From owner-freebsd-hackers  Mon Nov 27 22:59:32 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id WAA09526
          for hackers-outgoing; Mon, 27 Nov 1995 22:59:32 -0800
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id WAA09511
          for <FreeBSD-hackers@freebsd.org>; Mon, 27 Nov 1995 22:59:19 -0800
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id RAA15565; Tue, 28 Nov 1995 17:57:47 +1100
Date: Tue, 28 Nov 1995 17:57:47 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199511280657.RAA15565@godzilla.zeta.org.au>
To: roberto@keltia.freenix.fr, se@zpr.uni-koeln.de
Subject: Re: Translation Adaptec 1542 versus NCR810
Cc: FreeBSD-hackers@freebsd.org
Sender: owner-hackers@freebsd.org
Precedence: bulk

>> The NCR bios ought to use the values the drive
>> has been initialised with, if it has been used
>> with an Adaptec before. If its a clean install,

Is it as flexible as the BT445C?  I.e., does it guess a geometry from
whatever _valid_ values you put in the partition table?

>In that case, the   drive was out of  the  box, first configured  with  the
>geometry displayed by the drive and it  didn't boot. I  then took the total

Big drives are likely to have >= 64 sectors/track.  Don't even think about
using such a geometry.  It won't fit in the partition table and neither
BIOSes nor fdisks should allow you to specify it.

>number of sectors, divided by the values of sect/trk and heads displayed by
>"boot -v" and calculated the # of cylinders.

This should always work (except in 2.0.5 which displayed the wrong number
of heads and cylinders (off by one error)).

Bruce