From owner-freebsd-current Wed Mar 15 13: 4:18 2000 Delivered-To: freebsd-current@freebsd.org Received: from outgoing.redshift.com (outgoing.redshift.com [216.228.2.99]) by hub.freebsd.org (Postfix) with ESMTP id 2BA9837BABF; Wed, 15 Mar 2000 13:04:10 -0800 (PST) (envelope-from nobody@redshift.com) Received: from RedShift.com (mail.redshift.com [216.228.2.86]) by outgoing.redshift.com (8.9.3/) with ESMTP id NAA10453; Wed, 15 Mar 2000 13:04:09 -0800 Received: from www.redshift.com (www.redshift.com [216.228.2.89]) by RedShift.com (8.9.3/) with ESMTP id NAA28081; Wed, 15 Mar 2000 13:04:07 -0800 Received: by www.redshift.com (8.8.5) id NAA20079; Wed, 15 Mar 2000 13:04:06 -0800 Date: Wed, 15 Mar 2000 13:04:06 -0800 Message-Id: <200003152104.NAA20079@www.redshift.com> To: "Mike Smith " , freebsd-current@freebsd.org From: "yramin" Reply-To: "yramin" Subject: Re: Re: Best NIC for FBSD (was: Buffer Problems and hangs in 4.0-CURRENT..) X-Mail-Gateway: WebMail X-Real-Host-From: 131.120.8.59 Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="-----20078-----WWW-Mail-2.01-----" Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG This message is in MIME format. -------20078-----WWW-Mail-2.01----- Content-Type: text/plain Content-Transfer-Encoding: 8bit Mike, These are my own perosnal findings with machines I've tinkered with. Nothing scientific about them. I did a few trials FTPing a 600MB file around on a 100Mb switched network between to machines (they were not identical, but used the same ones for both tests), once with Intel PRO/100 adapters, and once with 3com parralel tasking (forgot the model number) adapters. I watched top both times, and noticed that the Intels had lower interupt %s, and finished the job about 5-30 seconds faster (varied on each run). Both machines used wu-ftpd and were running FreeBSD 3.2 - R at the time. > > fxp0: The Intel driver is by far the highest preformance model, > > beats the 3com (second best) hands down with much lower CPU > > overhead. > > Do you actually have any numbers to quantify this? There's nothing in > the driver architecture nor any of my testing that would suggest this is > actually the case at this point. > -------20078-----WWW-Mail-2.01----- -------20078-----WWW-Mail-2.01------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message