Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Tue, 21 Feb 2012 21:28:54 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: OpenCL vs CUDA

Is this a general difference you see too?

Benchmarking: CRYPTMD5-OPENCL [MD5-based CRYPT]... DONE
Raw:    147803 c/s real, 49267 c/s virtual

Benchmarking: cuda-cryptmd5 [MD5-based CRYPT]... DONE
Raw:    258048 c/s real, 258048 c/s virtual

(both formats use the exact same test vector)

Is this due to Lukas' version being better optimised at the moment, or
is it due to NVidia's OpenCL stuff not being up to par with CUDA?


For phpass, both versions run at almost exactly the same speed:

Benchmarking: PHPASS-OPENCL [PORTABLE-MD5]... DONE
Raw:    256984 c/s real, 129113 c/s virtual

Benchmarking: cuda-phpass [PORTABLE-MD5]... DONE
Raw:    252438 c/s real, 252438 c/s virtual

magnum

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.