From owner-freebsd-alpha Tue May 14 23:17:19 2002 Delivered-To: freebsd-alpha@freebsd.org Received: from alcanet.com.au (mail2.alcanet.com.au [203.62.196.17]) by hub.freebsd.org (Postfix) with ESMTP id B830037B405 for ; Tue, 14 May 2002 23:17:13 -0700 (PDT) Received: from mfg1.cim.alcatel.com.au (localhost.localdomain [127.0.0.1]) by alcanet.com.au (8.12.1/8.12.1/Alcanet1.2) with ESMTP id g4F6GX6u003793; Wed, 15 May 2002 16:16:34 +1000 Received: from gsmx07.alcatel.com.au by cim.alcatel.com.au (PMDF V5.2-32 #37640) with ESMTP id <01KHRH2GQTBK8X307U@cim.alcatel.com.au>; Wed, 15 May 2002 16:16:29 +1000 Received: from gsmx07.alcatel.com.au (localhost [127.0.0.1]) by gsmx07.alcatel.com.au (8.12.3/8.12.3) with ESMTP id g4F6GR01067385; Wed, 15 May 2002 16:16:27 +1000 Received: (from jeremyp@localhost) by gsmx07.alcatel.com.au (8.12.3/8.12.3/Submit) id g4F6GR58067384; Wed, 15 May 2002 16:16:27 +1000 (EST) Content-return: prohibited Date: Wed, 15 May 2002 16:16:27 +1000 From: Peter Jeremy Subject: Re: Alpha CPU Performance vs i386 In-reply-to: <20020514222344.A2915@freebie.xs4all.nl>; from wkb@freebie.xs4all.nl on Tue, May 14, 2002 at 10:23:44PM +0200 To: Wilko Bulte , Gorm Jorgensen Cc: alpha@FreeBSD.ORG Mail-Followup-To: Wilko Bulte , Gorm Jorgensen , alpha@FreeBSD.ORG Message-id: <20020515161627.O18023@gsmx07.alcatel.com.au> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-disposition: inline User-Agent: Mutt/1.2.5.1i References: <20020514201923.GK37326@Area51.DK> <20020514222344.A2915@freebie.xs4all.nl> X-Authentication-warning: gsmx07.alcatel.com.au: jeremyp set sender to peter.jeremy@alcatel.com.au using -f Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On 2002-May-14 22:23:44 +0200, Wilko Bulte wrote: >On Tue, May 14, 2002 at 10:19:23PM +0200, Gorm Jorgensen wrote: >> >> I can't seem to find any specs on how an Alpha 500 MHz is performing vs an >> Intel 500MHz. BTW, you can't just say "Alpha 500 MHz". An EV6/500 in an ES40 is roughly twice as fast as an EV5/466 in an AS4100. >> Is Alpha faster, slower or generally the same ? > >FP faster, integer is ?? Note that correctly supporting IEEE FP will significantly hit Alpha performance - maybe 20-30% slower, _much_ worse if you have lots of non-finite numbers (denormals, infinities and NaNs). The hit should be slightly less on the latest AXP chips which support precise FP exceptions. My gut feeling is that an AS4100 EV5/466 is roughly on a par with a PII-266 integer-wise. At the low end, a Multia (LCA/166) is slightly better than a 486DX2-50 on worldstones. Since I'm currently waiting for it to finish - ssh between an ES40 EV6/500 and a PIII-600 suggests that the PIII-600 is around 50% faster at running blowfish. [In the last case, ssh and ssl were compiled using Compaq ccc with '-tune host -arch host' on the Alpha]. As someone else noted, memory and I/O bandwidth on most Alphas is far better than PCs. (The Multia and other LCA machines are probably the only exceptions). Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message