From owner-freebsd-hackers  Sat May 24 04:39:24 1997
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id EAA22163
          for hackers-outgoing; Sat, 24 May 1997 04:39:24 -0700 (PDT)
Received: from verdi.nethelp.no (verdi.nethelp.no [195.1.171.130])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id EAA22158
          for <hackers@FreeBSD.ORG>; Sat, 24 May 1997 04:39:20 -0700 (PDT)
From: sthaug@nethelp.no
Received: (qmail 25680 invoked by uid 1001); 24 May 1997 11:39:16 +0000 (GMT)
To: shag@concentric.net
Cc: hackers@FreeBSD.ORG
Subject: Re: Controler for SCSI
In-Reply-To: Your message of "Sat, 24 May 1997 05:06:43 -0600"
References: <3386CBC3.88C37DBB@concentric.net>
X-Mailer: Mew version 1.05+ on Emacs 19.28.2
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Date: Sat, 24 May 1997 13:39:16 +0200
Message-ID: <25678.864473956@verdi.nethelp.no>
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> > Nope, EIDE supports PIO mode 4 (16.6MB/sec) and DMA at about the same
> > speed.  16.6MB/sec is standard for PCI EIDE controllers.  Crufty SCSI
> > PIO controllers are unlikely to be as fast.  New Quantum EIDE drives
> > support DMA at 33.3MB/sec.
> 
> 	Now if only it worked that way In Real Life... :-) I have a pretty
> decent Fijitsu drive, and my Quantum Lightning, which is among the worse
> SCSI-II drives manufactered in the last couple of years, takes it to
> town on every benchmark on the planet. I know the controller is crap,
> but it was cheap, and I was just hoping. =)

I have to disagree. I have a PPro-200 with a 1.7 GB Quantum FB, PIO mode.
I get a sustained 6.8 MB/s on sequential read from this, while my Seagate
Barracuda "only" gets around 6 MB/s.

The price differential between SCSI and EIDE is significant enough these
days that I think it's vital for FreeBSD to get EIDE DMA support.

Steinar Haug, Nethelp consulting, sthaug@nethelp.no