Path: utzoo!mnetor!uunet!tektronix!tekcrl!tekgvs!keithe
From: keithe@tekgvs.TEK.COM (Keith Ericson)
Newsgroups: comp.sys.ibm.pc
Subject: Re: AT ROM--drive type table entries
Message-ID: <3400@tekgvs.TEK.COM>
Date: 1 May 88 00:19:47 GMT
References: <423@upvax.UUCP>
Reply-To: keithe@tekgvs.UUCP (Keith Ericson)
Distribution: na
Organization: Tektronix, Inc., Beaverton,  OR.
Lines: 41
Keywords: AT, HardDisk, ROM, BIOS, SETUP
Summary:

Expires:

Sender:

Followup-To:


In article <423@upvax.UUCP> stevewa@upvax.UUCP (Steve Ward) writes:  >
>[problem with disk not being listed in the BIOS ROM table]
>
>I remember seeing an ad for a ROM chip(s) you could buy to put in the
>extra sockets in the AT to provide additional drive table entries.
>The problem is I don't remember where I saw it and I can't find it.

I've just started playing with modifying the BIOS ROM (copying it to a
PROM programmer, editing it, burning new PROMs) and indeed I can make
any disk type entry I need. (The following applies to my Intel
Motherboard 386AT with Phoenix BIOS ROMs and MS-DOS 3.3.)

Two problems:

1. Something is incapable of recognizing any more than 1024 cylinders.
I know it's not the controller (a Western Digital WD-1003A-WA2) because
I've used it with a UNIX installation that utilized the 1224 cylinders
of my Maxtor XT-2190. I can't get fdisk to acknowledge any more than
1023 cylinders. In fact, with the 1224 cylinder entry fdisk thought
there were only 199 cylinders. Somewhere the tenth bit of the cylinder
count is getting dropped. Anyone know where? Even Norton Utilities
Partition Table Editor can't handle it.

2. The IBM diagnostics disk - which contains the setup program to write
the information in the CMOS RAM - now reports that I'm now running it
(the setup program) on a non-IBM machine and I should use the setup disk
that came with the machine. Well. I hate to tell it this, but it wasn't
an IBM machine _before_ I modified the ROMs, either! What gives? I
remembered to change the last byte of the PROM to obtain a checksum of
0. Actually, I changed the last byte in both PROMs (high- and low-byte)
so that each PROM would have a 0 checksum (as they did before the
modification).

As the punchline to the old joke* goes: "How do it _know_?!?!"

keith

*One old geezer was remarking to another that the smartest thing in the
world had to be a thermos bottle because it knew how to keep hot things
hot and cold things cold. Replying to the rebuttal that "That ain't so
great" the first geezer exclaimed, "Yeah, but how do it _know_?"