From owner-freebsd-hardware Tue Mar 16 22:39:33 1999 Delivered-To: freebsd-hardware@freebsd.org Received: from mercury.gfit.net (ns.gfit.net [209.41.124.90]) by hub.freebsd.org (Postfix) with ESMTP id F105F150AF for ; Tue, 16 Mar 1999 22:38:57 -0800 (PST) (envelope-from tom@embt.com) Received: from gizmo (timembt.iinc.com [206.67.169.229]) by mercury.gfit.net (8.8.8/8.8.8) with SMTP id GAA22786 for ; Wed, 17 Mar 1999 06:38:47 GMT (envelope-from tom@embt.com) Message-Id: <3.0.3.32.19990317014025.006ffd90@mail.embt.com> X-Sender: tembt@mail.embt.com X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Wed, 17 Mar 1999 01:40:25 -0500 To: freebsd-hardware@freebsd.org From: Tom Embt Subject: Re: P-II vs K6-2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-hardware@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Whether you get better performance out of a small, fast L2 cache or a >larger, slower one depends entirely on the application. It would be >really interesting to see some figures here instead of theory. > >Greg This would be a fun area to do some benchmarking in. I think a few people have said that certain 3D first person shooter games under Win95 benefitted from the faster, but smaller, cache of the Celeron. I would not be surprised, however, if the big slow cache of a real P2 held a slight edge in an environment with lots of separate threads each doing, say, compiling or something. Anybody care to run some tests? (I don't have a P2-anything) Tom Embt ICQ UIN: 11245398 tom@embt.com d:-)> To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hardware" in the body of the message