From owner-freebsd-chat@FreeBSD.ORG Sat Mar 13 22:06:38 2004 Return-Path: Delivered-To: freebsd-chat@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4772D16A4CE; Sat, 13 Mar 2004 22:06:38 -0800 (PST) Received: from cain.gsoft.com.au (cain.gsoft.com.au [203.31.81.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3589F43D1D; Sat, 13 Mar 2004 22:06:37 -0800 (PST) (envelope-from doconnor@gsoft.com.au) Received: from inchoate.gsoft.com.au (localhost [127.0.0.1]) (authenticated bits=0) by cain.gsoft.com.au (8.12.9/8.12.8) with ESMTP id i2E66WQ9063517; Sun, 14 Mar 2004 16:36:33 +1030 (CST) (envelope-from doconnor@gsoft.com.au) From: "Daniel O'Connor" To: David Schultz Date: Sun, 14 Mar 2004 16:36:29 +1030 User-Agent: KMail/1.6 References: <404FAC50.6070603@pythonemproject.com> <200403111155.39591.doconnor@gsoft.com.au> <20040314051128.GA57404@VARK.homeunix.com> In-Reply-To: <20040314051128.GA57404@VARK.homeunix.com> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200403141636.29686.doconnor@gsoft.com.au> X-Spam-Score: -1.5 () CARRIAGE_RETURNS,IN_REP_TO,QUOTED_EMAIL_TEXT,REFERENCES,SPAM_PHRASE_00_01,USER_AGENT X-Scanned-By: MIMEDefang 2.16 (www . roaringpenguin . com / mimedefang) cc: Rob cc: freebsd-chat@FreeBSD.ORG Subject: Re: What makes Centrino so fast? X-BeenThere: freebsd-chat@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Non technical items related to the community List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Mar 2004 06:06:38 -0000 On Sun, 14 Mar 2004 15:41, David Schultz wrote: > > but in short.. better branch prediction, and micro-architecture > > improvments in general, and a slightly longer pipeline (for higher clocks > > vs a PIII) > > You're right that the longer pipeline allows the processor to be > clocked higher, but for a *given* speed (e.g. 1700 MHz), a longer > pipeline is actyually a disadvantage; longer pipelines cause more > stalls and higher branch misprediction costs. The better branch > predition merely attempts to hide the penalty of the longer > pipeline. I don't know why the Centrino performs better than the > Athlon in this case, though. If you really care, you'll probably > have to factor the benchmark into specific, simple tests that > demonstrate the performance difference, play with compiler > optimizations, etc. I said the longer pipeline was for higher clocks, the improved branch prediction could have a benefit over and above the penalty imposed by a longer pipeline, but as always I would imagine it would depend on the type of code being run through it :) -- Daniel O'Connor software and network engineer for Genesis Software - http://www.gsoft.com.au "The nice thing about standards is that there are so many of them to choose from." -- Andrew Tanenbaum GPG Fingerprint - 9A8C 569F 685A D928 5140 AE4B 319B 41F4 5D17 FDD5