Roe Peterson wrote:
I'm looking at upgrading a pdp11/73 in a
pdp11/23plus box to an 11/83. Is there a simple multimeter test I can do on a qbus slot
to determine if it is Q/Q -- which i understand can be serpentine, or Q/CD, which are PMI
capable for the 11/83 CPU.
I hope that Allison is available to determine if I am suggesting a
possibility that might damage a board of the backplane.
I doubt that is a possibility, but just in case, can anyone else please
comment as well.
A BA23 box is wired with the top 3 slots as Q/CD and the lower
5 slots as Q22 / Q22. If a backplane slot is wired as Q / CD,
then that slot can hold ONLY a single board, either dual OR quad,
on the left side of the slot. If a backplane slot is wired as Q/Q,
then either one quad board or two dual boards are used. The order
of the boards in a BA23 is:
!A 1
2A 2
3A 3
4A / 4B 4 / 5
5A / 5B 7 / 6
6A / 6B 8 / 9
etc.
Depending on if your PDP-11/73 uses any or none of the "B" portion
of the backplane when a dual board is present, you already know if
any of the backplane is Q/Q.
Since you have not provided a list of the boards and the location of
each board, I can not comment further.
However, you can always place a dual board in an "A" slot and leave
"B" portion empty. If that slot is Q/Q, then the bus grant will be broken
and any boards beyond that location in the chain will not interrupt.
If I have not explained the situation carefully enough, please ask more
questions.
Do you already have the M8190-AE and the PMI memory boards?
If you do, then you may want to compare the speed difference. When
I performed a similar benchmark under RT-11, I found that the PDP-11/83
was about 33% faster than the PDP-11/73. About 20% was due to using
the PMI memory (which usually works with the PDP-11/73 boards as well)
and about 13% was due to the faster crystal on the M8190-AE board.
If you have any more questions, please ask.
Jerome Fine