From owner-freebsd-hackers Thu Aug 7 00:49:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA25659 for hackers-outgoing; Thu, 7 Aug 1997 00:49:43 -0700 (PDT) Received: from bugs.us.dell.com (bugs.us.dell.com [143.166.169.147]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id AAA25653 for ; Thu, 7 Aug 1997 00:49:40 -0700 (PDT) Received: from ant.us.dell.com (ant.us.dell.com [198.64.66.34]) by bugs.us.dell.com (8.6.12/8.6.12) with SMTP id CAA15262; Thu, 7 Aug 1997 02:48:02 -0500 Message-Id: <3.0.2.32.19970807023455.006c4904@bugs.us.dell.com> X-Sender: tony@bugs.us.dell.com X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Thu, 07 Aug 1997 02:34:55 -0500 To: Mark Mayo From: Tony Overfield Subject: Re: Pentium II? Cc: Curt Sampson , hackers@FreeBSD.ORG In-Reply-To: <19970806223748.63083@vinyl.quickweb.com> References: <3.0.2.32.19970806043249.006df3e4@bugs.us.dell.com> <3.0.2.32.19970803041915.006a69e4@bugs.us.dell.com> <3.0.2.32.19970806043249.006df3e4@bugs.us.dell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 10:37 PM 8/6/97 -0400, Mark Mayo wrote: >Just one question: why are you comparing the PPro with 256K cache >instead of the PPro 200/512 ?? Everybody I know that builds high >end PC server (an oxymoron I know..) used the 512KB version of >the PPro... And most run it at 233MHz... Because my point was only to show that the L1 is faster than the L2 cache, even on the PPro. The size of the L2 cache isn't important to that specific point. I did describe the data which indicated the L2 cache size, but only because I wanted to explain the rest of the numbers for the sake of completeness. The PPro 200/512 has only two numbers different. The 384KB and 512KB numbers are similar to the 256KB number instead of the 768KB number. I don't have the time or will to properly compare the PPro to the PII for high end PC servers. It's my *opinion* that the PII at 266 or 300 MHz will always outperform the PPro, unless you need more than two processors, in which case the PPro is currently the only way to get that. You might also look at: http://www.alde.com/speed.html to see how at least one CPU bound application performs on a wide variety of processors. - Tony