On Sat, Mar 12, 2016 at 11:21 AM, jim s <jwsmail at jwsss.com> wrote:
What is the provenance / source of the panels?
Mine came from an acquisition by Nick Allen from a collection in Georgia.
I believe there was a Multics installation in Atlanta they were removed
from.
Multics site: SCSI. Southern Company Services, Inc., Atlanta GA. Nuclear
fuel
inventory. Installed 1982.
(2 L68, 4 MSU0451, 4 MSU0501)
2 Level 68 CPUs, 4 MSU0451 disk drives (309 kg, removable pack, 156 Million
9 bit bytes), 2 MSU0501 disk drives (non-removable pack, 1101 Million 9 bit
bytes)
The panels on the 6180 at USL were all inside side access panels for one of
the rows of hardware boxes. One box panel was usually
exposed with the door
removed, but it could be closed up. There were problems which required
access to one of the panels frequently in operations, so it was seldom
closed.
We probably could get access to Dockmaster with some advance arrangement
and good will on the part of the CHM when they have time to arrange access
to the storage to which it was moved to see actual installed panels.
I agree, the black panel has about the only interesting display.
+David Griffith
I might also suggest that once David Griffith finishes porting the PDP 10
Panda panel and has that design working and integrated that there may be
enough blink'n lights there to display a satisfying 6180 display on a
normal desktop case.
the advantage is that it is at least already 36 bits and has some of the
nonsense of having that bit count worked out already. I'd think we
(someone) could fork and add a second bank of lights, or use two of the
Panda usb devices to put out a lot of information about a 72 bit 6180.
His main problem now is with interfacing and coding PDP 10 assembly code
which is obviously not useful for re-purposing it for Multics use anyway,
and is internal to SIMh PDP10 emulation.
If a lot of people who are interested in blinking Multics Honeywell 6180
displays were interested it would contribute a lot to him selling out a run
of his board kits.
I don't think there is that much interest, but I'll keep a weather eye
out.
Interfacing to beaglebone would be simpler for me, as I already have the
infrastructure in-place for my simulated display. (All of the needed data
is in a shared memory segment, a standalone program just scrapes the data
and sends it out.)
-- Charles