The following is a list of people associated with the English
Electric DEUCE COMPUTER in those early days,
as they come to mind, from the memories of those listed here or they contact me
after visiting this website.
There are now over 490 names in the database. (JB September 2010)
Latest Updates and Additions to the website here
updates.html
Please use the alphabetical index below
to locate DEUCE PEOPLE on this and other pages of the website.
A - B - C - D - E - F - G - H - I - J - K - L - M - N - O - P - R - S - T - V - W - Y
Thumb Nail images are being added progressively. Click the image to
go to the full size photo
Use your back button to return
Use the SITE INDEX to locate a DEUCE installation.
To return to the DEUCE homepage click the image below.
Visit "The Staffordshire University Computing Futures Museum" "
Here
[Dead link, July 2024]
Which has photos of Nelson Research Labs where the early DEUCE were built and
tested.
Many of the staff listed here later transferred to Kidsgove
"In the 1950's there was no DEUCE at Whetstone and we were allowed to use the one in Stafford, staying at "The Swan" hotel and eating like kings. The lab was situated out in the country in Blackheath Lane and the three senior chaps there were Cliff Robinson, Allan Gilmour and John Denison. These three were unfailingly kind to us embryonic programmers and often went for an after-lunch walk up the lane to which we were sometimes invited." (Charles Broyden)
The 11th March 1955 edition of "Engineering" states that there were three DEUCE shown at the demonstration at English Electric on 17th February 1955. One was due to go to NPL, another was due to go to RAE Farnborough and the third was to be retained at NRL. A fourth (possibly under construction) was due to go to the company's London office later. (Robin Vowels)
"A name to conjure with, and strike terror into the hearts of all and sundry. I have similar recollections (to Rod Whitworth), but I was in the middle of a particularly trying test run, with post-mortems and and whatever-it- was-called-when-you-punched-the-instructions-out-as-they-were-being-obeyed. (Program display? Key somewhere in the middle of the board in a down position, use purple striped cards?) He snuck up behind me, made me jump a mile, and looked a bit taken aback when I wasn't at my most friendly and forthcoming!
[In fact, away from NRL, he was a very nice chap, but I think he left that persona with Ray at Security!] "(David Leigh)
"On the day that the Princess was present, JK Brown himself came down (to RAE Farnborough) and, after seeing the Princess, JB remembers that JK Brown was all puffed up and reported back to the workshop in a loud voice - "Isn't she tiny!" (John Boothroyd - Deuce Reflections)
Cliff Robinson reported to Wilf Scott, who protected us from the fearsome J.K.Brown." (Mike Wetherfield)
I think it must be 50 years since I last demonstrated the Easter Sunday program, and you may be interested in its background. We were due to formally demonstrate Deuce to the first batch of experts, (mainly from NPL) who were due to come to Kidsgrove, and about 3 days before they came, J K Brown insisted that we invent and make about 5 demonstration programs to show it off. One of these was the Easter Sunday program, which, if I remember rightly, was written on time by John Denison. Another was the program to factorise any number up to a million. The task was duly completed and they remained the backbone of demonstrations to curious laymen for years to come.
JK also wanted a portable stand to be built so that 12 or more visitors could see demonstrations at the same time. The front of this "bandstand", as it was called, can be seen on the photo of A C D Haley doing a demonstration, and it was brought out regularly for visitors who came in batches. "bandstand" photo
I was very impressed to see how much of the early days you have managed to record for posterity, with so few of the original documents available.
"Cliff Robinson ran the show - one couldn't have asked for a better manager. Others, like myself, would have followed him anywhere." (Mike Wetherfield)
The Computer Journal , Volume 1, Issue 4, 1958
DEUCE interpretive programs
C Robinson
The English Electric Company Ltd., Stafford, UK
This paper describes the principal features of (i) The General Interpretive
Program, (ii) The Tabular Interpretive Program, and (iii) Alphacode, which are
the interpretive programs which have been most extensively used in solving
problems on DEUCE. The characteristics of these three schemes are compared and
contrasted. - Full text in TIFF form available
here.
My connection with Deuce is that my father is Cliff Robinson, who worked at NRL until 1961, and then at Kidsgrove.
My earliest memory (of anything) is sitting down on the living room
floor with Allan Gilmour and my father, helping them with an upgrade.
(I have always thought it was ACE to Deuce but reading your documents it seems
more likely to have been Deuce Mk I to Mk II.)
The task involved punching one new hole in the same place on each card (my
task) and putting the chad into an adjacent hole (their task).
In my early teens I used to go in to Kidsgrove at weekends to use
Deuce, where Alphacode was my first programming language.
When a Deuce was decommissioned, my brother and I acquired a lot of the
switches, lamps and relays, and used them to make our own simple binary logic
circuits.
My father had originally come in to computing from the world of
calculator arithmetic. (He had compiled tables of logarithms, sines, cosines
etc.) I was introduced to the Brunsviga calculator as soon as I could count,
and I was then taught all the tricks and shortcuts.
I see one of them is published on David Green's web site:
here.
My father moved on to ICL, where he ran their computer bureaux and supplies subsidiaries and was responsible for the introduction of the electronic calculator to the UK. I remember he wanted to supplement the Japanese manufacturer's instructions with step by step guides to common calculations, clearly an idea rooted in the Deuce Programming Notes.
He is extremely numerate to this day!
As far as Deuce is concerned, I joined English Electric at Nelson Research Labs, Stafford, in the second half of 1957. I was employed as a "Mathematician", but this effectively meant "Deuce programmer".
I learnt Deuce programming from the excellent Vic Price / George Davis manual. I took to Deuce programming like a duck to water - the first program I wrote, to teach myself, calculated PI to 150 decimal places, punching the result onto 5 cards.
Later I helped to introduce, and was heavily involved with, the various nefarious programming practices known as "frigging the Multiplier" and I devised the first "Read eight 8-digit integers" subroutine, R24T - and later, after John O'Brien (Marconi) produced an improved version (R24T/1), I inevitably had to go one better and produced R24T/2, which used even fewer instructions.
Competitive days, them was!
I also wrote a "Scheme B" Brick which read eight 8-digit numbers. "Scheme B" was a matrix-handling system, the successor to "Scheme A".
About the last serious Deuce program I wrote was the Mark 2 Alphacode Compiler (which was really an Assembler) - thanks to "Mult-frigging" this version kept the card reader going continuously instead of stopping after each card.
Read Mike's recollections of programming DEUCE in the late 1950's here
See also Allan Gilmour's entry
Brian had left before I joined NRL but was always spoken of as a real expert.
I believe he developed both "Scheme A" and "Scheme B", and also wrote "Willie the Worm" which danced around in one of the circular CRT displays on the Deuce console! (Mike Wetherfield)
Deuce programmer, already at NRL when I joined. Extrovert and a keen hockey player. (Mike Wetherfield)
Deuce programmer, the next to join after Jim Lucking. (Mike Wetherfield)
Deuce programmer. At Kidsgrove, when software development for KDF9 became a properly organised department, he took over the running of it. (Mike Wetherfield)
Neville Hawkins wrote both the Interpreters (Mark 1 and Mark 2) which executed the code produced by the corresponding Alphacode Compilers. (Mike Wetherfield)
Bernard was left in charge of the programmers who remained at NPL, who included Neville Hawkins (and Mike Kingsbury?), after the rest of us decamped for Kidsgrove. (Mike Wetherfield)
I have always understood that he personally wrote the DEUCE program which worked so brilliantly when ITV did their live broadcast from Nelson Research Labs on the night of the 1959 General Election. DEUCE correctly predicted the final outcome after comparatively few results had been announced, and in this respect did much better than the competition - I forget whose computer the BBC were using. (It was as an RAE Deuce - JB)
Many of the NRL staff took part on that evening, some of us just carrying bits of paper from one place to another (I was apparently "seen on television"), though there must have been people punching the results onto cards and feeding them into the DEUCEs. I forget who the ITV "front man" was, but there on-the-spot "psephologist" was the celebrated statistician Maurice Kendall - I remember him sitting next to Cliff Robinson as they were being interviewed in the early hours of the morning at the end of the show.
(Mike Wetherfield)
The solution of railway problems on a digital computer:
This paper gives an account of railway problems solved on the English Electric
DEUCE The
Computer Journal, Volume 1, Issue 1
"The brightest engineer and programmer in the team."
(John Boothroyd - Deuce Reflections )
"(known as "Speedy"), was responsible for the music programme, we used to gather round the DEUCE at Christmas time to listen to it playing carols." (David Leigh)
John Denison wrote the Mark 1 Alphacode Compiler. (Mike Wetherfield)
Author of Programmes to Aid Programmers - DEUCE Programme News - No. 16. November, 1957.
Roger was responsible for Deuce News, distribution of subroutines, etc. (Mike Wetherfield)
|
|
John program testing at the Deuce control desk. |
"JB was expert in setting up the Multiplier/ Divider function on the DEUCE. In earlier computers multiplication and division had been undertaken through software / programming. Then the NPL designed the first hardware divider and, while JB was not involved in this design process, JB was the first to set it up fully. JB then became expert at fixing problems that went wrong with the Multiplier/Dividers in DEUCE machines."
