Path: utzoo!attcan!uunet!cs.utexas.edu!utastro!bigtex!james
From: james@bigtex.cactus.org (James Van Artsdalen)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: RLL vs IDE
Message-ID: <48831@bigtex.cactus.org>
Date: 14 Oct 90 02:59:13 GMT
References: <40552@eerie.acsu.Buffalo.EDU>
Reply-To: james@bigtex.cactus.org (James Van Artsdalen)
Organization: Institute of Applied Cosmology, Austin TX
Lines: 30

In <40552@eerie.acsu.Buffalo.EDU>, axaris@acsu.buffalo.edu
	(vassilios e axaris) wrote:

> However, I have received personal messages mentioning that Unix and
> Xenix do not work properly with IDE drives. Does this apply to all IDE
> drives or just a few particular ones?

Some versions of unix / xenix try to figure out the geometry of a
drive.  I recall that ISC did this to drives on the secondary
controller in at least some cases if not all.  This may not work for
an IDE drive.  ISC may have changed this.

The problem is that many IDE drives adjust their apparent geometry to
whatever the system wants.  If you ask for 64 sectors per track, you
got it, with fewer cylinders or heads.

The way you figure out the geometry is to tell the drive it has 255
sectors per track, 16 heads and 1024 cylinders (for example), and see
if it works by seeking to those values.  The code detecting the
geometry may be assuming that the drive's geometry is fixed, whereas
each time you program a new geometry into an IDE, it changes.

The bottom line is that you'll just have to try it out.  I don't know
if anyone ever used the above algorithm for a drive on the primary
controller/connector.

PS. Remember to keep the IDE cable length SHORT!
-- 
James R. Van Artsdalen          james@bigtex.cactus.org   "Live Free or Die"
Dell Computer Co    9505 Arboretum Blvd Austin TX 78759         512-338-8789