Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!zaphod.mps.ohio-state.edu!rpi!uupsi!rodan!amichiel
From: amichiel@rodan.acs.syr.edu (Allen J Michielsen)
Newsgroups: comp.sys.ibm.pc
Subject: Re: MFM on RLL?
Message-ID: <2450@rodan.acs.syr.edu>
Date: 14 Mar 90 18:58:34 GMT
References: <1828@crash.cts.com>
Reply-To: amichiel@rodan.acs.syr.edu (Allen J Michielsen)
Organization: Syracuse University, Syracuse, NY
Lines: 28

In article <1828@crash.cts.com> jca@pnet01.cts.com (John C. Archambeau) writes:
>paravia@csd4.csd.uwm.edu (Mark David Kakatsch) writes:
>>Can an MFM drive be connected to an RLL controller?
>>I 
>>know that MFM is pretty slow for a 386, but since it would be free,...
>>
>
>Yes it can, but you will get mixed answers on the reliability of the
>configuration.  I personally would NEVER hook an MFM drive to an RLL
>controller.
> 
>

   At risk of starting up, again, the perpetual arguement on this subject, I'll
add the following.  I have run disk intensive, real world applications on 386
systems that have had the interleave optimized for that system & not, and have
measured the following. 1. The transfer speed difference between mfm & rll is
a very small factor in the big scheme.  2. The track to track & access times
are a MUCH bigger factor in the big scheme.
   I too wouldn't do a MFM with a RLL either, but have in a spinch & would only
say the following 3 things.
1. seagate has several drives that they specify are RLL only & 'claim' they 
   be 'damaged' by mfm use.
2. Try it, if it works & runs read/write diagnostics for a couple days non stop
   without errors, it probably will be fine for a long time a average use.
3. No matter what, you should keep up to date backups of your hard disk files.

al