I hope I will see it working in my lifetime!
Although to be honest, to "understand how computers work" you only really need an abacus. Less fancy, but also much less expensive!
A project to build British mathematician Charles Babbage's mechanical computer has won assistance from the Science Museum in London. The museum has begun digitising Babbage's plans and notebooks so that John Graham-Cumming, the programmer and computer historian behind the project, and his team can begin work. Babbage first …
In the case of the abacus the program is "entered" by the user's fingers, not punchcards, but it is perfectly possible to demonstrate how computers work (including some pretty complex notions of programming) using only an abacus. Some people even build binary abaci especially for the purpose of teaching how binary computers work, although with the right conception you can manipulate pretty much any set of conceptual objects with an abacus.
Calling abaci "mere calculators" is an insult to both abaci and calculators. Although an abacus can be used for calculation, it is only an abstraction tool; the "calculator" would be the system [abacist abacus]. The same system can be a full-fat -although not very fast- "computer", and more, with the right abacist. The abacus is not limited to binary, either, which is very interesting for CS teaching purpose; although it is not possible to model a real quantic computer at the iron level with a reasonnably-sized single-frame abacus, you'll note that Babbage's machine is not more helpful in that case.
Actually it might just be possible to model a quantic computer using a multiframe abacus, I'll give it some thought.
Bits of the Analytical Engine have been built in Meccano and also in Lego, but nobody has yet built the whole thing (that i know of).
As for differential analysers, ie analog computers, Meccano is a favourite material for these, Barnes Wallis even had one for helping with his Bouncing Bomb calculations.
... is that of heat transmission.
One of the "computing engines" in the Science Museum's collection is Carter's Ringing Machine. This was devised circa 1900 by Mr Carter, who was a Birmingham Bell-ringer, and -- for afficionados it was capable of ringing Stedman Triples on hand bells via an electro-mechanical linkage. The problems was that with the original steam engine, heat was transmitted via the drive shaft into the system causing sufficient expansion for the tolerances to be exceeded and the machine to seize.
In more recent times, the machine has rung a "full extent" of Stedman Triples (that's all 7! = 5040 changes or about three hours), using an electric motor as power supply. It's worth a look if you can get the curator to give you a private demonstration.
Youll need the upgrade. That version has two boilers powering it, providing Steam Linked Injection. Each boiler has two fireboxes and the whole thing is arranged so that the four fireboxes are in a cruciform arrangement to maximise space for stoking, known as four-way cross fire.
Coming soon from Antique Macro Devices......
.. it was lack of government, or other, funding that shelved both major projects. Had funding and foresight been in abundance at the time, the machines could have, amongst other things of course, calculated highly accurate ranging tables for the navy etc and we would have really, really have been ruling the waves then.
I have not looked at the actual specs, but I'd be very surprised if it can do *any* floating point operations, let alone 3 per second.
Ok, sure you can probably emulate a flop ... once per week.
I think performance should be measured in MIPW (Million Instructions Per Week).
Go steel chains. Strip a few hundred of those no-muffler-noisy-two-strokes motorcycles off their chains.
Rubber belts for power transmission require tension, which force the bearings over time. Chains eventually need spring fasteners for long distances, but wear out at slower rate, with proper application of axle grease. At a plus, chains provide synchronicity, should that ever be needed. Plus they agree with high temperature, soot... as long the grease is there.
That would take away the heat transmission problem and make it a 'green' computing initiative. You could probably govern it so the water cooling at one end counteracts the heat from friction in the gears. It might be worth investigating running it in an inert atmosphere to avoid corrosion and bugs.
Biting the hand that feeds IT © 1998–2019