On 2016-May-24, at 8:38 AM, Swift Griggs wrote:
Since I'm an igmo about most machines before the
mid-eighties (and still
fuzzy even on most of those), I'm curious about all these older machines
with front panel buttons and switches. What all did they do?
Primary front-panel operations were:
1 - examine and load memory locations
2 - examine and load registers, including the instruction counter
3 - start and stop processor execution (run state)
Thus with (2) you can start execution where-ever you want in memory
Also, depending on the machine:
4 - single-step the processor, sometimes per instruction, sometimes per clock cycle or
instruction phase
5 - twiddle I/O devices
You could
actually program them using the front panel right?
Yes
Some of them
bootstrapped this way, too?
Yes. In the core days, the common thing to do would be to use the front panel to
'toggle in' the first bootstrap loader of some few dozen instructions/words.
Core being non-volatile, that bootstrap program would continue to reside there across
power-down/up so you didn't have to re-toggle at every power-up.
Just set the execution-start address with the toggles, and run.
What kind of "language" was used for that
(ie.. what were the basic mechanics)?
Raw machine code. Typically expressed in octal or hex depending on the machine instruction
layout or policy, and translated to binary by the operator as you toggle in.
Did the buttons ever change color?
Were you considered a badass if you had switch
flipping all memorized down
to an art?
Were they mainly multi-position toggle switches or
on/off
buttons?
Most common was two-position (up-down) toggles.
Some machines had rotary switches for such as register-display selection
Some used momentary-push-buttons for 1-0 bit cycling.
They just seem to be a lot more important on older
mainframes and minis.
Also, what was the main reason for the blinkenlights? Was it to show
system load or specific system states?
Primary intentions were:
1 - initial bootstrap as mentioned above
2 - hardware diagnosis and servicing
Earlier and larger machines might have the state of every or near-every flip-flop in the
machine brought out to an indicator,
so you could see the entire processor state on the front panel. You could then use that
to trace logic faults in the machine.
Simple example: if the carry flip-flop lamp isn't turning on after single-stepping
through an ADD instruction with registers
loaded with operands which should result in carry . . .
3 - (less common) program diagnosis , i.e. used as an instruction-level software
debugger, if one had monopoly machine access.
The improved reliablity of LSI logic over discrete and SSI, and the creation of ROM chips
of reasonable capacity (to hold the bootstrap or a monitor), would bring about the demise
of the blinkenlight front panel.
Note that only a couple of the first microcomputers had blinkenlight front panels, and
they were pretty much gone from minis and mainframes by the late-70s.
Just curious. I'm learning a ton from reading
these threads on older
machines, but there is so much I don't know.