From owner-freebsd-chat  Fri Feb 28 20:00:44 1997
Return-Path: <owner-chat>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id UAA07489
          for chat-outgoing; Fri, 28 Feb 1997 20:00:44 -0800 (PST)
Received: from r74h25.res.gatech.edu (ken@r74h25.res.gatech.edu [128.61.74.25])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA07484
          for <freebsd-chat@freebsd.org>; Fri, 28 Feb 1997 20:00:41 -0800 (PST)
Received: (from ken@localhost)
          by r74h25.res.gatech.edu (8.8.5/8.8.5) id XAA01541;
          Fri, 28 Feb 1997 23:00:08 -0500 (EST)
From: "Kenneth D. Merry" <ken@r74h25.res.gatech.edu>
Message-Id: <199703010400.XAA01541@r74h25.res.gatech.edu>
Subject: Re: RSA 56-bit key challenge
In-Reply-To: <199703010123.RAA12036@rah.star-gate.com> from Amancio Hasty at "Feb 28, 97 05:23:30 pm"
To: hasty@rah.star-gate.com (Amancio Hasty)
Date: Fri, 28 Feb 1997 23:00:07 -0500 (EST)
Cc: taob@vex.net, freebsd-chat@freebsd.org
X-Mailer: ELM [version 2.4ME+ PL25 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-chat@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


> I ran this test on my PPRO 200Mhz 3.0-current as of a couple of weeks
> ago.
> 
> Got this :
> 
> {hasty} ./rc5-client-freebsd -m
> rc5-56-client: Performance testing with 1000000 crypts
> rc5-56-client: Complete in 3.347 seconds. (298812.07 keys/sec)
> 
> The result looks to fast for me 8)
> 
> So the next question am I am running the reference benchmark correctly?

	I think that is correct.  Here's what I got, dual PPRO 200, 256K
cache.  (so I assume I could get about twice this if I ran two copies)

{r74h118:/usr/home/ken/dl:23:0} uname -a
FreeBSD r74h118.res.gatech.edu 3.0-SMP FreeBSD 3.0-SMP #0: Mon Jan 27 01:27:38 EST 1997 ken@r74h118.res.gatech.edu:/usr/src/sys-SMP/compile/panzer  i386
{r74h118:/usr/home/ken/dl:24:0} ./rc5-client-freebsd -m
rc5-56-client: Performance testing with 1000000 crypts
rc5-56-client: Complete in 3.300 seconds. (303063.36 keys/sec)


> >From The Desk Of Brian Tao :
> >     Anyone else dedicating spare cycles (or whole machines) to this
> > effort?  I've got a 200-MHz Pentium cranking out just over 148000 keys
> > per second, which is pretty fast compared to numbers posted for
> > various Suns, SGI's and RS/6000's (see http://www.vex.net/~rasmus/rsa/
> > for our particular team effort).  Linux systems seem to be 15% to 30%
> > faster on similar hardware though.  :(  Does anyone have numbers for a
> > dual-CPU system or a fast PPro system?
> > 
> > http://www.rsa.com/rsalabs/97challenge/
> > http://zero.genx.net/
> > 
> > --
> > Brian Tao (BT300, taob@risc.org)
> > "Though this be madness, yet there is method in't"



Ken
-- 
Kenneth Merry
ken@r74h25.res.gatech.edu
Disclaimer:  I don't speak for GTRI, GT, or Elvis.