Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!ncar!gatech!rutgers!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Anyone own an ICD SCSI hd controler? Others mentioned..
Message-ID: <20297@cbmvax.commodore.com>
Date: 2 Apr 91 22:07:52 GMT
References: <91084.191834IO91461@MAINE.BITNET> <2254@wet.UUCP>
Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 35

In article <2254@wet.UUCP> pk@wet.UUCP (Philip King) writes:

>I do notice with the ICD that screen updates don't appear to happen until the
>transfer/directory scan finishes.  This may be due to the fact that non-DMA 
>controllers tie up the CPU more.  

Sure 'nuf.  And if the ICD that's shipping is the same design as the one we
tested last year on the A3000, it's using up ALL the CPU time during disk 
reads.  Even during seeks.  It triggered the A3000's default timeout of just
over 8us (that's about 200 68030 wait states).  No functional problem on 
production A3000s, which enable to longterm bus error timeout at 1/4 sec,
but an awful waste of your CPU time.  The A2091 uses around 40% of the CPU
time during a disk transfer, the A3000 about 6%.

>The reason I bought the ICD was the price, the (reputed) speed, and the
>removeable-media support.  

That is the fastest way to run a hard disk, if all you care about is disk
performance (eg, CPU time doesn't limit you anywhere).  This technique is
very popular on single-tasking systems like PClones and Macs.

>By the way, I tested the speed with Diskspeed 3.1, which is by all 
>accounts a superior benchmark program to Dperf2.  

It really is.  DPerf doesn't have enough grain at higher speeds, and will
apparently die if it goes too fast.  Mike Sinz could fill in all the details.
One interesting test of SYSTEM performance is to run DiskSpeed and Dhrystone
2.1 at the same time.  What's really needed, though, is a simple "remaining
CPU time" mode in DiskSpeed, so you can not only guage the disk performance,
but the amount of CPU time you're paying for that speed.

-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
      "That's me in the corner, that's me in the spotlight" -R.E.M.