From owner-freebsd-net@FreeBSD.ORG Fri Dec 23 22:13:50 2005 Return-Path: X-Original-To: freebsd-net@freebsd.org Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7301F16A41F for ; Fri, 23 Dec 2005 22:13:50 +0000 (GMT) (envelope-from cswiger@mac.com) Received: from pi.codefab.com (pi.codefab.com [199.103.21.227]) by mx1.FreeBSD.org (Postfix) with ESMTP id 029A943D46 for ; Fri, 23 Dec 2005 22:13:49 +0000 (GMT) (envelope-from cswiger@mac.com) Received: from localhost (localhost [127.0.0.1]) by pi.codefab.com (Postfix) with ESMTP id 246AB5DD5; Fri, 23 Dec 2005 17:13:47 -0500 (EST) Received: from pi.codefab.com ([127.0.0.1]) by localhost (pi.codefab.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 37007-04; Fri, 23 Dec 2005 17:13:45 -0500 (EST) Received: from [199.103.21.238] (pan.codefab.com [199.103.21.238]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by pi.codefab.com (Postfix) with ESMTP id E3FAD5C56; Fri, 23 Dec 2005 17:13:45 -0500 (EST) In-Reply-To: <20051223214718.GA55809@il.fontys.nl> References: <20051223193237.GA80590@wrongcrowd.com> <20051223214718.GA55809@il.fontys.nl> Mime-Version: 1.0 (Apple Message framework v746.2) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <5B2BFA0C-B1D5-45A5-AC9C-74A29BCD3113@mac.com> Content-Transfer-Encoding: 7bit From: Charles Swiger Date: Fri, 23 Dec 2005 17:13:45 -0500 To: Rink Springer X-Mailer: Apple Mail (2.746.2) X-Virus-Scanned: amavisd-new at codefab.com Cc: freebsd-net@freebsd.org, Matt Staroscik Subject: Re: Good gigabit NIC for 4.11? X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 23 Dec 2005 22:13:50 -0000 On Dec 23, 2005, at 4:47 PM, Rink Springer wrote: >> However, there is little point to trying to use GB and jumbo frames >> on a NIC in a standard 33MHz PCI slot; unless you have PCI Express >> slots available or a GB card integrated with the chipset, the PCI bus >> will bottleneck the system from doing much better than a 100Mbs NIC >> would perform... > > Well, most desktop boards with a gigabit card don't have very good > performance; I wasn't able to get any decent performance out of a sk > (4) > on an ASUS P4P800-E Deluxe motherboard anyway (Even an em(4) in a > 32bit > PCI slot did much better). I've had very decent results with 64bit > PCI-X, > however, which sk(4)'s, ti(4)'s and em(4)'s. It all depends on the details. :-) Any decent server motherboard which supports PCI-X ought to have the backplane bus to handle that kinda bandwidth, whereas a consumer grade motherboard where the GB NIC is hanging off the PCI bus rather than being integrated into the northbridge is not going to do nearly so well. The onboard em NICs in the Dell PE28x0's and whatever is in the HP DL370/380's (bge's) seem to work well... -- -Chuck