From owner-freebsd-hardware Mon Nov 3 08:57:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA14109 for hardware-outgoing; Mon, 3 Nov 1997 08:57:33 -0800 (PST) (envelope-from owner-freebsd-hardware) Received: from ns.mt.sri.com (SRI-56K-FR.mt.net [206.127.65.42]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA14099 for ; Mon, 3 Nov 1997 08:57:25 -0800 (PST) (envelope-from nate@rocky.mt.sri.com) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.7/8.8.7) with ESMTP id JAA17357; Mon, 3 Nov 1997 09:57:20 -0700 (MST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA07309; Mon, 3 Nov 1997 09:57:16 -0700 (MST) Date: Mon, 3 Nov 1997 09:57:16 -0700 (MST) Message-Id: <199711031657.JAA07309@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: "Michael L. VanLoon -- HeadCandy.com" Cc: asami@cs.berkeley.edu (Satoshi Asami), hardware@freebsd.org Subject: Re: fxp0 and full duplex In-Reply-To: <199711031000.CAA00941@MindBender.serv.net> References: <199711030544.VAA07343@bubble.didi.com> <199711031000.CAA00941@MindBender.serv.net> X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > * > Do you know, for a fact, that MS IP stacks (from Win95 thru NT Server) > > * > are significantly less efficient than the BSD variety? Or are you > > * > just slamming MS for the hell of it? [ Many stories deleted where FreeBSD outperformed Win95/NT on indentical hardware ] > >That is the speed of ftp transfer of large files from disk to disk. > >(Of course, this could be the filesystem and not the network driver, > >but it's just a single datapoint anyway.) > > I wouldn't be surprised if this is also at least partially to do with > "suckage" in the ftp client. The NT ftp and telnet clients have been > known to suck badly in many other ways. I never mentioned ftp in my statements, although I can't give 'real' #'s to back it up. Doing clock timing tests on an idle network from a Win95 box to a NT Server (4.0-latest patchlevel (3 or 4, don't remember), we're getting ~100K/minute using 'network neighberhood', or whatever M$ calls their network. Heck, NFS on the same hardware get's about 500-600K/sec. Nate