From owner-freebsd-chat  Mon Aug  5 21:54:08 1996
Return-Path: owner-chat
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id VAA22005
          for chat-outgoing; Mon, 5 Aug 1996 21:54:08 -0700 (PDT)
Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id VAA21999
          for <freebsd-chat@freebsd.org>; Mon, 5 Aug 1996 21:54:07 -0700 (PDT)
Received: by agora.rdrop.com (Smail3.1.29.1 #17)
	id m0une8u-0008w2C; Mon, 5 Aug 96 21:53 PDT
Message-Id: <m0une8u-0008w2C@agora.rdrop.com>
From: garyh@agora.rdrop.com (Gary Hanson)
Subject: Re: FreeBSD and Mersenne Primes
To: joelh@gnu.ai.mit.edu
Date: Mon, 5 Aug 1996 21:53:20 -0700 (PDT)
Cc: freebsd-chat@freebsd.org
In-Reply-To: <199608031941.PAA17213@kropotkin.gnu.ai.mit.edu> from "Joel Ray Holveck" at Aug 3, 96 03:41:38 pm
X-Mailer: ELM [version 2.4 PL24 ME8a]
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

>  > There was some data in the first announcement that I saw about the
>  > project (or perhaps on the web site) that *in this application* 4
>  > Pentium 133(?)s = one Cray.
> 
> Is this factoring in code optimized for a Cray?  On typical mundane
> code, the Cray is nothing spectacular; just for code that has been
> vectorized.

  Considering that the heart of the algorithm involves ffts, I'd have to
assume that it's vectorized. Quite accidently, I came across the original
source of my (slightly mangled) paraphrase above. For the dejanews fans
or the extremely anal, it was in article <4o36jd$7br@nntp1.best.com> by
Luke Welsh:

   In 1992, a Cray-2 took 19 hours to find that 2^756839-1 was prime.
   Today, a Pentium 166 takes only 21 hours...A Cray C90 took 7 hours
   to find 2^859433-1 in 1994, a Pentium takes just 24 hours. That
   means that 3 Pentiums almost equal one supercomputer, 4 will beat it.
 
 
  --Gary Hanson