I'm fully supportive of this
Too many UK achievements get lost in time and get shipped out due to lack of money.
Bletchley are doing a wonderful job on very little money, I applaud their determination and enthusiasm
In a project described as "the computing equivalent of the raising of the Mary Rose", engineers at Bletchley Park intend to restore a 1950s-era computer - featuring a magnificent 112.5 bytes of memory* - to working order. The machine in question was built at the Atomic Energy Research Establishment at Harwell in Oxfordshire. …
Interesting article but may I make a minor suggestion:
"but also featured 900 gas-filled tubes" - in the UK, these devices are known as valves (they're only 'tubes' in the US).
God knows where they're going to locate these damn things; it's hard enough locating them for power transmitters (where transistors / FET, et al still don't cut it ) - getting them for systems that the transistor took over in the 70's will be a lot more fun.
When they do get it working, is there any chance the UK TAX office could borrow it for their calcs.?
"featuring a magnificent 112.5 bytes of memory"
I first read this as bits, not bytes, and immediately wondered what half a bit was...
Does this imply that its word size was only a nibble? I thought that way-back-when, a byte was whatever size the hardware defined it to be?
Still utterly confused, but less so than half a bit.
But yay to Bletchley. Please build it out of Meccano(c)
"but also featured 900 gas-filled tubes" - in the UK, these devices are known as valves (they're only 'tubes' in the US).
Indeed they are known in the UK as Valves. You missed the obvious boner though. They are not 'gas filled' They are also known as (and here's a clue) VACUAM tubes. Right ...they are 'filled' with nothing but vacuam - and plenty of it...(guffaw!).
Thermionic emission devices. When I were doing techy stuff I made 'em as a project for my OND. All the best Amplifiers used them. I used to have an old cinema amp that you could play bass through and REALLY honk off the neighbours.
Curiously the cold cathode tubes, are still called tubes in the UK. They are not switching elements, and are not therefore valves. They are not thermionic either. Even Mullard - who no doubt made the ones used in the computer - call them "tubes" in their own literature.
One assumes that the computer is largely intact, but given the passage of time probably not safe to power up. No doubt lots of icky perished rubber insulated wire, dried out cpacitors, and the accumulated grime and dust of the ages. Relays will be the same as used in telephone exchanges of the time, and used in Colossus, and relatively easy to rebuild or source. The memory tubes will probably be fine since there is essentially nothing to go wrong, unless they leak. Being cold cathode, they have no heaters, and don't run hot.
I wonder, did Thommy Flowers design the beast? If ever there was an unsung hero in the history of computing he is it.
"the computing equivalent of the raising of the Mary Rose". OK, without the giant crane, massive diving and engineering effort, not to mention lousy weather and bloody cold water. Perhaps a little more like reconstructing the Mary Rose...
Still, a great project and I hope Bletchley gets the funding it deserves. Good luck to them.
As I understood it from the BBC version of the story, storage was 900 dekatron tubes, which could each hold a single digit. Most dekatrons would store 1 of 10 states, so it's a single decimal digit rather a binary one, so the memory could exist in 10^900 states, which in binary terms can be represented in about 2990 bits, or 373.75 of our modern day bytes.
I spotted a minor error in your report: according to the BBC story, the tubes are decatrons i.e. they have 10 stable states. Therefore, they store just over 3 bits per tube, not one - those are thyratrons.
They're also quite rare as they are prone to leaking or going bad in storage. I wonder where such a large quantity of presumably good ones was found.
- A. Pedant
First the correction....
The Harwell computer is being restored by the 'engineers' (actually volunteers) of The National Museum of Computing which is based at Bletchley Park, rather than Bletchley park itself.
Now the info...
The 'tubes' are Dekatrons and the system has 90 (not 900) of them for it's main storage. Each one is capable of storing a decimal number 0 -> 9. It uses loops of paper tape on 4 readers to load the program and access subroutines and can store intermediate numeric data in the Dekatrons with the final results printed out or punched to paper tape. All the original Dekatrons are still with the system and it also has a number of spares for various parts of the system. How well the Dekatrons have survived is unknown. We already have a stock of Dekatrons so should have enough to replace any that have failed.
It wasn't particularly quick in operation, taking 2 seconds to add two number and over 15 seconds to do a division but it was reliable, running for over 10 days one Christmas / new year period.
