El Reg posts dirty pics for old computer buffs Reckon you know your computer history? Well, let's find out. Here's a set of 11 pictures taken from assorted old computers. Your mission Jim, should you accept it, is to identify what's in these pictures and what's the computer or computer device they were found in. I'll follow this article up with another in a week or two's …
Yes and those even have links to descriptions of those computers on them.
So 2 is a Packard Bell PB-250 from 1961 using delay lines.
3 is a PDP-12
4 is a PDP-8/E
5 is a PDP-10
6 is from a Honeywell 316
8 is again from a PDP-10
9 is from the Honeywell again
10 is from a PDP-8/I
Not linked are:
1 probably some dual pentode or dual triode
11 is a barrel or drum printer
So please if you do something like that, don't just use easily accessible pictures.
I remember those from my days as an operator in the 70s. Noisy buggers they were. 120 columns on the ICT one I operated. A hammer came from behind the paper to push it onto the drum forming a letter. The one in the photo is smarter than the ICT one which has a whole row of the same character in line. That meant a line of hyphens used to make a separator on a report all fired at once and could blow a fuse. I'm told a cleverly designed report could play interesting rhythms but I never heard one. An icon for nostalgia is needed.
Brings back happy (?) memories as an operator in the 70s, we had to scrub the barrels with toothbrush and alcohol every 4 hours. Dirty work. Otherwise the print got fuzzy as the letters got clogged with fluff off the inked printer ribbon. If I recall we had about 16 of these printers each about twice the size of a large domestic fridge and despite acoustic hoods on a peak run (when end of month, weekend and bank holiday all come together) they'd all be going hard at it for 48 hours. I think someone said the paper passed through at 3 mph - then the laser printers turned up printing at 30mph!. Yes our drums did have rows of all one letter but didn't blow a fuse. Test prints would print rows of all the same letter so you could spot degradation of the speed of an individual hammer.
Piccie 1) Bag of ECC83 Dual triodes.
2) no idea.
3) That colour's got Data General written all over it
4) for the same reason, "Isn't every computer a DIGITAL computer?" PDP 8 or 11.
5) Can't name the DEC machine.
6) 'Cor', blimey! I had 128Kwords of that! Read, Write and sense are the 3 wires through each toroid.
7) Boot ROM on a PDP?
8) No idea which backplane this is.
9) Remember having to do this kind of work. Did it once, by hand on the back seat of a car (ooer, missus) as I was being driven to Faslane. It was a prototype, we had to design and build it on-the-fly, before we got there! Used a small hand-wrapping tool, and pre-cut wire. Battery had run out of my power-wrapper, so I was envious of my mate, whose machine was good. My wrist ached like buggery after 6 hours of that!
The rest? Can't help, but piccie 10 looks strangely familiar...
Well what's particularly odd is that it's not in a module IBM used to combine tubes and passive components into modules which could be easily replaced:
http://en.wikipedia.org/wiki/File:RoehreIBM_090325.jpg
However the IBM701 doesn't seem to have. It had modules modules just like in that picture. Therefore I believe it's an IBM 701 or a very close relative.
http://www-03.ibm.com/ibm/history/exhibits/701/701_141510.html
Also, if you compare http://upload.wikimedia.org/wikipedia/commons/e/e7/IBMVacuumTubeModule.jpg, where you can see the part number 317261 on one of the valves, to the picture here, or the enlarged version at http://www.rcsri.org/gallery/ibm-tubes.jpg, where you can make out "PART NO. 3...." and "...261" on a couple of the tubes, it's definitely got to be one of those modules.
How about guessing what this is:
http://casandro.dyndns.org/theregister/image.jpg
Particularly note the logo of that company, which already existed before WW II.
Whoever guesses it first gets free entrance to the museum this device is in, and perhaps even the right to sleep on my couch. However I need to warn you, it's in Germany.
Having first eliminated it from being the maching that goes PING from monty python I ventured down the train of thought of it being a early video/audio handerling type of kit and then thought pre WWII and german. I then thought Hmmm was this involved in the Olympic games in germany that was televised and left it at that. Then looked at the logo and took a break knowing I'd tracked it down to one of the sub links of this page: http://www.radiomuseum.org/dsp_sammlungen.cfm.
Hmm, trying to guess there. Meter is calibrated in % and db. % is almost certainly modulation depth. There is a switch for audio, tv & audio, tv. There are a heap of devices that look like parametric equalisers. Given the clues this looks like part of a very early satellite ground station. Given it is in Germany it is possible it was even used for the first Telstar, which would make it a very significant bit of history. Then again the above comment about the Olympic games might mean it was used for Syncom 3, and receiving the 1964 summer Olympics. It could of course have been used for both.
Not bad, yes there are parametric equalisers, they are in groups of 4, and that's a clue.
However it's not satellite releated. Again, it's something, you have seen in a smaller and somewhat cheaper form before. You probably even owned a consumer version of it.
Maybe it's noteworthy that some of those devices needed pressurized air.
professional Electronics Engineer, been doing it for the past 12 1/2 years.. just turned 35 and starting to feel like an old fart with all the young-uns starting now without a clue about what real engineering is about.. but having seen those pics.. yay! I feel like I'm still and at the cutting edge (or at least the bread knife edge.. still sharp and able to cut, but can't quite manage the real crusty stuff!)
Anyway, while I'm please to say I don't recognise ANY of those old computers.. (in todays terms a 50p PIC), I recognise some of the technology and why its out of date :-)
Next time my wife, six years younger than me, gives me crap for being an old fart I'll just point to this article and let her know that I don't recognize any of them. Transistors? Vacuum tubes? Register switches? Heh, right. My first computer was a 8086. What chance do I have of identifying these things.
Anon to keep all you old farts from telling me what a whippersnapper I am. :-P
One of the first jobs I did as an apprentice in 1969 was to wire up what we called 'Handbag Amps'. These were Analogue Computer Amps used on the Gnat Flight Simulators. They were 'mobile'. One Furniture Van for the cockpit and another for the electronics. The term Handbag referred to the large chrom handle (like that on a handbag) that made removing the amp easy. The unit was secured into the frame with two dzus fasteners.
The wiring bays also brought back memories. I spent nearly two weeks wiring up one bay by hand. This was for at SAA 747 Simulator. I made only 9 mistakes from nearly 2k wires. The wiring was deliberately not made neat as to reduce 'crosstalk' between adjacent wires.
The DEC Kit also brings back memories. I spent more than 10 years working for them on all sorts of wonderful bits of kit. Sadly, I can remember the PDP/11-70 Bootstrap secuence once I've consumed a few pints of Brakspears Best hence the 'Anon'.
I'm looking forward to getting a few Raspberry Pi's to automate the '00' layout in the loft.
For the ancient ones, have you:
Used a machine with real core memory?
Used Punch cards?
Made up a drum card for a keypunch?
EBCDIC?
Assembler on 5 different machines (with different instruction sets)?
Entered a program with the front panel switches?
That did some actual work?
Found a hardware bug?
Fixed the hardware bug you found?
Paper tape?
ASR 33 Teletype?
Acoustic modem?
Bootstrapped the operating system from the console with out using the "boot" function?
Build up a microprocessor system from the raw parts, and wired it yourself?
Replaced a defective core memory stack with a working one (soldering it yourself)?
<<<<<The list abounds, as I'm sure others will add entries!>>>>>
Almost all!!! (Exception: Drum card)
H/W fix?
A real H/W bug fix was discovering the setup-and-hold times of a D-type were violated sometimes. The D-type flip-flop would possibly go metastable, i.e., both outputs (Q and !Q) going to the same level, or even half-rail. The stupid fix (that is still suggested in some textbooks ) is to put another D-type before the original...
OK, Fixes the original D-type's problem - except it has no clue what to do about it's new friend when it goes metastable...
The solution to metastability of putting two (or more) D-types in series is not a stupid solution.
If you have an asynchronous input which needs to be used by a synchronous system you have to synchronise the input somehow. Depending on the clock frequency, rate of change of the input, metastable window and the metastable lifetime you can calculate the probability of a metastable, oscillating or invalid output from the last D type and you can then choose sufficent D-types to make that probability sufficently low to meet the reliability requirmenst for the system.
