Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!texbell!ssbn!bill
From: bill@ssbn.WLK.COM (Bill Kennedy)
Newsgroups: comp.unix.i386
Subject: Re: Adaptec controller problems
Summary: It uses sectors in 2.0.2, tracks in 2.2
Message-ID: <1245@ssbn.WLK.COM>
Date: 18 Nov 89 01:52:15 GMT
References: <36506@ism780c.isc.com> <444@inebriae.UUCP>
Reply-To: bill@ssbn.WLK.COM (Bill Kennedy)
Distribution: na
Organization: W. L. Kennedy Jr. and Associates, Pipe Creek, TX
Lines: 27

In article <444@inebriae.UUCP> I write:
[ A whole lot of stuff ... ]

I said I'd shut up, so I'll be brief.  I was misinformed about how many
defects there could be or how they were handled.  I'll delete everything
except the clarification.  Thanks to ISC for straightening me out on this,
it's important.

>In article <36506@ism780c.isc.com> support (Signed Mike Alcorn) writes:
[ al lot of stuff too ... ]
>>and Adaptec 2322B-8 controller.  ISC did request the use of the drive because
>>it had too many defects to work on 2.0.2.  The problem was related to Mr.
>	^^^^^^^^ How many are "too many"?  Bigger drives always have more
>	defects, is there a limit to the size of the bad track table?  To
>	the number of alternate tracks available for assignment?  If so, what
>	is it so that we don't buy a drive with more than that?

Turns out that 2.0.2 maps sectors.  Bill Miskovetz' drive has 56 sectors
per track, so the 36 reported defects were a lot more sectors than just
36 tracks.  I'm told that one of the differences in 2.2 is that it maps
tracks.  That means that you lose a track (not unusual in bigger systems),
but you have so many on a large capacity drive it doesn't hurt as much as
trying to bookkeep on a sector basis.  I thought that 386/ix was doing
that all along, I was mistaken.
-- 
Bill Kennedy  usenet      {attctc,att,cs.utexas.edu,sun!daver}!ssbn!bill
              internet    bill@ssbn.WLK.COM   or attmail!ssbn!bill