It is likely to take the best part of a year to restore and will be viewable at TNMOC throughout its restoration period - http://www.tnmoc.org
Yes, Russia and China, in fact. China manufactures and sells exotic valves that we no longer make here or in the rest of Europe. They also tend to be cheaper than the Russian equivalents.
However, I still have a civilian wartime receiver c.1940 and I can still get valves (Mullard) for that no problem. Pricey though.
I have a very fine collection of mini/micro hi-speed computer valves from the 50's which still work.
The counting style of valve, along with the massive thyratron switches used in the old high power pulsed radar, are not vaccum "filled" but have a low pressure gas (hydrogen, argon, mercury vapour, etc). Once conducting, they stay on until the current drops to a low level, which is useful for some jobs such as digital counters.
Take this from a 67 year old. In England they were called "valves"
not only in my time but also in my father's time. Until recently
only the US called them "tubes". The advent of the US dominated
Internet now means everyone under 35 calls them tubes.
I still have some EL34s and EL85s for a Marshall 100 watt bass
amplifier which I used in the 1960s. They are still in their
original boxes and on the ends and sides of the boxes it plainly
states "electronic valve".
How about this picture (from an American company)
"I spotted a minor error in your report: according to the BBC story, the tubes are decatrons i.e. they have 10 stable states. Therefore, they store just over 3 bits per tube, not one - those are thyratrons."
You know what decatrons are? Looked at end-on you see ten little neon dots in a circle, one of which is glowing at any time. When you apply a pulse the glow moves round to the next cathode. At the end of a complete circuit there's an output pulse to trigger the next decatron. Thus they are capable of counting in decimal and do not in fact store any bits (binary digits) at all.
Don'd confuse decatrons with numicators which display numerals 0 to 9.
still have one - bought when I was a nipper for -IIRC- 5 shillings, i.e., two weeks' pocket money.
Built it into a super-regen 0-V-1. Took it out, but I reckon if I fired it up now, it'd still be OK. Always worried about that plastic bit to the cathode (it's the red bit at the bottom in the piccie in the link) would leak, but after 20 years I tried it - good as new!
Just like my Sinclair Programmable. Works fine. Now, if I could get my HP-35 calculator working again...
People totaly wig out when they hear of nuclear power - "ooh, it's nasty, you can make bombs out of it, thats EEEEVIL" Then they do bad street theatre by dressing up as skeletons and Uncle Sam.
Wierd thing is, you NEVER see a crowd of hippies waving signs saying "NO INTERNAL COMBUSTION ENGINES! NO CHEMICAL EXPLOSIVES!" Both of which have killed way, way more people that nyookyular-anythings.
Funny old world, aint it?
Oh, and er, Jean-Luc? Unless they invented time machines, there is NO WAY Wolverhampton College could have been teaching FORTRAN on IMB PCJRs in 1973. I'm just saying.
Ah, so the tubes are decatrons.
Web page for the comupter: http://www.scit.wlv.ac.uk/university/scit/history/witch.html
One they are tubes. Mullard would almost certainly have made them (being a UK located company), and they call them tubes. For a set of pictures look here. http://www.tube-tester.com/sites/nixie/datdekat/Z502S_mul/z502s-mullard.htm
Note the label on the box, Mullard made both valves and tubes. If it doesn't control electron flow it isn't a valve. The fact that our American cousins call everything a tube does not mean that we call everything a valve.
Also, it is, rather counter intinuively, perfectly reasonable to express the memory capacity in bits. Everyone gets taught that a bit is a binary-digit. But this is only half the story. A bit is more formally defined as the amount of information that is stored in a binary digit. There are other units of information, three are usually defined: the bit, the nat, and the ban. These are the binary, natural, and decimal units of information. One bit equals ln(2) nats = 0.683 nats, and about 0.301 bans. Thus a decimal digit contains about 3.32 bits of information. Since the computer had 900 decatron tubes each capable of storing a single decimal number (and thus held 900 bans), it contained 900 * 3.32 = 2998 bits of information. The mistake is converting that to bytes. That isn't a well defined operation. A bytes does hold exactly 8 bits of information (and thus also holds 5.45 nats or 2.409 bans), so it is forgivable, but still wrong. Anyway, the conversion yields 373.5 equivalent bytes of information. So something may have wrong anyway. However, not all of the tubes will have been used for storage, many will have been used as computational elements, so it may be that the number of tubes devoted to actual data storage was more in line with the computation. But I suspect someone just got the number wrong.
