Newsgroups: comp.unix.aux
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!yale.edu!cmcl2!panix!alexis
From: alexis@panix.uucp (Alexis Rosen)
Subject: Re: Comparing disk performance
Message-ID: <1991Jun26.153204.1107@panix.uucp>
Date: Wed, 26 Jun 91 15:32:04 GMT
References: <5805@dftsrv.gsfc.nasa.gov>
Organization: PANIX - Public Access Unix Systems of NY

jim@jagubox.gsfc.nasa.gov (Jim Jagielski) writes:
>This is the output of the "Time Drive" application from the LaCie
>SilverLining disk. I have used it to look at the performance of two types
>of hard disks: the Seagate 330MB Wren Runner (4385M) and the Seagate 175MB
>Wren (sorry, don't recall the part number). [...]
>
>Note that this was run under the Mac OS (6.0.7) on an accelerated MacII
>(the 50MHz DayStar PowerCard) so we might be hitting the SCSI port speed limit
>on the Runner... 

Not really. The Runner and the II are well matched. Only the IIci is a little
faster than the runner.

>                 Once my IIfx upgrade is up and running, I'll do this again.

Don't bother. The IIfx SCSI port is a most embarassing botch. Under MacOS,
it's no faster than a Mac II. The IIci has significantly better performance.
Under A/UX, it _could_ be faster, somewhat. It wasn't under 2.0, and I haven't
done any tests on the released version of 2.0.1.

I recommend that anyone interested in drive speed measurements order a price
list from MicroNet. They're incredibly useful. Next to each unit is a
measurement of the sustainable transfer speed. Unlike marketing literature
I've seen from other companies, though, I've never seem their speed ratings
overestimate drive performance measurably. I used to check them whenever I
got a new model to play with, to see if they were telling the truth. Now I
don't feel any need to. They've established themselves as reliable, as far as
I'm concerned.

---
Alexis Rosen
Owner/Sysadmin, PANIX Public Access Unix, NY
alexis@panix.com
{cmcl2,apple}!panix!alexis