PC power supplies (the Joy of Standards)

[Andy Wade]

... or *was* there, it now seems. Thanks for that enlightenment.

[Bob Eager]

Not long ago, there was a working 803 at Bletchley Park. If they get the project off the ground, I'm supposed to provide software for a 2900 they have been promised.

(I just missed the 803 at work; I arrived a year too late and started on the Elliott 4130).

[Roland Perry]

In message , at

22:47:57 on Mon, 22 May 2006, CWatters remarked:

Mine has two: USB and 1394 (Firewire).

[Andy Wade]

That'll be the one Jules was talking about, I suspect.

Weren't they ICL by the time the 4130 arrived? These machines must have been standard issue to universities at the time - Bangor, where I arrived as an undergrad in 1970, had followed the same sequence. By that time research work had pretty-well all moved over the the 4130, but they kept the 803 going for undergraduate use. We were taught Algol programming on the 4130, but jobs had to be submitted on punched cards to the operators through a hatch in the wall and you picked up your line printer output (failed to compile again, surely not?) later in the day - quite a grind. Using the 803, OTOH, was much more fun since they let you operate the thing yourself, after taking a fairly trivial test. I think it finally got pensioned-off sometime around '75-76, when the hardware started to get rather unreliable (frequent core parity errors, ISTR).

[Julian King]

Actually it is surprisingly less difficult than you might think. Several of the machines we had in for tender had serial. Some had this via a stupid dongle, but the HP NX6120 for example has a serial port on the side, just as you might hope.

Julian

[Bob Eager]

Some of the documentation had 'Elliott' on it...I suspect it was on the cusp!

Bangor had an online system at one point - KOS, the Kent On-Line System. It went out to all of the 4130 universities AFAICR. I arrived at Kent in

1970 as a student, and had fun making KOS do things it wasn't supposed to. I ended up managing the successor to the 4130.

[Nick Shipman]

No, I think the 4130 was an Elliott Automation machine. It got renamed the ICT 4130 then the ICL 4130.

It was the first computer I ever used, in 1976 or thereabouts. We submitted coding forms and got back line printer output and a punched card deck, which we corrected on a 12-key manual punch. Turnround time was one week. It made for careful coding; most job failures were introduced by the punch operators, and if the second run failed that was cause for concern...

Nick

[Jules]

That's the one :-) It's got a minor core fault which needs tracing, but otherwise ran - unfortunately a few months ago we were told to move it to a different building (at the time the park trust claimed ownership and we had a heck of a job proving otherwise) and so we dismantled it into separate cabinets. That problem went away (thankfully) but as we're planning for the room that it's currently in to be the large systems hall when the new museum opens, I expect it'll stay decoupled for the time being.

We've got a film drive for it - but unfortunately no controller hardware. If somebody knows of one lurking in a shed, shout! Worst-case, the Science Museum have one in storage that we *might* be able to have on loan one day.

Indeed! As soon as the lawyers have sorted out lease details for H Block on the park we can start getting the building into shape, and after that start filling it with interesting stuff :-) Going from about 500sqft of space to 13000sqft sure will be nice.

We've actually got the 2966 on site in storage along with boxes of spares and documentation - where it's sat for ten years awaiting room to actually put it on display :-( Cosmetically the machine's not in the best of shape - usual panel rust etc. - but so far as we've looked the innards all look to be healthy enough and moisture doesn't seem to have caused any major problems. I expect there may be rubber components in the drive mechanisms that need replacing.

The media situation isn't so good though - the disk packs need rebuilding before they'd work (and so contents will be lost), and I doubt the tapes are still readable. Hence the interest in your 29xx media preservation efforts...

All in all the machine's a shade over 40 cabinets I think, so should look quite impressive in the intended room (I gather that's pretty small for a typical 29xx though). The Elliott will go in there too, the Marconi TAC, DECSystem 570, and various large machines that are more modern (HP T500, Sun 6500 etc.)

Our power bills are going to be interesting.

cheers

Jules

[Jules]

How much of Kent's computing history is preserved? As I studied there I've got a particular interest...

I know the ICL 29xx is long-gone.

