IMSAI 8080 (our show so far...)

From: Dwight Elvey <elvey_at_hal.com>
Date: Tue Jun 29 14:13:10 1999

"Bill Sudbrink" <bill_at_chipware.com> wrote:
> Ok. There is a 16 pin ribbon cable between
> the front panel and the MPU-A card.

Hi Bill
 I'm sure there is a scrambling of this connection
someplace. This is the only place that can do the
things you mentioned. The front panel is hard
wired to the S-100 bus. This means that wires can't
get mixed here. The data lines from the CPU board
to the front panel can get swap around. Remember that the
way the front panel does things is by jamming instructions
and data directly onto the CPU's bus through the cable.
The results are read from the S-100 bus. In other words,
when you use a front panel switch, like examine, it
forces a jump instruction into the CPU with
the address you specify with the switches over the cable,
in sequence. A real clever trick I'd say. You'll note that
the C3 bit sequence still works if the data bits 7-0 are mirrored
to 0-7. This is why the panel seems to do anything at all.
 If reversing the ribbon connector in the header doesn't
work, try removing the socket from the CPU card or front
panel and installing it on the other side of the PC
board. I would suspect that this was an error in the construction
of the front panel from day one. It is most likely the
front panel that is backwards. Just flipping the IDC header
on the cable won't mirror the wiring to the connector.
Putting the socket on the other side of the board will
mirror things.
 At worst, you'll have to hand wire the cable to make up for a
non-standard connector but I'd think it was more likely
just getting the header socket right at both ends. I don't
have a schematic of the panel in front of me but it is a
simple mapping problem.
Dwight
Received on Tue Jun 29 1999 - 14:13:10 BST