Extract form [ "A few reflections from
John Boothroyd about working with DEUCE at Kidsgrove"]
Recorded by his daughter, Alison Hutchison - January 2007
"I gather that he was a mature man back in the late 1950s, based
on his brilliant monograph on the multiplier and divider in 1959.
It was he who discovered that you could use the divider to convert binary to
BCD by changing the content of the divisor and dividend during division. He
noticed that the result was always wrong by a constant amount, and was able to
correct that.
Arthur Bailey told us about that achievement in 1962, though the name J. Boothroyd was already familiar to me from the DEUCE program write-ups. ( Robin Vowels)
JB left Kidsgrove in 1964 to take up a position as Officer-In-Charge of the Hydro-University Computing Centre, University of Tasmania. (Alison Hutchison)
I noticed that he contributed a few algorithms to the ACM, one of
which I've put on my web site last month."
http://members.dodo.com.au/~robin51/acm-290.inc
( Robin Vowels)[dead link]
"At just about the time I left Boots for NRL, John Boothroyd made the reverse trip - Boots recruited him to be their chief computer engineer, but of course the machine wasn't forthcoming. I only met him when he rejoined NRL about a year (I think) later, when he became more a programmer than an engineer - he became head of training; he and David Ozanne wrote the "new" Deuce training manual." (Mike Wetherfield)
"Colin describes the ACE pilot as a "dog's breakfast" and from 1949 through to 1951/1952 they set about engineering what was very much a 'laboratory model' into a more robust entity, to be known as DEUCE" - (Jeremy Walker)
"Colin always looked ahead and never stopped thinking about wider
issues - such as what problems would lie ahead after this one was tackled, or
what other way to consider the current problem."
(John Boothroyd - Deuce Reflections )
"During the time I was at NRL, Colin Haley and Ray Ellison sat in the machine room with, and maintained, the two Deuces; one took one's hardware problems to them." (Mike Wetherfield)
"A natural electronics engineer - absolutely brilliant - no one better in the team." (John Boothroyd - Deuce Reflections )
Engineer who (I think) was responsible for attaching Magnetic Tape units to Deuce. He worked in a little round glass lab. of his own, perched on top of the NRL building. (Mike Wetherfield)
"I also remember David Ozanne who was a wizard with the DEUCE control panel and could almost program the computer using the panel alone. He came somewhere between Cliff R and Co. and we neophytes in seniority and was regarded with awe for his skills." (Charles Broyden)
"Wrote the Mult/Div test programme which ALWAYS failed if you single-shotted through it." (Jeremy Walker)
"I was in at the beginning of the computer revolution. As an
undergraduate reading Electronics I spent a long vacation working in the
English Electric Research Laboratories at Stafford. I was working on the
famous Deuce computer. It was massive, housed in a air conditioned building
about half the size of this church. It rarely worked for more than an hour
or two. That was in the early sixties. About ten years later I was university
chaplain and lecturer in Electronics working on an IBM transistor driven
machine in the University of Zambia in the heart of Africa, and having to
rewrite all my Maths and Engineering lectures, because real problems could
now be solved by available, reliable computers of great power. Now I carry
in my pocket a Psion computer that possibly has as much power as that IBM
machine, and certainly has more power than all the 1960s computers in
England put together."
[ Extract from "Weaving the Web" - Bishop of Southwark's address to
the Weavers Company 22 February 1999 ]
"I left school in 1961 and went to work at Nelson Research Labs
(NRL), which was just down the road for me, where they used to build DEUCE.
(You can still see the marks on the floor - I checked last week.) I was a
kind of second-string engineer, and helped bring the machine up to speed in
the mornings (among other things) - 8.40 - 10.40 each day. Trouble-shooting
during thunder storms has vivid memories for me. And testing - and retesting
- diodes, valves, and the like. I went to college later that year, but kept
vac jobs going at English Elastic (!), which also brought me into contact
with KDF9, KDP10 and all that."
