RT @gat3way: @solardiz @passware Looks like. I am impressed how close 570 is to 7970. 580/680 could be even faster than Tahiti...
zo3> <@...ardiz> @gat3way @passware NVIDIA cards are in general surprisingly good at SHA-512 - or rather ATI/AMD are surprisingly bad at it
zo4> <@...ardiz> @gat3way @passware Anyhow, I just benchmarked JtR bleeding-jumbo's office2013-opencl on the two cards in bull ...
zo5> <@...ardiz> @gat3way @passware ... I got 261 c/s on the GTX 570 o/c (need to ask magnum to optimize more!), but a whopping 911 c/s on 7970
zo6> <@...ardiz> @gat3way @passware Yeah, per @hashcat's benchmarks 6990 is faster than 7970 per-GPU at iterated SHA-512 - but all are close to GTX 580
zo7> <@...ardiz> @gat3way @passware Split kernel. Looks like as few as 64 iterations per kernel invocation, until the 100k iterations are reached.

zp0> <@...ardiz> @thorsheim With deliberate anti-GPU design (not what we see with Office 2013), it could as well be < 10 c/s per GPU for encrypted documents
zp1> <@...ardiz> @gat3way Increased iterations per kernel invocation to 1000, got 915 c/s on 7970 (and two degrees hotter GPU, hmm)
zp2> <solardiz> BTW, this is the original 7970 at 925 MHz stock clock. Not a 7970 GE, nor overclocked. It could definitely run at a higher clock fine.

zp9> <@...ardiz> @gat3way @thorsheim OK, 7970 at 1100 MHz to match @jmgosney's, 1000 iterations per kernel, 1087 c/s at Office 2013, 80C after 8 min
zq0> <solardiz> 925 MHz, 64 iterations/kernel was 76C; 925 MHz, 1000 iterations was 78C; 1100 MHz, 1000 iterations is 80C

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.

Observation: it takes maybe a couple of minutes for --test of this
format to complete on GTX 570, but about 8 minutes on 7970.  Why?


Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