Date: Wed, 15 Mar 2000 16:04:33 -0500 (EST) From: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> To: Mike Smith <msmith@FreeBSD.ORG> Cc: freebsd-current@FreeBSD.ORG Subject: Re: Best NIC for FBSD (was: Buffer Problems and hangs in 4.0-CURRENT..) Message-ID: <200003152104.QAA59565@khavrinen.lcs.mit.edu> In-Reply-To: <200003152053.MAA01346@mass.cdrom.com> References: <200003152017.MAA19573@www.redshift.com> <200003152053.MAA01346@mass.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
<<On Wed, 15 Mar 2000 12:53:43 -0800, Mike Smith <msmith@FreeBSD.ORG> said: > 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. As of about four years ago, the relative performance of the network interfaces was fairly clear: 1) Intel 2) DEC (now also Intel) 3) vendor not disclosed to me 4) 3Com The reason why Intel's NIC did better in these tests was that their PCI interfaces were comparatively very efficient. Since every major vendor has gone through several revs of silicon, there is some reason to believe that quality vendors' NICs now differ only in the inherent quality of their programming model. (So these old results should mean less today.) I always specify Intel NICs today, unless I don't care about performance (in which case I get a $20 RealTek card). There's probably no reason to prefer it over the current 3Com offering (for example) except that 3Com always seems to cost more. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same wollman@lcs.mit.edu | O Siem / The fires of freedom Opinions not those of| Dance in the burning flame MIT, LCS, CRS, or NSA| - Susan Aglukark and Chad Irschick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200003152104.QAA59565>