Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!jwt!john
From: john@jwt.UUCP (John Temples)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: IDE drives: good or evil?
Message-ID: <1991Mar24.173308.3337@jwt.UUCP>
Date: 24 Mar 91 17:33:08 GMT
References: <12492@darkstar.ucsc.edu> <12070002@hplvec.LVLD.HP.COM>
Organization: Private System -- Orlando, FL
Lines: 15

In article <12070002@hplvec.LVLD.HP.COM> calloway@hplvec.LVLD.HP.COM (Frank Calloway) writes:
>I got 117 MBytes under DOS 4.01 and a data tranfer rate of just over
>800K, as measured with Coretest

Since most IDE drives have on-board caches, all Coretest tells you is how
fast you can read from that cache.  Not particularly meaningful unless your
application reads the same block of 64K from the drive over and over...
But Coretest comes up with an impressively high number which looks good in
ad copy -- too bad it doesn't mean anything.

Is there a Coretest-like program out there which reads, say, 10 contiguous
megabytes from a drive to measure the transfer rate?  This should
defeat just about any hardware or software cacheing mechanism.
-- 
John W. Temples -- john@jwt.UUCP (uunet!jwt!john)