27 Jul
2015
27 Jul
'15
10:42 a.m.
From: Brian L. Stuart: Monday, July 27, 2015 10:00 AM
My drawings are for version D on revision C etch, but they
agree with yours on these connections. I was misreading
which signal went to pin 9 vs pin 7.
So it appears the KM8E does read a constant 0 for bit 3
during GTF (and RIB). This will apparently be different
everywhere else (DECmates, KT8A, etc.).
BTW, my replies have been bouncing from the Yahoo
groups, so you may want to forward stuff or otherwise
keep those folks in the loop.
Vince
I can see
where this happens in the M837 schematic (E50),
whenever DF is gated to AC9-11. That, in turn, seems to
be for GTF or RIB.
I wonder if there are differences in different versions of the
schematics. If I'm reading the versions I got from bitsavers
correctly, on E50, Pins 2, 5, 11, and 14 are gated on to
DATA3, 9, 10, and 11 by a low on Pin 9. It looks like Pin 9
does identify the RIB and GTF instructions. But on this
version, it looks like Pin 2 is pulled high through R8. The
interrupt inhibit signal does come into Pin 1 which is gated
onto DATA3 by a low on Pin 7, and it's controlled by the
Omnibus IND signals for getting info to the front panel.
So it appears that these versions of the drawings agree
with the maindec and the observed behavior rather than
the documentation.
The versions I'm looking at appear to be for the Revs D
and E. Are the drawings you have that agree with the
documentation for an earlier rev, perhaps?