"I returned to NRL on graduation, but used to pop across the road to the
college to use the DEUCE with paper tape. I was poached in 1966 to lecture
at the" college -
Jim Lucking was so enamoured with Deuce that later on, when he was running a programming section (which I think wrote test programs for System 4) he made all his programmers learn Deuce machine code - he thought there was no better training. (Mike Wetherfield)
At NRL and Kidsgrove, he was the tabulator expert, and knew how to plug-up the necessary "boards" to produce the best printed output from punched cards. (Mike Wetherfield)
David ran an Operational Research department (which I presume developed software) at NRL and subsequently in Kidsgrove - basically for Bureau customers, possibly including English Electric itself, who needed this expertise. (Mike Wetherfield)
"The only other person that I remember from Blackheath Lane was Janet, who was only a bit senior to us and was a wizard with Scheme B. Scheme B was a matrix manipulation program which required the correct shuffling of vast stacks of the Hollerith cards on which the matrices were stored, and I think even Cliff was outshone by Janet in this area." (Charles Broyden)
My younger brother - Joseph Galvin - was photographer at English Electric in the 1950's (Bill Galvin)
My own qualifications for having anything to say about it (Deuce)
are much more tenuous, since my only contact with the machine was from late
1962 onwards. By that time, it had become so "old hat" that the Nelson Research
Laboratories would agree to leave their machine switched on after normal
working hours so that we mere apprentices and junior engineers could arrive on
our bicycles, carrying haversacks full of punched cards, to run our own
programmes in the evening.
(Extract from a photo copied document dated 13-1-91 held by David Leigh. The
full text will be added when the photos have been processed)
I was at NRL from 1964 to 67. Mike St.Oakes was a friend of David
Leigh who I remember was always inventing some electronic gadget or other. I
was a Mathematician - well that was my title - I used to run routine ray
tracing applications on the Deuce machine.
My main work was converting math models to Algol to run on a KDF9. My clearest
memory of the Deuce was when the NRL Director was showing the Deuce to some
visitors and I was asleep at the console having succumbed to the warm
atmosphere within the Deuce room.
I worked in the Operations Research department at the NRL in Blackheath Lane, Stafford for 6 months between July and December 1966. This was part of my "thick" sandwich course from Bradford University (formerly Bradford Institute of Technology). The university course had already introduced me to programming computers so I was assigned to developing a computer program, written in Alphacode, on the DEUCE computer.
I remember the laboratory in which the DEUCE was housed as being stifling hot in the summer and beautifully warm in the winter. On a number of occasions I watched Pete , the support engineer, do the start up tests. It took a number of attempts to turn on the power as it was not uncommon for a few valves to fail on power up. Once passed that hurdle Pete then played the DEUCE keyboard by toggling the keys to count upwards in binary amongst other things. Pete breathed a sigh of relief when the magic words "THANK GOD FOR THAT" appeared on one of the cathode ray tubes indicating that all start up tests had been successfully completed. He then handed the DEUCE over for production use.
I wrote a program to simulate random air bubbles in a metal casting. A support group was next door to my office led by Frank? and having Fay Colclough and Jenny Cole as punch card operatives. At least the chore of punching cards was removed from my duties!
Thanks for a good website and keep up the good work of maintaining a historical survey of those pioneering computer days.
[ Website ]
From September 1967 to April 1968 I worked at Nelson Research
Laboratories. I was a member of a small group at the laboratory undertaking
research into programming languages, compilers and compiler-compilers.
I didn't have any material involvement with the Deuce during my 18 months at
NRL. There was still one there at the time and I had the guided tour, but never
had occasion to use the machine. I had a look at your website which was very
interesting. Only a very few of the people involved with the Deuce were known
to me, I guess I arrived on the scene in the twilight years, probably not long
before they became museum pieces.
Check out Mark Woods excellent " History of Kidsgrove Works - Nelson Industrial Estate ".
See also "The Model Deuce" from an article in the local Sentinel newspaper.
"Chief Engineer when the prototype Deuce was to be 'commercialised, further-developed and put into quantity production' by a new team at Kidsgrove under Derek Royle, in 1955." (Cliff Robinson)
"Replaced J. K. Todd as Chief Engineer. Certainly by 1958, Asbury was in charge of the whole Development Establishment at Kidsgrove - of which Computers was only part." (Cliff Robinson)
"After 4 months I went to help Frank Thompson install the BP machine in the City of London. We went in his car and daily he used to scare me by racing up Ludgate Hill and demonstrating a handbrake turn in front of St.Paul's Cathedral.
The installation was in a small existing room of a Victorian building. The layout was dreadfully cramped, the Delay Line mushroom having to be placed on a ledge 3-feet high, making adjustment awkward." (Steve Allcock)
"When I was first let out of the classroom by "Arfur" as fit to be let loose, under guidance, on an unsuspecting computer, I was put under the care of Fortunato MARCIANO, also known as "Luigi" who explained to me that the "Luigi" was a nickname awarded by Arthur Bailey, when he first started training, who obviously decided that "Fortunato" was too much of a mouthful." (John Ryan)
Special Trainee - March to September 1959. As the result of a suggestion made by my school that I should get some industrial experience before going up to Cambridge that Autumn, I went to Kidsgrove one Friday in March 1959 where I was interviewed by Derek ROYLE, who told me that I had the right kind of mind for computers and asked me if I could start work there the following Monday! I said "Yes!" - so began my involvement with interesting computers.
