From owner-freebsd-hackers  Wed May 17 21:39:22 1995
Return-Path: hackers-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id VAA24687
          for hackers-outgoing; Wed, 17 May 1995 21:39:22 -0700
Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id VAA24678
          for <hackers@FreeBSD.org>; Wed, 17 May 1995 21:39:19 -0700
Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id VAA13272; Wed, 17 May 1995 21:37:48 -0700
From: "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>
Message-Id: <199505180437.VAA13272@gndrsh.aac.dev.com>
Subject: Re: Adaptec 2940?
To: tom@haven.uniserve.com (Tom Samplonius)
Date: Wed, 17 May 1995 21:37:48 -0700 (PDT)
Cc: gibbs@estienne.CS.Berkeley.EDU, hackers@FreeBSD.org
In-Reply-To: <Pine.BSF.3.91.950517192321.5578C-100000@haven.uniserve.com> from "Tom Samplonius" at May 17, 95 07:30:38 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1154      
Sender: hackers-owner@FreeBSD.org
Precedence: bulk

> 
> 
> On Mon, 15 May 1995, Justin T. Gibbs wrote:
> 
> > >  Is it just me, or is the 2940 slower than the 1742 and 2742?
> > >
> > >  (BTW, I'm running a kernel supped last week)
> > >
> > >Tom
> > 
> > I'll let you know once I get my Pentium machine. :)  I don't see 
> > any reason why it would be slower than a 2742.
> 
>   I swapped a ASUS AMD486DX4100 PCI with a 2940, for a AMI 486DX266 EISA 
> with a 2742, and found that "iozone auto" would give consistently better
> results.  Running top and running two dd's or two iozone's revealed that EISA
> system was using less system and interrupt time for the same job.
> I used a almost current kerenel and the same drives for both.

Which ASUS model of board, there are no less than 3 ASUS PCI 486 MB and
the memory bandwidth on all 3 of them are different.

Wich AMI model motherboard, I don't know how many EISA boards they make.

I suspect you may be comparing CPU <-> memory bandwidth when you look
at the system load more than anything.


-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                   Custom computers for FreeBSD