This is how I have always seen it done and i can't think of way of avoiding it. I have seen people try but they end up creating their own latches out of async logic whch is really just the same but with worse and ill defined characteristics.
There should be no setup and hold time violations from the synchronous part of the systems. That is a design mistake.
Paper tape?
ASR 33 Teletype?
Acoustic modem?
How else would you transfer your Fortran from an ICL 1903A to a Dec 20?
You print out you ICL program onto paper tape.
Attach the ASR33 to the acoustic coupler.
Ring up the OU's Dec 20 and then shove the handset into the acoustic coupler.
Feed the paper tape into the ASR33 and upload the program at 300 Baud.
Error correction, you ask? You disconnect the ASR 33 and connect the
VT100 and eyeball the Fortran looking for any missing commas or
whatever. Fortunately most errors tended to be bursty, but not all...
Geoff
Picture 5 is certainly, as it proclaims, a KL-10 (A or B model) which is the CPU unit of, in this case, a DECsystem-10 which was the blue variant running TOPS-10, as distinct from the orange version which ran TOPS-20 (DECSYSTEM-20) - (KL-10E). KL-10 based DEC-10s also were different from DEC-20s in that they booted from DECtape where -20s booted from RX01 (5 inch) floppies - in both cases using a PDP-11/40 as the primary front-end computer, and -10s had external memory busses while -20s had internal memory. (Exception for the purists who remember these things, the 1091 was indeed a blue 2060 and did boot from floppy - I think that there was also a -1095 with the same relationship to the 2065.)
If picture 8 isn indeed the central backplane wire-wrap of a KL-10, then I have an enduring memory of having this Field Replacable Module' being replaced not once but twice on a DECSYSTEM-20 (2060) in the mid/late 1980s for ICI on Teesside. Shipment on a pallet some 3m x 1.5m x 2m as I recall!
In addition, Digital eventually shipped us a 2065 (and then downgraded it to 2060 because I'd never go round to upgrading the OS - failing to maintain n-1 is nothing new!) diverted from initial installation to keep us running until the main machine was eventually fixed.
PS: If anybody wants a picture of it, I have part of a KI-10 control console in my garrage!
Absolutely brilliant.
Congratulations to everybody who got the answers.
Just one tiny question - how can you be so superb at getting the answers and so dumb you don't understand the simple instructions "Email your entries to me here", and "post any public comments about this in the Reg Forum Picture Puzzler topic?"
We don' need to read no stinkin' documentation
or
I didn't get where I an today by reading the instructions
I must admit though Kwac, I can only conclude that at the very least its been an awfully long time since you worked in end user support if you honestly believed this lot would follow those simple instructions... You don't show off how clever you are and massage your ego by sending emails... Especially as its so much more fun to spoil the fun for everyone else.
There is a need here to clear the air. Several posters have taken exception to the posting of these pictures because they are available at www.rcsri.org, the website of the Retro-Computing Society of Rhode Island. The implication is that The Register snagged, scarfed or otherwise somehow misappropriated these images from us. To the contrary, the RCS/RI has been working closely with Chris Mellor of The Register to make these images available to the wider world; to those folks who would not otherwise know of the RCS/RI. We are rather delighted to find out that we are already better known than we'd expected, and we hope that this article will make us even better known. The RCS/RI is an all-volunteer private organization, funded by its members (Although if you'd like to kick in a few bucks, let us know!), that collects, refurbishes and operates antique computers, generally but not exclusively associated with New England. We feel that the preservation of our technological heritage is well worth the pain and trouble that doing so requires. It's also a lot of fun. We have an Open House on the third Saturday of every month at our facility at the Atlantic Mill in Providence, RI, and we invite you to come visit our collection. And thereby to see for yourself what's in these pictures. Details may be found on our website, www.rcsri.org.
-Geoffrey G. Rochat, Member, Retro-Computing Society of Rhode Island
And Ma Bell used to use the terms longitudinal and metallic when they meant differential and single-ended, and measured resistance in feet. The several computer companies that I've worked for, Back In Ye Olde Dayes, had as engineering policy to wrap insulation around the post in what was known as "modified wrap," and would reject wire-wrapping jobs otherwise. So what you're seeing was mostly likely intentional.
There's just no accounting for taste.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
