Path: utzoo!attcan!uunet!samsung!usc!rutgers!cbmvax!steveb
From: steveb@cbmvax.commodore.com (Steve Beats)
Newsgroups: comp.sys.amiga
Subject: Re: 3000 autobooting a 1096N
Message-ID: <13766@cbmvax.commodore.com>
Date: 13 Aug 90 15:11:30 GMT
References: <754.26C4F281@weyr.FIDONET.ORG>
Reply-To: steveb@cbmvax (Steve Beats)
Organization: Commodore, West Chester, PA
Lines: 29

In article <754.26C4F281@weyr.FIDONET.ORG> Randy.Coghill@f70.n140.z1.FIDONET.ORG (Randy Coghill) writes:
>Alright I would like someone at CBM (Dave Haynie this is probably
>right up your alley).
>
>I recently purchased my 3000 about 2 weeks ago.  At that time my
>dealer (and boss) allowed change the Quantum 40S for a Seagate 1096N
>hard drive (80 megs compared to 40 on the Quantum).  Well everything
>goes fine util I power down an turn on my system....No autoboot!! 
>Seems my little newly born 3000 does not like to wait till the hard
>drive has initialized.
>
The problem IS with the Seagate drives.  Seagate don`t have any ROMs on
their drives so the firmware has to be loaded in from disk.  While this
is happening, the drive does not respond to SCSI selections at all.  Most
other drives will at least respond and return a not-ready status.  This 
is acceptable, since the A3000 SCSI driver will wait until the drive is
ready.  As far as the A3000 is concerned, a drive that does not respond
to selection is not even plugged in!  

However, there is hope on the horizon.  The A3000 SCSI driver is based
on the same one that was used for the 2091.  The 2091 had some jumpers
that alter the behaviour of various SCSI options.  One of these options
is the time spent waiting for drives to power up.  Since the A3000 has
no SCSI jumpers, we used bits in the non-volatile RAM of the clock
instead.  Unfortunately, an editor has not yet been written for twiddling
these bits.  If you feel like doing a bit of programming, take a look at
resources/battmembits.[h|i] which should clue you in enough.

	Steve