A little history. The history of the definition of the bit is fully documented. It was first defined and used by Shannon in his seminal paper on information theory: A Mathematical Theory of Communication, The Bell System Technical Journal, Vol. 27, pp. 379–423, 623–656, July, October, 1948. (He gives credit to his college J. W. Tukey for the term.) It is perhaps one of the very few cases where a common term has its entire etymology and history perfectly understood. I get arguments all the time (mostly from IT people, who somehow think that they own the term) that a bit is, and can only be, what a memory element holds. But the guys that invented and defined the term would disagree.
"Don'd confuse decatrons with numicators which display numerals 0 to 9."
Numicator[tm] was a trade name. The generic is "Nixie Tube", and trust me they came in a lot more varieties than just numeric. I still have a clock, a VOM, and a couple frequency counters that use Nixies ... Yes, it was "Nixie Tube", never "Nixie Valve" ;-)
Yes, they are gas filled, usually low pressure neon with a little mercury or argon (?? From memory, I could be off on the Penning mixture details, it's been a LONG time since I've even thought about this stuff.)
 Generic in the same sense as "kleenex"; I think it was Burroughs that owned the Nixie trademark, but as far as I know they never defended it in court.
Decatrons are valves and they are gas filled.
The glow is around the Anode not the Cathode.
While the only thing they can do is to count, they do hold state equiv. to 3.3 bits. The state can be examined by applying 10 count pulses an watching for the carry output. So if the 7th anode is glowing the carry will come after three pulses. After the carry has pulsed, start pulsing the decatron in the adder until all 10 pulses have been applied. The adder will have moved on 7 positions and may have generated a carry. This is roughly the way the Witch adds decimal numbers together.
" When a first generation computer is already running in Australia.. http://museumvictoria.com.au/csirac/"
According to Wikipedia "The machine finally found a permanent home in the Melbourne Museum in 2000. It has not been operable since its shutdown, but many of the programs that ran on it have been preserved, and an emulator has been written for it. "
No mention of it working on the museum web site either....
When I was there in the late 70's as a Post office Apprentice we nicked a Reliant Robin form the car park and put it on the flat roof of our accommodation. Owner was NOT happy.
Motorbikes up and down the 100 yard long corridoors.
Got the Pole Erection Unit and 'planted' a pole in the lawn outside the front door of the mansion.
The bar in the mansion, well, those hairy arsed welsh jointers drunk it dry every night..
At 53, I remember it like yesterday.
Ahhhhhh Happy days.
>> "the computing equivalent of the raising of the Mary Rose". OK, without the giant crane, massive diving and engineering effort, not to mention lousy weather and bloody cold water.
Did you miss the bit were they said the computer was being stored in Birmingham?
Valves or Tubes in the U.S are very very easy to get. One previous poster thinks they have gone out of fashion. The fact is they are better than the modern equivalent in many cases. Langrex down in Croydon still have and buy in stock from all over the world. Many shortwave transmission station use valves and being a radion amateur myself still use them.
Apart from dropping them on the floor they are a lot more rugged electrically speaking as they physically move electrons through a near vacuum, with a few gasses as opposed to a solid substrate.
Listen to any valve radio and then listen to its modern equivalent. Modern ones just just rubbish by comparison.
Going back to an article done by the late Paul Young for Everyday Electronics in the 80's, when he was lucky enough to meet the boss of Sony (I think), even the boss (in Japanese) agreed.
Why do you think places like Maplin sell a Valve Amplifier kit ?
Good luck to those at Bletchley - Keep up the good work !
The main problem with decatrons was comparatively slow switching, the ones used in WITCH are HivAC GC10B's - they glow purple - some high speed (and high voltage) ones are used in the multiplication and division unit, these have a pinkish glow.
It was a most impressive sight at night !
A really neat side effect of the use of decatrons was that you could see exactly what number was in every memory location simply by standing in front of the machine and looking at it -
great for programme debugging.
You could also stand in front of the multiplication/division unit and watch it doing long multiplication and long division step by step.
The reason that the Colossus does not count as the world's oldest working computer
is simply that the machine on display at Bletchley Park is only about 10 years old.
There are other replicas and rebuilds of early machines around such as
the Manchester Baby, the Atanasoff machine and the Zeus Z3, the WITCH
however is original and is, AFAIK, the 3rd oldest surviving computer, the
oldest being Australia's CSIRAC and the 2nd oldest the Pilot ACE in the
London Science Museum.
Biting the hand that feeds IT © 1998–2021