Power Consumption: 1.5kW
Size 2m high x 6m wide x 1m deep
Weight: 2.5 tonnes
Number of Dekatron counter tubes: 828
Number of other valves: 131
Number of relays: 480
Number of contacts or relay switches: 7073
Number of high speed relays: 26
Number of lamps: 199
Number of switches: 18
I believe, when they first turned it on and it powered up and booted after not really having been booted for decades .. this is a fair spring, dear fellow..
Is the poetry of talking of systems that spring back to life without a focus on the 3 years of human effort all that different from the poetry of cat walk models, photoshop body images, etc?
Poetry can be great, but not when it enforces unrealistic expectations. In technology we are not above the limerick.
I found it interesting that they compare the ouput to a man with a pocket calculator when the people these devices replaced were overwhelmingly female. I can remember reading respected mathematicians approximating the problem-solving power of early computing machines in “girl-years” and describing machine labor in “kilo-girl” units.
The sound is great. Its a shame when you walk into a server room now you don't hear anything except the AC running. It just doesn't have the same feel.
"At a time when iPhones are swapped out every 12 months, the TI-83 calculator and this computer are among the only pieces of technological machinery that have survived for decades."
The TI-83 with a graphic display is recent by comparison to the venerable HP12C.
It had a built-in watchdog with automatic restart of the program: "The normal alarm gives a warning if the computer is inactive for 30-60 seconds. The delayed alarm returns the computer to the beginning of the programme when a period of inactivity is detected if the inactivity is not due to either Signal order."
Rudimentary 'stack trace': "When the computer stops owing to a programme error on a faulty tape it is useful to know what order is stored in the control section of the computer or the address from which an order is being read. Five keys on the Control racks switch in a lamp display showing this information."
Programs were written, debugged and staged on paper tape, then went into production: "The normal procedure is to make a paper original using the keyboard perforator. From this a printed slip is produced and checked against the manuscript programme. Any errors are marked up and if necessary a corrected paper copy is made. The paper original or copy may then be proved in the computer and any further corrections inserted. Where the final loops of tape are required to pass through the computer more than a few dozen times linen copies are made from the paper originals."
Typing errors on the tape were corrected by backspacing over the error, then punching all holes open. I guess the reading unit recognized this as a NOP or NULL character.
Deployment of programs were done by glueing both ends of the tape together, so it formed a loop: "Splice the tape into loops by cutting the ends to a blunt point, allowing an overlap of two or three rows. Coat both ends sparingly with thick paste on one side and join, with the leading end on top. The joint should be lined up carefully to ensure that there is no discontinuity at the edges and that the feed holes are clear. Press the joint together firmly and dry thoroughly before use. Avoid excess paste on the linen tape as the starch impregnation is easily removed by moisture."
The chief reason (at least in the US) is probably due to standardized testing having a white-list of allowed calculators. Your phone can emulate the TI-89 just fine (and run other tools like LAPACK, PARI, Octave, etc.) but good luck getting it approved for a test.
I was chatting with my dad about this today. (He teaches classes that use graphing calculators.). It's cheap to buy a calculator emulator app so they must allow it soon. Though students sometimes use the camera and email to cheat.
Anyone know how this really could be useful? If it took the thing 5-10 seconds to multiply two numbers, and it also needed input as of punch cards. How could it at any point be faster than doing calculations manually?
I'm genuinely interested in how it was really used, any hints?
The BBC article claims that its advantages over humans were that it didn't make mistakes and would happily churn on for quite a while, with a running time of up to about 80 hours per week.
I don't know how large the numbers were that needed 10 seconds to multiply, but even if it is not much faster (or even if it is slower) than a human for a single operation, a human won't be able to reliably do arithmetic over any comparable time frame.
It didn't use cards, but paper tape (as seen in the BBC clip). So I guess, in theory, it could have gone on forever if you just glued a fresh tape to the end of the old one.
>"It was, of course, slow, not much faster than hand calculation on single operations, but fully automatic, extremely reliable and utterly relentless. It took little power and could be left unattended for long periods; I think the record was over one Christmas-New Year holiday when it was all by itself, with miles of input data on punched tape to keep it happy, for at least ten days and was still ticking away when we came back."
It 'sprung' back to life after a three year restoration project.
The original article is a much more appropriate read: http://www.tnmoc.org/news/news-releases/worlds-oldest-origin...
The tech spec is brilliant: