Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1334233996.83186.YahooMailNeo@web161302.mail.bf1.yahoo.com>
Date: Thu, 12 Apr 2012 05:33:16 -0700 (PDT)
From: deepika dutta <deepikadutta_19@...oo.com>
To: "john-dev@...ts.openwall.com" <john-dev@...ts.openwall.com>
Subject: Re: bitslice des

Hi, I was doing speed test between openssl des and john des. I get following statistics for openssl

type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
des cbc         100225.76k    89521.76k    89778.20k    95060.70k    96158.84k


and for john 


Benchmarking: Traditional DES [32/32 BS]... DONE
Many salts:    434566 c/s real, 997527 c/s virtual
Only one salt:    426208 c/s real, 568277 c/s virtual


Benchmarking: LM DES [32/32 BS]... DONE
Raw:    9306K c/s real, 12086K c/s virtual

Now considering openssl, it can process 100225.76 x 1000 = 100225760 bytes/sec which should account to 100225760 /8 = 12528220 encryptions/sec (since DES block size is 8 bytes)

With john, considering LM DES (which according to what I read does 2 DES encryption), the result is  9306 x 1000 = 9306000 x 2 = 18612000 encryption/sec

This provided 1.48 times speedup with john des (non sse or other optimizations). Am I right in my calculation?

Deepika



________________________________
 From: Simon Marechal <simon@...quise.net>
To: john-dev@...ts.openwall.com 
Sent: Wednesday, April 11, 2012 3:42 PM
Subject: Re: [john-dev] bitslice des
 
On 11/04/2012 10:57, Deepika Dutta Mishra wrote:
> I read the mailing lists, I will take up one of the algorithm soon.
> Right now I am going through the DES BS code. Have you compared your
> code to openssl implementation (which is not bit sliced)? How much speed
> up has it given? I ran the test on my system, for DES I have following
> result. I could not understand the meaning of the result, I mean how you
> calculate c/s, what is real/virtual and how to relate c/s to
> encryptions/sec.
> 
> Benchmarking: Traditional DES [32/32 BS]... DONE
> Many salts:    439104 c/s real, 1514K c/s virtual
> Only one salt:    433184 c/s real, 1353K c/s virtual

Real refers to how many checks/seconds were computed during the time the
benchmark was run. Virtual is some calculation on how much it could be
if JtR was able to seize all CPU resources. When in doubt you will want
to run this on an idle system and only trust the "real" value.

A check uses 25 DES encryptions. You can also check the LM format.
Content of type "text/html" skipped

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.