Date: Tue, 10 Jun 1997 14:00:53 -0400 From: Dave Alderman <dave@persprog.com> To: Steve Passe <smp@csn.net> Cc: Chuck Robey <chuckr@glue.umd.edu>, Tom Samplonius <tom@sdf.com>, freebsd-hardware@FreeBSD.ORG Subject: Re: fastest possible FreeBSD system? Message-ID: <339D9655.6B9285BE@persprog.com> References: <199706100358.VAA18537@Ilsa.StevesCafe.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Steve Passe wrote: > > Chuck, > > > Steve, since the PPro cache is accessed at the clock rate, and the > > Pentium II cache is accessed at the bus rate, I would think the > > PPro would win hands down, in performance, no? > > the benches I have seen say no, check out: > > http://sysdoc.pair.com/pentiumII.html Unfortunately, these benchmarks do not evaluate multiprocessor performance. In multiprocessor tests (unfortunately with Windows NT), A Dell dual Pentium Pro 200 server was slightly outperforming a dual Pentium II 266 server from Dell. I saw this in PC Week (I think). If anyone can cite the article, please do so. For some reason, Dell had priced the Pentium Pro system MUCH higher that the Pentium II - far more than the simple hardware differences would justify. I would speculate that SMP is more sensitive to changes in the L2 cache performance uniprocessor designs since the Pentium II is otherwise superior unless there is something fundamentally flawed with Slot One (which will be replaced by the end of the year anyhow with Slot Two). Another possibility is that the quite mature FX chip set is less optimal with Pentium II's when attempting SMP. -- It's not my fault! It's some guy named "General Protection"! --Ratbert David W. Alderman dave@persprog.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?339D9655.6B9285BE>