Xref: utzoo comp.sys.mac.hardware:3216 comp.sys.mac:54120
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!think!leander.think.com!ephraim
From: ephraim@leander.think.com (Ephraim Vishniac)
Newsgroups: comp.sys.mac.hardware,comp.sys.mac
Subject: Re: Hard disk benchmarks for Mac IIfx?
Keywords: disk, fx
Message-ID: <36069@think.Think.COM>
Date: 3 May 90 20:10:50 GMT
References: <1800@ndmath.UUCP>
Sender: news@Think.COM
Reply-To: ephraim@think.com (Ephraim Vishniac)
Followup-To: comp.sys.mac.hardware
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 36

In article <1800@ndmath.UUCP> milo@ndmath.UUCP (Greg Corson) writes:
>Has anyone seen any benchmarks of hard disks yet that include performance
>figures for the Mac IIfx?  I'm getting ready to buy one and I'm wondering
>if I can get a faster transfer rate on my SCSI disk by getting a CDC Wren-runner
>drive instead of the usual apple Quantum.

>I've heard the standard Apple Quantum drive is no faster on an FX than it
>is on a CI...does anyone know this to be true?

I just ran a simple test using SCSI Evaluator 1.02 on a IIfx.  My
internal disk is an Apple (Quantum) 80MB disk; the external disk is a
CDC Wren III.  Both disks are formatted with 1:1 interleaving.  I ran
SCSI Evaluator's default tests, for reads only.  For small transfers
(10K, 15K), the Apple 80MB disk was literally off the chart.  I
suspect this means that the controller caches more than 15K, but less
than 20K.  For 20K transfers and larger, the Apple disk did about
6000kbps. 

For comparison, I ran the same test on my external disk, a CDC Wren
III.  Note that this is an old disk (1987?), presumably slower than
the Wren Runner.  I don't believe it does any local caching.
Transfers rates increased fairly smoothly with transfer size, starting
at about 6000kbps and increasing to nearly 8000kbps for the largest
transfers. 

So, the Apple disk scored better on small transfers, probably due to
caching, and the CDC disk scored better on large transfers.  In
practice, which is "better" depends on the cache hit rate, the typical
transfer size, and the seek rate of the drives.  I didn't test seek
times for two reasons: (1) the test is lengthy and (2) my results on
a Wren III have absolutely no bearing on the performance of the Wren
Runner.
Ephraim Vishniac    ephraim@think.com   ThinkingCorp@applelink.apple.com
 Thinking Machines Corporation / 245 First Street / Cambridge, MA 02142
        One of the flaws in the anarchic bopper society was
        the ease with which such crazed rumors could spread.