Tony Duell wrote:
I also started out learning computer hardware on
the PDP 11/45 at Maynard.
The only thing hard about the class was trying to figure out what those
You went to a class? I had the machine, the printsets and a logic probe.
And no list like this one to ask for help when I got confused...
I was sent there by the company I worked for at the time in preparation for
starting up a new bakery (just happened to be the largest one in the world
at that time.) It ended up that I became rather proficient at repairing the
PDP 11/05 and the PDP-16s and never worked on the 11/45s at the plant.
One cute (to me) story. I knew the 16s pretty well, but I was switched to
another area and another foreman took my place. Well a problem developed
with one of the computers and I was *told* not to take care of the problem.
After a week of downtime, I was told to go up and see what the problem was.
About 5 minutes after that, everything was back to normal. I had left
specific instructions that the machines were not to be exchanged out except
when a power supply went bad. A computer went down and the new foreman told
the tech to change out the machine. Apparently the foreman was not well
respected, and the tech just did as he was told (so I found out later when I
asked the tech why.) A number of cards in the PDP-16 were specific to the
location the PDP-16 was installed, and changing the computer out took these
specific cards (constants cards) and put them on the bench with who knows
what in the computer at that point. A quick change of cards solved that
problem.
hyroglyphics
were that described somehow the internal workings of the thing
:). Hardware repair was pretty much limited to putting tape on one of the
board gold fingers so it didn't make proper contact when the board was
reinstalled.
Eh? I can understand why that might introduce a 'deliberate fault' which
you'd have to find, but I can't think of a diagnostic technique where
isolating odd pins of the 11/45 CPU modules would be that useful.
An understanding of how the machine worked was what was being taught. Being
able to find and define what wasn't working was one way of showing at least
some understanding of how it worked. We only ran into trouble once when we
taped the wrong pin and put the card back in. We ended up creating a
problem with the microcode execution, and that one proved a bit difficult to
locate (took us about 45 minutes or so to find the problem.)