From: Noel Chiappa
(I suspect, but have yet to verify, that this is in
part because bus
grant lines are run directly from that slot to the UNIBUS adapter slot.)
So it turns out I was half-wrong, but half-right. I was paging through a copy
of EK-PDP84-TM-PR4 (PDP-11/84 Technical Manual) which I recently acquired: in
section 2.1.14 "Backplane (H9277-A)", pg. 2-6, the following sentence appears:
Bus signals BDMGI (pin AR2) and BIAKI (pin AM2) for slots 2 & 3 are jumpered
on the front of the backplane.
(This is not in any other version of EK-PDP84-TM, of which I have several.)
So I got out my spare 11/84 backplane, and sure enough there are two jumpers,
W1 and W2. The traces connected to them are, luckily, on the surface, so it's
possible to see where they go: one end runs to a trace connected from slot 1
to slot 2, and the other to a trace connected from slot 3 to slot 4.
EK-PDP84-TM-PR2 actually contains (pp. D-4 to D-8) prints for the backplane.
Unfortunately, they are very low-res, and only partially readable, but one
can see, on the bottom of the first print (D-4), the two jumpers. The
associated traces do indeed connect to AM2 (BIAKI), AN2 (BIAKO), AR2 (BDMGI)
and AS2 (BDMGO).
In other words, when the jumpers are in, the CPU's BIAKO/BDMGO pins are
connected directly to the UNIBUS adapter's BIAKI/BDMGI pins; when they are
out, those signals are routed through the two 'memory' slots, in the normal
QBUS manner.
What this strongly suggests to me is that those two slots _can_ function as
real QBUS slots. Otherwise, why arrange so that the grant lines can run
through them? I haven't examined all the other pins, to make sure they
contain the full set of QBUS signals - in part because I don't want to pore
over those poor images!
(Speaking of which, does anyone have a copy of either 11/84 Field
Maintainance Print Set - MP-01955 or MP-02536? Neither one seems to be online
- at least, as best I can tell, from some Googling, although others may have
better Google-fu than me - and it would be fabulous to have access to them.)
But I do strongly suspect they can function that way; at some point, when I
have an /84 running, I'll actually try them out. As to why DEC put this
capability in, and then didn't document it or use it - who knows? Maybe they
though it would introduce extra complixity in the user instructions, or
something.
Anyway, I think we are close to cracking this particular puzzle...
Noel