From owner-freebsd-hackers Wed Jul 30 02:03:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id CAA20363 for hackers-outgoing; Wed, 30 Jul 1997 02:03:08 -0700 (PDT) Received: from sparkie.gnofn.org (sparkie.gnofn.org [206.27.168.35]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA20352 for ; Wed, 30 Jul 1997 02:03:05 -0700 (PDT) Received: from sparkie.gnofn.org (sparkie.gnofn.org [206.27.168.35]) by sparkie.gnofn.org (8.7.Beta.10/8.7.Beta.10) with SMTP id EAA02602; Wed, 30 Jul 1997 04:02:39 -0500 (CDT) Date: Wed, 30 Jul 1997 04:02:39 -0500 (CDT) From: Craig Johnston To: Tom Samplonius cc: FreeBSD Mailing List , Rod Ebrahimi , hackers@FreeBSD.ORG Subject: Re: Pentium II? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, 29 Jul 1997, Tom Samplonius wrote: > > On Tue, 29 Jul 1997, Craig Johnston wrote: > > > > > The PII runs 16-bit software better and adds MMX extensions, but for a > > > > network server the PPro will still be faster. > > > > > > The PII/266 will be faster than a PPro/200. > > > > Not buying into Intel's slot 1 ploy is a good enough reason not to run > > PII's. Slot 1 is not going to be around very long and I wouldn't count > > on not running into bugs in the relatively untested slot 1 chipset. > > If you use that arguement, you shouldn't by anything then! The PPro Um, no. There's this idea of relative quantities, you see? Slot 1 is going to be short-lived _compared_ to other solutions. How long has socket 7 been ticking along now? It's well known that slot 2 is coming right on the heels of slot 1. Why bother with the Intel "chipset of the week" game? Slot 1 systems don't offer anything you can't do better with the same cash with PPro and socket 7 systems. The only folks likely to benefit from slot 1 at all are likely to be Intel -- they'll make a bundle off upgrade-happy power-lusers who've been exposed to one too many Ziff-Davis magazines. > (socket 8?) is doomed too, as Intel will not be developing it further (a > 233mhz version would be nice). The Pentium MMX is going to max at 233, > before being put out to pasture. So, every option is "doomed"... The real issue is that the PII is untried, whereas FreeBSD systems on PPro hardware, like ftp.cdrom.com (which serves 2000 simultaneous ftp users) have been ticking along reliably for quite some time now. > > The PPro 200 offers all the CPU horsepower you're going to need on > > a FreeBSD network server. I'd worry about the amount of RAM and > > the speed, latency, and number of hard drives. SCSI, of course. > > Really? You need to get out more... Oh, I should have said "unless you're going to be doing a significant fraction of what ftp.cdrom.com does" -- pardon me. ( Notice I said "on a network server.") So exactly whose critical FreeBSD systems are you building with PIIs right now, Tom? -Craig