Differences between PDP-11/83-84 M8190-AB/AC and AD/AE CPU cards

[Jerome H. Fine]

 >Noel Chiappa wrote:

>    > From: Jerome H. Fine
>
>    > With regard to the MSV11-JD in the PDP-11/84, was this board placed
>    > after the KDJ11-AE board or before it?
>
>After. The CPU _has_ to go in slot 1 on the 11/84 backplane. (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.)
>  
>
In that case, I would assume that the PDP-11/84 backplane is wired
so that the PMI memory is activated correctly WHEN the CPU is in
slot 1 and the first memory board is in slot 2.  Note that I am repeating
what you just stated.

The reason that I have stated this for the PDP-11/84 is because for
the PDP-11/83 (usually in a BA23 or a BA123), the backplane is
wired so that the PMI memory is activated correctly WHEN the
first memory board is in slot 1 and the CPU follows the memory
boards.

>    > Also, has anyone ever used a KDJ11-AC board with the 15 MHz crystal in
>    > the PDP-11/84 backplane? I can't see why it should not work, but just
>    > maybe the PDP-11/84 backplane and a simple KDJ11-AC board are not
>    > compatible.
>
>Would a KDJ11-A board even work in an 11/84 backplane? It wouldn't (couldn't)
>do the special PMI (actually, we ought to distinguish between 'QBUS PMI',
>such as in the 11/83, and 'UNIBUS PMI', as in the 11/84; the second is a
>superset of the first, I believe) protocol, so it wouldn't work with the
>UNIBUS adapter.
>
>See also Micronote #39, "Differences between KDJ11-A and KDJ11-B".
>
I made another err.  I meant the KDJ11-BD board or the M8190-AC.

Since I will never have an opportunity to ever use a PDP-11/84, it will 
never
be possible to check my question.

HOWEVER,  NOTE that the identical CPU board, KDJ11-BF (M8190-AE),
is used in BOTH the PDP-11/83 and the PDP-11/84.  The KDJ11-BD is the
M8190-AC, a quad board (if I remember correctly).  Although the KDJ11-AC
has a 15 MHz crystal, when used in the correct position in a BA23 box, 
namely
AFTER the PMI memory, the KDJ11-BD is reported as a PDP-11/83 by the
RESORC.SAV program and the speed of the system is about 20% faster than
when the PMI memory is placed into the ABCD slots AFTER the CPU at which
point RESORC.SAV reports the CPU as the PDP-11/73B.

That is why it seems reasonable to hope that the PDP-11/84 can also use the
KDJ11-BD instead of the KDJ11-BF, except that it would then run about
10% to 15% slower.

Jerome Fine