Tech #27: How A Computer System Works (1975)

I thought it might be fun to take a quick tour through a book on computing from 1975.  There's just something about those gigantic whirring boxes that makes my heart beat faster.  How A Computer System Works is fully illustrated and a wonder to behold.  Let's take a walk through the pages of antiquated technology, shall we?

It's amazing to think that your cell phone can do more than this behemoth could ever dream of.  Plus we have our first miniskirt sighting, so we're off to a good start.

CRT = cathode ray tube


Here's the Large Scale Univac - basically a maze of cabinets.  I'll bet that room was loud.  Whoever sat at that desk probably felt like Captain Kirk.

The early seventies: high hemlines and huge computers.  It's amazing to think that we shot people into space with this technology.

This system would have trouble handling the data from this single blog post.

"Characters of Information" is the antiquated way of saying "bytes".  So 8.4 MB.

I don't think these pens worked too well.  It would be another three decades before the touch screen really became a part of everyday life.  I'm glad to see he's got his ashtray situated nearby.

Early "green screen" terminal connected to a mainframe.

Link to the newspaper article about this new device

When the photo shoot was over, he went back to playing solitaire.

This is just too awesome to put into words, so I won't even try.

I miss the dot matrix printer.  There was something cool about the way it struck those dots line by line right in front of you.  Laser and Inkjet are so "hidden" - the page just blandly spits out fully complete.  Dot matrix put some elbow grease into their printouts.

I sure hope there's no mistakes on that copy, because you'll have to type the whole damned thing again.  Typos could be devastating back then.

If I walked up and found this going on in a telephone booth, I'd think some crazy shit was about to go down.

One invention I could do without is the autotune.  It makes even the lousiest singer sound on key...... of course, it also sounds cold and inhuman, and has contributed to the destruction of the music industry as we know it.  But I digress.

I remember certain software required multiple diskettes for installation.  So, let's say you had Aldus Pagemaker in 1987 - you were looking at maybe 15 disks, each with a label "1 of 15", "2 of 15," etc. Your 386 would huff and puff with each one, making the whole thing pretty time and labor intensive.

I think I've heard the first item ever to be scanned at a register was pack of Wrigley's chewing gum.

Forget the microfiche, how about that micro-mini!

I sure hope these ladies weren't under the impression technology was going to make their jobs easier.

He's not driving while working that thing, is he?  Seems a bit cumbersome.

Well, that's all from this great book.  All vintage technophiles out there, please chime in on any of these devices.  I'd love to hear some info.


  1. Ahh, yes, the card punch reader! That is how I learned to program computers - one bloody card at a time. The upside is that the little punch outs made great confetti!

  2. AnonymousJune 12, 2013

    NCR Century computers - ahh - good memories - another great blog post

  3. That last shot reminds me: We were watching Scarecrow and Mrs. King and Scarecrow had a computer in his car! A whole stinkin' computer with CRT and everything! And he was communicating with the home office!

  4. AnonymousJune 12, 2013

    My job was to fault isolate and repair such large systems down to the individual bit-level; I mean trace and manipulate individual bit(s) in the CPU(s), memory, and disk, actually repairing the bit's faulty logic. What they don't show you in the large systems is the floor is raised to allow massive cabling required to interconnect every one of those devices to their controllers and power - a veritable root-covered forest floor of cabling, each often several times larger than a garden hose, snaking to every device at least once, often many times. The floor was designed to pass massive quantities of extremely cool air into each cabinet, often generated outside by huge air handlers, forcing it through huge analog power supplies and past all the extremely hot logic chips with even more blowers. If you lost cooling for just a short time, you could end up destroying an entire cabinets if the automatic power down malfunctioned (happens) - if you were lucky just one chip would blow its top off from thermal stress; not so lucky and the entire boxes semiconductors could become very flaky, crashing randomly until sorted out or even scrapped in extreme cases. It was so noisy and cold! all that air being moved around. Dot-matrix had a distinct sound, but so did those Line Printers - so loud they were enclosed in large cabinets, which you see in the pics. Some of the fastest had steel letters embossed on heavy metal belted chains which whirled around in a blur by a half-horse motor, aligning just long enough for a fast electromagnetic hammer to strike it onto the page, every column which required timing adjustment. I could go on and on, but it was the end of the common mainframe era, and I was one of last who were trained by the old school just as the PC era began. Heavily involved with HW and SW during the PC renaissance, the change was not traumatic for me, just bittersweet. S-A-H-D

  5. I was thinking that the NCR disk might not be 8.4MB, because a lot of early mainframes used 6-bit character sets and "packed" 5 characters into 4 bytes!

    That said, I looked up the NCR and it was ASCII (modern 8-bit-byte stuff) so it's likely that 8.4MB is right :)

    Still... these days if you had a USB stick of 8.4MB you'd probably throw it away - too small! :-P

    Nice post :)

  6. I first took computer programming in 1970. You got your assignment and wrote out the program by hand. Then you stood in line for a keypunch machine so you could type it onto cards, one line per card. Even a very simple program could have hundreds of cards. On the way to the computer center you would drop your deck, which is how you learn to number each card in order while you're punching them. At the computer center you handed the deck over to someone who would input the cards into the computer. After awhile you'd get back your cards, and a thick wad of computer paper which would say something like "INPUT ERROR" on every page. If you were lucky, there would be something in there that would tell you where you made the error, most likely you forgot to separate your input data by commas. So you'd repunch your data cards and try again. The next wad of paper would come back with maybe one number on each page. That was your indication that you'd left an open do loop somewhere. And oh my god! This assignment is due tomorrow and all the key punch machines are broken!!!