The Meiko is at Wroughton, but I'm not sure what state it's in (infuriatingly, according to what I've been told, staff skipped it rather than offering it to any museums, and it was left to the students to retrieve it and find a home for it. That probably means it's not in the best of shape though, and manual / software may not be present)

A couple of the HLH Orions seem to have survived in private hands.

ICL DAP - no idea where this went. Would love to hear that it's still intact somewhere (and it'd be even better to secure it for the museum)

Parsys - no idea again. It was just being installed circa 1996 I think, so

*may* still be on site. Questions to the computing dept. have so far fallen of deaf ears though.

HP Apollos - I've been unable to trace what happened to these too. Kent had a few in the first half of the 90's, and they're reasonably significant machines due to the networked nature of the Domain OS (plus the proprietary keyboards are like gold dust)

Motorola 68k crates - I have vague memories of using these when I was there, somewhere over in the electronics labs. I suppose I'm more interested in those for personal memories, but they were quite nice examples of 68k trainers.

PDP-8 - there was one over in Physics on the top floor when I was there; no idea if it's still there or not!

Cambridge Ring stuff - some still existed on site in the early 90's; expect it's long-gone by now though. (we don't have any examples of Cambridge Ring hardware unfortunately)

... I bet there were all sorts of other interesting machines there over the years though. Whether any are still on-site and in a position to be rescued or not, I don't know...

cheers

Jules

[Jules]

Are they real serial ports though, or something hooked up internally to the USB? I've not used USB-serial adapters myself before, but I've only ever heard bad things about them from people who have tried to use them in anger (particularly talking to older devices).

I'm sticking with my old Thinkpad for as long as I can - it might be a little slow, but at least it's got serial, parallel and USB on board so can cope with most things. I just need to find a floppy drive and SCSI card for it at some point...

cheers

Jules

[August West]

Much the same happened when Meiko closed their office in Aztec West (Bristol). I was workin next door at inmos at the time, and we could see them skipping all sorts of stuff, but we couldn;t save it, as the skips were locked. But at least we did manage to save an awful lot of transputer stuff when ST took over inmos, and closed the test floor.

[Nick Wagg]

My first love, sigh.

[Bob Eager]

Oops! Didn't realise that! Did I teach you?

I have a CPU card left....!

We never had a DAP at Kent, unfortunately.

Chemistry had one before that...running the mass spectrometer in the early 70s.

I'm always asking but there seems little left...will have another look though.

I got involved in the Edinburgh history project because I worked on their operating system so much...and I kept all the source code.

[Bob Eager]

The emulator is quite fun. I built a large program on OS/8 a couple of weeks ago...!

[Bob Eager]

I have all the necessary stuff and have been working away at getting to the stage of building a system using a compiler on one of my VAXes. I am still looking for a 9 track drive to attach to my PC, which would make it a whole lot easier...

Didn't realise you hung out here...nice to meet you! If you want to make contact by email, the address given here DOES work.. but easier to use

(bob) (at) (eager) (dot) (cx)

[Roland Perry]

In message , at 10:47:56 on Tue,

23 May 2006, Andy Wade remarked:

I did my Algol programming on a 4120 (iirc), perhaps that's a very similar machine (would have been 1968-70). On punched cards, which are quite handy because you can build a library of often used "lines" of code [1], and shuffle them into a program. The trick was to make friends with the operators, and have them run your stuff first (because you had a track record of not writing code that got stuck in futile endless loops); and then later they might let into the computer room and run the stuff yourself.

[1] eg: "Begin" "End" "For X = 1 to 100" etc.

[Tim Ward]

Oh yes, the little blue decks of duplicate common blocks (before the days of #include). Which we put through the big duplicator, as we needed dozens of copies, and didn't bother to interpret them, so we had all these cards with holes in and no printing along the top ... got quite good at reading the holes.

[Bob Eager]

Basically the low end version of the 4130. A lot of complex instructions (called 'extracodes') were implemented in software; when used, the operating system was called and it interpreted them. Mainly stuff like floating point.

[CWatters]

No twiglets in the tuck shop.

[CWatters]

Ok Mr Clever. Which one of those uses -5V then?