Date: Wed, 15 Oct 1997 09:31:57 -0700 (PDT) From: "Eric J. Schwertfeger" <ejs@bfd.com> To: dkelly@hiwaay.net Cc: Jonathan Lemon <jlemon@americantv.com>, don@PartsNow.com, Marty Leisner <leisner@sdsp.mc.xerox.com>, Jacques Hugo <jacques@wired.ctech.ac.za>, questions@FreeBSD.ORG Subject: Re: mmx or ppro Message-ID: <Pine.BSF.3.95.971015092605.329A-100000@harlie.bfd.com> In-Reply-To: <199710150135.UAA20999@nospam.hiwaay.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 14 Oct 1997 dkelly@hiwaay.net wrote: > When faced with this choice myself a couple of months ago, I chose the PPro > at 166 MHz with 512k cache at about $325 over a cheaper 180 with 256k or > more expensive 200 with only 256k. In addition to the larger L2 cache I got > 66 MHz bus speed vs. 60 for the 150 or 180. Not rerribly interested in > overclocking as I'd guess the PPro-166 is a -200 that failed the grade. Every PPro150 I've had (5 so far) has been happy to run at 166 (though admittedly with a smaller cache), and all but one would run solid at 180. Since the official 166 comes off a different production line, I'm sure that the 150's aren't tested at 166, so I've been running them all at 166. In playing around with 180 vs 166, I noticed that probably due to the on-chip L2 cache, external bus speed means a lot less to a PPro than it does to a Pentium. I've also noticed that for very small programs that fit in the L1 cache, the PPro really doesn't run that much faster than a Pentium of similar clock speed, but for larger tasks (kernel compiling, make world) the PPro easily outpaces the Pentium.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.971015092605.329A-100000>