Oops, thought I'd mentioned that originally, this
is the Sigma chassis for
two reasons, it fits nicely in the bottom of a rack, and I'm saving the
BA123 for a spare for my /73.
Here is the original config, didn't work:
KA650 -----
MS650 -----
MS650 -----
RQDX3 TQK50
-rest empty-
I *suspect* that the whole right-hand-side of the Sigma box is CD-only.
The only way to tell for sure is to pull the backplane and trace it out.
This worked:
KA650 -----
MS650 -----
MS650 -----
RQDX3 GRANT
-rest empty-
and this:
KA650 -----
MS650 -----
MS650 -----
RQDX3 empty
GRANT empty
-rest empty-
Final solution:
KA650 -----
MS650 -----
MS650 -----
RQDX3 empty
TQK50 empty
-rest empty-
This confirms my suspicion that the Sigma backplane you have has
CD slots all the way down the right side (though it's not conclusive;
it might only be CD for the first four slots.)
If, indeed, it is CD all the way down, then putting a TQK50 in
next to a memory card there can heavily confuse the memory mapping and
testing.
IIRC, the Diagnostic it stopped on was 34 (but might be
off a few).
Yes, the early tests are memory bus testing and sizing.
Seem
to work great this way, but it sounds like I should swap the RQDX3 and
TQK50 from what Allison says.
Well, if you're playing fully by the DEC rules you should swap
the cards, but both the RQDX3 and TQK50 are smart heavily buffered
devices, and in such a tiny configuration as you have I doubt there'd
be any differences seen at all. OTOH, if you have 3 Q-bus backplanes
with dozens of "stupid" (unbuffered) cards in each, it does eliminate a
lot of heartache to play by the rules!
Do you happen to know the configuration of a Sigma
Backplane? Somehow I
suspect that would answer my questions.