2 ms·

My first paying job as an undergrad was maintaining a remote job entry system running on a PDP-11. It was a revelation that the instruction set was simple, orth

by Zigurd 3y ago

My first paying job as an undergrad was maintaining a remote job entry system running on a PDP-11. It was a revelation that the instruction set was simple, orthogonal, and made sense. Computers could be elegant! Later I had a full time job maintaining an driver written in C and 68k assembler and that PDP-11 experience was a big help. 32-bit x86 by comparison looked messy and weird, when I had to grok that for a gig ploughing through an impossibly large pile of Mitch Kapor's code at Lotus trying to track down and fix bugs.

inetsee 3y ago
My first job after graduation was programming and running a PDP-8 in a Psych lab (my degree was in Psychology with a minor in Computer Science). Data transfer was interesting from today's perspective. I would have to print out data to a teletype machine with the paper tape writer turned on, then take the paper tape to the computer center (where the Burroughs B5500 lived) load the paper tape into a teletype machine connected to the mainframe, and after that the mainframe would output the data on punched cards. After all the typos were corrected on the punched cards, the Psych Grad students could run their data analysis using programs also on punched cards. One interesting episode I remember to this day was when we got some error messages from the PDP-8. I ran a diagnostic that said the one of the boards in the computer was experiencing intermittent errors. I ran a diagnostic which said yes one of the boards was failing. We had a maintenance contract, so I called DEC and they sent out a repair person. He ran the diagnostic again and confirmed that a board was failing. I expected him to just replace the board. But No! He pulled the board out of the computer, unsoldered a transistor, and soldered in a new transistor. After running the diagnostic to confirm that the problem was fixed he went on his way and we were back in business. Nowadays if you get a hardware failure you have to replace an entire Motherboard, or a memory board.

by Zigurd 3y ago

inetsee 3y ago

My first job after graduation was programming and running a PDP-8 in a Psych lab (my degree was in Psychology with a minor in Computer Science). Data transfer was interesting from today's perspective. I would have to print out data to a teletype machine with the paper tape writer turned on, then take the paper tape to the computer center (where the Burroughs B5500 lived) load the paper tape into a teletype machine connected to the mainframe, and after that the mainframe would output the data on punched cards. After all the typos were corrected on the punched cards, the Psych Grad students could run their data analysis using programs also on punched cards. One interesting episode I remember to this day was when we got some error messages from the PDP-8. I ran a diagnostic that said the one of the boards in the computer was experiencing intermittent errors. I ran a diagnostic which said yes one of the boards was failing. We had a maintenance contract, so I called DEC and they sent out a repair person. He ran the diagnostic again and confirmed that a board was failing. I expected him to just replace the board. But No! He pulled the board out of the computer, unsoldered a transistor, and soldered in a new transistor. After running the diagnostic to confirm that the problem was fixed he went on his way and we were back in business. Nowadays if you get a hardware failure you have to replace an entire Motherboard, or a memory board.