    1. AnonymousJune 15, 2013

      Our university shared our mainframe with three others in the 70's. To accommodate the load, they limited each student to a whopping four runs per day (FOUR!). This made you very, very careful about your cards. So in addition to dealing with lines at the punch card machines, crazy program logic rife with GOTO's, you only had four chances a day to get it done. I remember sitting down one afternoon after having blown thru two of my four runs with silly syntax errors and pouring through my cards for and entire hour 'desk checking' the cards looking them over to make sure that they were pristine before I went for run #3. Kids today learning programming with their auto correcting, compile as you type IDEs do not have a clue of how great they have it.

  7. 1981 - my first real job - IBM 360 - 16 mb of memory! And a whole room full of disk drives, tape drives, and air conditioning! Does anyone else remember what a write ring was? I still have my collection of different colored ones in a box somewhere....

    It actually turned into the start of a varied and interesting career--and I credit much of it to taking typing in high school!

    1. Funny you should mention write rings. I was just digging through a box tonight and found a half dozen I took from my first computer job, working on an IBM AS/400.

      And yeah, typing class was the best thing I ever did. "Ready… begin."

  8. You know where else the technology has improved? All those chairs those people are sitting on. I had an old four-wheeled banker's chair... nostalgia junkie that I am, even I tired of having this uncomfortable, heavy, unstable thing that would tip over and fling me backwards to the ground at the slightest adjustment. The five-wheeled chairs are lighter, much more stable, and far easier to adjust up or down. Also: armrests.

  9. Man those ladies are as great as the computers - brilliant work as always!

  10. We had a room like this in high school, believe it or not, in the early 80's. I remember feeling like I was part of the future. But then the massive power of tech that should have given us a Sci-fi life became Syfi with all that potential is used to show how a poor human you can be in a Dunkin-Donuts youtube video.

  11. Fig. 60 - Good luck *finding* a phone booth these days.

  12. I too miss dot matrix printers. Ribbon running dry? Just run the sheet of paper through a couple more times and it'll be legible, unlike ink jets or lasers. Wanna run off a 6' banner for Judy's birthday? Only tractor-feed paper will do that.

  13. AnonymousJune 18, 2013

    Wow, I wasn't even born when all these fascinating devices were in use. But I am old enough to remember those dot matrix printers. The sound they made was very cool, and printing band posters and banners with Metallica logos on them looked real geeky and pixelated :)

  14. AnonymousJune 24, 2013

    The printers are most likely chain printers rather than dot matrix. Digital Equipment Corporation made a dot matrix printer, the LA36, and it was fast at 180 characters per sec. Centronics made a slower dot matrix printers, but it was less expensive that the chain printers. But work horses in terms of reliabilty and speed in 1975 were the line printers using chain printer technology.

    Chain printing consisted of a continous moving chain of characters passing between a gang of hammers positioned at each character position on the printer and the paper. When the proper character was in front of the hammer, the hammer struck and printed the paper, when the characters all passed by the paper rolled forward one line, and then the next line was printed.

  15. Thank God cops used to drive those Fury III "land yachts" so they could use the "latest technology"!
    My best friend's dad loved the Fury models and they were HUGE! The hood alone was about 6 foot wide!
    Drank gasoline like a pig.

  16. The Phone Booth Gadget advertising copy only tells half the story - the terminal speaks because the handset is in use, nestled into the built-in _acoustic coupler_. This was the shit: Bytes converted into audio tones and transmitted as a fast-warbling whistling noise! Data transfer rate maybe 300 baud with a tailwind, but that was better than nothing and any phone could become an “ethernet cable,” so to speak.

    Some years back, I picked one up on eBay - it didn't work but I didn't care; I parked it next to my PC with my (working but not in use) rotary-dial phone beside it, and happily went on-line. All was set right again.

  17. where is the front panel for the Gould/SEL 32/55 or 32/75 ??

    with punched cards remember the rule, "fan and joggle the cards, 9 edge... " what? how does the rest of that go? (worth 25 points)

  18. Anyone have any photos of the Gould SEL 32/55 or 32/75 computers?

    ... Punched cards, remember the rule: fan and joggle the cards. 9 edge ___ ? (25 points)
    a. Eagles
    b. Juggles
    c. Mark Lipinski
    d. other (specify)

  19. Jim SweeneyMay 13, 2016

    I started programming in 1960 on an IBM 705. It was so big and expensive that we were not allowed to hook any card or print equipment to it. Everything had to go in via tape and come out on tape, using a 1401 to convert the cards to tape (for input) and to print the output from tape. It was a tube monster - imagine the cooling and raised floor cabling needed.
    But my first (computer) love was the IBM 1401 (the first IBM transistor computer), we had three of them to handle small jobs and the input output from the 705.
    Then we got rid of the 705 and went to two UNIVAC Octal 30 bit word machines (UNIVAC 490 series) - actually a commercial version of a system built for the military. We wrote a "real-time" program on one to handle queries from terminals in various offices and used a second one for "batch" processing.
    At night I moonlighted for SW Bell, programming their 2 IBM 1401's. Half the time programming everything in machine code because they didn't have the computer time to do assemblies on. They had a goal of all systems having to run 90% running time. I would program a loop into the systems from the CPU console so the poor operator (2 system 5 tape drives 2 card reader/punches, and 2 600 line-a-minute printers (and she was supposed to keep both systems running 90% of the time - only 1 operator).
    In the video at the top, I also worked on the UNIVAC 9400 pictured (a IBM 360 compatible system developed in Blue Bell, Pa). All in all, had a great time for 36 years doing applications (280 programs for Bell) and system software for various UNIVAC/UNISYS systems
    They truly were THE GOOD OLE DAYS. Jim Sweeney W3DCT

    1. Fun stuff. I've alway liked UNIVACs, mainly because of the name. :)