Classic_-_Concepts@cup.portal.com (01/02/91)
This is an update on my previous post for a friend with the following
system:
Amiga 2000 revision 4.2 motherboard
ST-225 20 meg hard drive (no other peripherals except 2 floppies)
2090 controller (this is the earlier, nonautoboot version of the 2090A)
SupraRAM 2000 populated with 2 meg, memory diagnostics are fine
In the previous post, the hard drive was urping on the binddrivers command
and doing a few seconds of disk seeks before failing.
A talk with Supra resulted in a chip being swapped, one of the four labelled
ones on the board. This chip was specific to the 4.2 revision motherboard.
Supra supplied the chip free of charge. The unpleasant disk seeks went away.
Helpful suggestions from the previous post were tried, but we still can't
make it work. I have tried the hard drive controller card and the RAM
chips in adjacent slots (both sides) and in the farthest-from-each-other
slots (both configurations) with no success.
The memory card works perfectly without the hard drive controller and,
apparently, vice versa, but not together.
Further to the previous post, I notice that the info command from shell
yields some strange information:
The boot drive, df0:, is listed as having a read/write error. There
is no read/write error and all the commands on the diskette continue to
function perfectly. Removing one or the other of the cards removes the
fictitious read/write error message.
The error message that comes up when the binddrivers command is run
says that the hard drive device is not a DOS device.
The mount command seems to work correctly, but the system can't find
the device (presumably due to binddrivers command not succeeding).
I tried changing the memory type in the mountlist from 1 to 0, as I
have heard this helps on other types of controllers, but this didn't change
the symptoms or error messages.
Oh, I also found out he is running the 1.3 operating system with the
1.2 Kickstart in ROM. One call to a pretty savvy technician resulted in
him saying this probably wasn't the source of the problem. Could anyone
confirm/deny the possibility? I think he'd rather wait for the ECS than
spend the money on 1.3 at this late date. However, if that would solve
the problem, I think he'd rather spring for the $30 upgrade than not be
able to use the extra RAM. If it doesn't solve the problem, it's $30
which would be better saved up for the ECS.
So has anyone heard of/solved this problem? Help greatly needed and
appreciated, otherwise the memory card was a waste of time/money, or it's
going to take a whole new controller and hard drive, a major expense.
Thanks, J. Petersen
ps, as mentioned in the previous post, removing one or the other of the
cards results in everything working normally again.