From owner-freebsd-hackers Wed Mar 6 14:06:46 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id OAA29537 for hackers-outgoing; Wed, 6 Mar 1996 14:06:46 -0800 (PST) Received: from etinc.com (etinc.com [204.141.244.98]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id OAA29515 for ; Wed, 6 Mar 1996 14:06:38 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.6.12/8.6.9) with SMTP id RAA01997; Wed, 6 Mar 1996 17:06:27 -0500 Date: Wed, 6 Mar 1996 17:06:27 -0500 Message-Id: <199603062206.RAA01997@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Version 2.0.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: Alexey Pialkin From: dennis@etinc.com (dennis) Subject: Re: Linux vs FreeBSD comparison - it's time, I think! Cc: hackers@freebsd.org Sender: owner-hackers@freebsd.org Precedence: bulk >> >Sounds fine to me, but I think we should make an effort to be absolutely >> >fair and to show some integrity. All numbers should be verified with >> >2 or more people, and pointers to the results should be maintained. If we >> >get a rep for playing fast and loose with the facts then we have a >> >problem. If we get a rep for being honest and being willing to give linux >> >its due (and it does have its good points) then we get to be an honest >> >broker. >> >> Its very difficult to be objective with benchmarks because they are so >> highly dependent on a particular piece of hardware or driver. For example >> whenever someone mentions FreeBSD vs BSDI some banana pulls >> out a benchmark with dual buslogic EISA adapters where BSDI may >> be superior, when the same test with an Adaptec PCI controller >> might be completely the opposite. Its almost impossible to have a >> single benchmark comparison for unix systems. You need to chose >> several "reasonable" systems with different hardware and show several >> benchmarks with each..... > >Why ? Are you sure ? >Let's take some system(some P5/16mb/1Gb ... and so on :) and make some >tests on it.. For example - compiling,news processing,ftp,www.... >We'll install Linux & FreeBS on the same system, so i think compasion will >be rather objective. >Any problem ? then things that you selected aren't the issue. But if you do network testing with a card that has a particularly bad LINUX driver and a very good FreeBSD driver, then the test is only valid for that one particular card. The Linux people could do the same test with a very good Linux driver and a buggy FreeBSD driver and get opposite results. So what have you shown? For example BSDI with an Adaptec SCSI and an NE2000 card is not a very good system, but with a buslogic VLB or EISA card (they've done a lot of work with these cards) and an SMC ethernet they're very good. dennis ---------------------------------------------------------------------------- Emerging Technologies, Inc. http://www.etinc.com Synchronous Communications Cards and Routers For Discriminating Tastes. 56k to T1 and beyond. Frame Relay, PPP, HDLC, and X.25 for BSD/OS, FreeBSD and LINUX