Path: utzoo!utgpu!water!watmath!isishq!Geoffrey_Welsh
From: Geoffrey_Welsh@isishq.UUCP (Geoffrey Welsh)
Newsgroups: comp.sys.cbm
Subject: Re: is it possible to read/write Commodore disks on an IBM PC?
Message-ID: <55.22CBBFB0@isishq.UUCP>
Date: 27 Jun 88 18:12:17 GMT
Organization: FidoNet node 221/162 - ISIS International, Waterloo ON
Lines: 40


 > From: erict@flatline.UUCP (j eric townsend)
 > Date: 22 Jun 88 05:38:54 GMT
 > Message-ID: <935@flatline.UUCP>


 > Instead of making

 > a "where is the head now" sensor, they used GCR encoding.
 
   While the rest of your message is quite accurate, this part suggests 

that the two are mutually exclusive. GCR is an encoding scheme, just 
like FM, MFM, or RLL (in fact, GCR is a form of RLL). There is nothing 
inherent in GCR encoding that prevent the operation of a head position 
sensor. It's just that Commodore decided not to.
 
   With the 4040's, that wasn't a problem. The SA-390 drives were pillars 

of reliability and, when the disk controller couln't figure out what 
track the head was on, it sent about 75 half-steps toward the outside 
of the disk and started looking for track 1. The SA-390 alignment was 
not damaged by the "knocking" because the stop bumper was slightly 
elastic, so the circular platter in which the spiral head positioning 
groove was molded never (well, almost never) slipped on the spindle. 
The head, because of the way it was guided along the spiral groove, 
never actually hit anything (if my memory serves me correctly).
 
   Personally, I blame the choice of the ALPS drive for the majority 

of the problem, though the slipping problem is present to varying degrees 
in all half-height drives (there's no room for the platter with the 
spiral groove in it). Fortunately, most disk operating systems don't 
knock the heads often...
 
   Geoff
 
   {allegra|decvax|utzoo|clyde|etc.!}watmath!isishq!Geoffrey_Welsh