1 Mar
2005
1 Mar
'05
11:26 a.m.
I finally got an hour free to dig back into my /45 restoration. Thanks to
some spare boards to help test and some very good "CE quickdocs" (thanks a
million ashley!), I got a little further. However, it brings me to two
questions.
1) In the absence of full docs on the MF11-L... I'm not sure about the
following. It looks like the 1st slot of the MF11-L backplan gets a unibus
in (A-B) from the last slot of the cpu backplane. The 2nd slot is my G110,
the 3rd is my G231, and the 4th slot has my H214 (C-F) in it. But looking at
some of the docs, it appears that an H214 can be in slots 1, 4, and a few
others per the diagram. My system arrived with one H214 in slot 4. Should it
be first... in slot 1 under the unibus in? Since I had a spare from ashley,
I put one in slot 1 and one in slot 4. I know the jumpers on the other
boards may not reflect 16K, but I figured it might change things. So if I
have just 8K set up, should the H214 be in slot 1, 4, or does it matter?
2) My front panel has an oddity (to me). If I select say address 10000 and
hit load addr the address lights respond correctly with 10000. I then hit
examine and get nothing (blank data lights). Whatever... but here's the
interesting part. If I keep pressing examine over and over again, the
address lights count up just like they should - except they skip bit zero
(the rightmost bit, either 0 or 15 whatever dec nomenclature is). I don't
mean that the rightmost bit just isn't displaying correctly, the actual
counting sequence acts like the rightmost bit is the 2nd light from the
right. I can tell it's not just a bulb problem because I don't have to hit
load addr twice to "move on", the count is smooth and sequential without the
rightmost bit. Any thoughts offhand?
I know my memory system is likely not jumpered right. I'm looking through
those jumpers either tonight or tomorrow and verify them.
Thanks for any thoughts!
Jay
1 Mar
1 Mar
11:35 a.m.
I finally got an hour free to dig back into my /45
restoration. Thanks to
some spare boards to help test and some very good "CE quickdocs" (thanks a
million ashley!), I got a little further. However, it brings me to two
questions.
1) In the absence of full docs on the MF11-L... I'm not sure about the
following. It looks like the 1st slot of the MF11-L backplan gets a unibus
in (A-B) from the last slot of the cpu backplane. The 2nd slot is my G110,
the 3rd is my G231, and the 4th slot has my H214 (C-F) in it. But looking
at
some of the docs, it appears that an H214 can be in
slots 1, 4, and a few
others per the diagram. My system arrived with one H214 in slot 4. Should
it
be first... in slot 1 under the unibus in? Since I had
a spare from
ashley,
I put one in slot 1 and one in slot 4. I know the
jumpers on the other
boards may not reflect 16K, but I figured it might change things. So if I
have just 8K set up, should the H214 be in slot 1, 4, or does it matter?
Jay, I think I have all the original documentation on the MF11-L, both the
big
field maintenance prints and the user manual. I'll double check, but I do
believe that these came with my 11/35.
Ashley
11:50 a.m.
On Tue, 2005-03-01 at 10:26 -0600, Jay West wrote:
1) In the absence of full docs on the MF11-L...
I'm not sure about the
following. It looks like the 1st slot of the MF11-L backplan gets a unibus
in (A-B) from the last slot of the cpu backplane. The 2nd slot is my G110,
the 3rd is my G231, and the 4th slot has my H214 (C-F) in it. But looking at
some of the docs, it appears that an H214 can be in slots 1, 4, and a few
others per the diagram. My system arrived with one H214 in slot 4. Should it
be first... in slot 1 under the unibus in? Since I had a spare from ashley,
I put one in slot 1 and one in slot 4. I know the jumpers on the other
boards may not reflect 16K, but I figured it might change things. So if I
have just 8K set up, should the H214 be in slot 1, 4, or does it matter?
I *think* I have full docs for it. Let me look it up in the next couple
of days and I'll get back to you. I should at least have a module
utilization for an MF11-L.
2) My front panel has an oddity (to me). If I select say address 10000 and
hit load addr the address lights respond correctly with 10000. I then hit
examine and get nothing (blank data lights). Whatever... but here's the
interesting part. If I keep pressing examine over and over again, the
address lights count up just like they should - except they skip bit zero
(the rightmost bit, either 0 or 15 whatever dec nomenclature is). I don't
mean that the rightmost bit just isn't displaying correctly, the actual
counting sequence acts like the rightmost bit is the 2nd light from the
right. I can tell it's not just a bulb problem because I don't have to hit
load addr twice to "move on", the count is smooth and sequential without the
rightmost bit. Any thoughts offhand?
That is correct behavior. The 11's are 16 bit machines but are capable
of byte addressing. When they're doing words, bit 0 is well 0. When
they're doing bytes, bit 0 can be either to represent the correct byte.
Front panel operations are done as words, hence bit 0 will be 0.
--
TTFN - Guy
noon
Guy wrote...
That is correct behavior. The 11's are 16 bit
machines but are capable
of byte addressing. When they're doing words, bit 0 is well 0. When
they're doing bytes, bit 0 can be either to represent the correct byte.
Front panel operations are done as words, hence bit 0 will be 0.
DOH! Of course! Got it, and THANKS!
Jay
3:03 p.m.
On Tue, 2005-03-01 at 10:26 -0600, Jay West wrote:
2) My front panel has an oddity (to me). If I select
say address 10000 and
hit load addr the address lights respond correctly with 10000. I then hit
examine and get nothing (blank data lights). Whatever... but here's the
interesting part. If I keep pressing examine over and over again, the
address lights count up just like they should - except they skip bit zero
(the rightmost bit, either 0 or 15 whatever dec nomenclature is). I don't
mean that the rightmost bit just isn't displaying correctly, the actual
counting sequence acts like the rightmost bit is the 2nd light from the
right. I can tell it's not just a bulb problem because I don't have to hit
load addr twice to "move on", the count is smooth and sequential without the
rightmost bit. Any thoughts offhand?
(I'm not sure of this, yay, forth if I be mistaken, may the PDP-11
gurus, ye who lived and worked in the age wherein people talked thusly,
step forth and smite me with a hardware referance manual)
IIRC the PDP-11 is byte-addressed... It's counting up two bytes at a
time.
--
Tore S Bekkedal <toresbe at ifi.uio.no>
6:38 p.m.
2) My front panel has an oddity (to me). If I select
say address 10000 and
hit load addr the address lights respond correctly with 10000. I then hit
examine and get nothing (blank data lights). Whatever... but here's the
interesting part. If I keep pressing examine over and over again, the
address lights count up just like they should - except they skip bit zero
(the rightmost bit, either 0 or 15 whatever dec nomenclature is). I don't
DEC have 0 as the LSB (rightmost bit on the panel)
mean that the rightmost bit just isn't displaying
correctly, the actual
counting sequence acts like the rightmost bit is the 2nd light from the
right. I can tell it's not just a bulb problem because I don't have to hit
load addr twice to "move on", the count is smooth and sequential without the
rightmost bit. Any thoughts offhand?
YEs, it's working correctly. Remember that the address is actually a
_byte_ address (the LSB of the address selects the byte within the word
for byte operations). The front panel accesses a word (2 bytes) at a
time, the address is always even (LSB = 0) and increments by 2 after each
EXAMine or DEPosit.
-tony
7078
days inactive
7078
days old
5 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
ggs＠shiresoft.com -
jwest＠classiccmp.org -
toresbe＠ifi.uio.no -
wacarder＠usit.net