Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!wuarchive!uwm.edu!ux1.cso.uiuc.edu!deimos.cis.ksu.edu!eecea!khc
From: khc@eecea.eece.ksu.edu (Ken Carpenter)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Perstor HD controllers
Summary: Slower than WD but can format when WD fails
Message-ID: <897@eecea.eece.ksu.edu>
Date: 4 Dec 89 20:57:33 GMT
References: <6544@shlump.nac.dec.com> <[2579aaa5:372.1]comp.sys.ibm.pc;1@nstar.UUCP>
Reply-To: khc@eecea.UUCP (Ken Carpenter) khc@eecea.eece.ksu.edu
Organization: Kansas State University, Manhattan
Lines: 15


I formatted a Miniscribe 3085 using the Perstor HD controller last August.
It gave the 1.9 times capacity as advertised, but could not run at 1:1
interleave, requiring 2:1 instead (on a 16MHz 386 box).  It ran without
error until the board died about 6 weeks later.  I replaced it with the
WD1006V-SR2.  This controller can run at 1:1 interleave and so gives
transfer rates about a factor of two higher.  However, the Miniscribe
could not be used with it.  The Miniscribe 3085 reported errors on
about one of 10 tracks when formatted using the WD RLL controller, where
it had reported none using the Perstor.  Further, the bad tracks changed
randomly when one repeated the test scans.  Trying to use it led to such
behaviour as being able to load a file to it, but then not be able to
read it back immediately, even with most of the tracks marked bad.
The conclusion:  MS3085 won't support RLL.
The question: how is it able to support ARLL?