I know what you mean about tech advances: At TI's SRDL in 1965, I drew my first ECL IC layout on grid paper, copied the coordinates off onto an 80-column form, keypunched the cards, used them to produce a plot on a CalComp plotter, and then took the plot to the "Zip-cutter" guys who used a manual X-Y rig to cut the "Ruby-Lith" -- and then peeled off the red stuff by hand.
Then I had to check their work by eyeball (patching mistakes with red tape) -- without getting fingerprints on it. Then it took several weeks to get masks back from the mask-making vendor...
(I'm not going to admit how many times I had to re-punch cards before the CalComp plot finally matched my original...) ;-}
Again, thanks for "sorting me out!" '-)
Ah, us old war horses can swap stories.
I don’t go back quite as far as you do, but I did spend the first few years of my professional career using a cardpunch to write COBOL 68 programs (64K manual segment swapping — and we were HAPPY TO HAVE IT!).
Then we got *gasp* TP monitors! And you just hit the RETURN key, you would get an error since the TP monitor handler (this was an RCA Spectra/Sperry Univac 90-series system — no TSO for us, thank God) thought it had encountered a blank card!