Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 20 Sep 2011 00:03:44 +0400
From: Solar Designer <>
Subject: Re: What does the dummy bench actually mean?

On Mon, Sep 19, 2011 at 09:13:08PM +0200, Pablo Fernandez wrote:
> How is the dummy benchmark calculated? I see the code, it seems simple, but 
> don't really understand it. 
> Does it actually correspond to the time it takes John to go to the next 
> candidate?

No.  It benchmarks dummy hashing, which includes overhead of the
"formats" interface and actual time it takes to compare candidate
passwords against known passwords (dummy hashes).

On a Pentium 3 1.0 GHz, I am getting 17M c/s at "dummy".
On a typical Core 2'ish machine, it reports a speed of about 90M c/s.
On Core i7-2600 3.4 GHz (Turbo Boost to 3.8 GHz), it goes up to 130M c/s.
Indeed, all of these are for one CPU core.

> My guess it that NO, that I am making a mistake here... so, is 
> there a way to measure the original time to "go-to-the-next-candidate" without 
> filters, so that I can check the efficiency of my filter?

No, but you can use a trivial filter() as your baseline:

void filter()
	word = 0;


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.