I was employed as a "Special Trainee" ( which I suspected was a useful personnel label for otherwise unclassifiable employees who would not be paid very much ) and was put through a slightly abbreviated form of the Maintenence Engineers Training course. So I remember Arthur BAILEY and his perpetual roll-up.
My lab time was spent, mainly, working on a Mk IIA machine ( which may have been the first to be built) and which subsequent casual enquiries led me to believe ended up as the Kidsgrove Bureau Machine. I suspect it was the first, as, during the build-up and test work we did in those days, we had trouble in using the AIM unit, which also housed the logic needed to access the extension memory.
This was traced to an E90CC valve, which did not seem to want to switch over - that, when tested on the lab's valve test box proved itself to be "dead" on one side, very low emission "dead".
A replacement was plugged in and we tried again - no joy! Valve tested - same as before - one, same, side "dead".
A further, pre-tested, replacement was plugged in - same outcome!
A further replacement was plugged in, but this time, we stood inside the machine and watched what happened when the power came on - and one anode began to glow red hot!
At this point , power off, and look at circuit diagram, where it was revealed that the circuit in question was wired between something like the +300 and -300 HT lines, where something like +100 and -100 would have been much kinder. At this point, the test engineers did something sneaky, but I would suggest, sensible. We altered the HT connections to be more civilised, powered up and checked that we could now access the extra delay lines, then powered down, put things back the way they were - and went humbly in search of a design engineer to help us solve our problem - who was then encouraged to tell us to try that which we had already established would work.
That DEUCE test frame was also the only machine I have literally worked ON, is in ON TOP OF , as I had to climb on top one day to repair a failed address tree.
"Group Engineer Computer Engineering D.P. & C.S.D. He was a brilliant engineer, highly regarded by everyone. His favourite trick--when a resistor or other soldered in component was needing (to be) changed - no need to switch off the cabinet power, just touch the earthed soldering iron tip on the + or - power rail, the over- current circuit would kick in, and drop the cabinet power automatically. Needless to say, nobody else was inclined to switch off this way!" - (Eddie Poole - BAC Warton)
"I remember going on site with Derek Royle to attend to problems with a rolling mill at Ebbw Vale, in Wales. He would work on the live panels with one hand in his pocket. Not bothered at all about any danger. Again, a nice man." (Jack MERRITT)
Derek Royle's secretary - Married Jeremy Walker
|
|
Jeremy checking the contents of a short tank in the CRT. |
" I joined the Industrial Electronics Department (IED) at Kidsgrove, from English Electric at Bradford and my then boss, Derek Royle, was tasked with the job of recruiting a team to take the machine forward. Despite the re-engineering of the product that had taken place at NRL, it was hardly reliable and so a good deal of work was to be done on making it reliable and enhancing it as necessary, with all further development and output taking place from Kidsgrove. "
Jeremy presented this talk, DEUCE - I'ts Life and Times , to the North West Group of the Computer Conservation Society in 1995.
DEUCE Maintenance Engineer. "Brian visited MAFF Guildford in the summer of 1961. Alex Robinson, my girlfriend Jackie Rawkins from the machine room and I, were in the engineers room when this enormous blow fly buzzed around. Brian grabbed the can of fly spray and chased the insect around the room. A while latter we all had a cup of tea and Jackie choked on her first sip of tea and went a very strange colour. "Should have told you" says Brian "That's where I killed the pest, in amongst the teacups." (John Barrett)
DEUCE Maintenance Engineer. I worked on DEUCE as a young support engineer, employed by Derek Royle, many many moons ago.... with Bill Beckett, Keith Powell etc.
I was interviewed by Derek Royle at Liverpool University who were advertising for a site engineer. Derek poached off the applicants he fancied for his DEUCE support section at Kidsgrove run by Keith Powell, and two of us, Eric Dobson and myself came to Kidsgrove in 1962 'ish. I remember especially being asked " If TWO + TWO = FOUR, what number does each letter represent" - It took my four brainy adult kids quite a while to solve it.....I worked it out with a little prompting from Derek.
The DEUCE support team had two jobs, to do annual overhauls on each machine - the A.O.T. 'ing (adjust on test) of the arith unit was one job we sometimes made a right cock-up of - generally speaking the on-site engineers knew far more about their machines than we did - but I suppose it was in the maintenance contract that we visited. The other job was to whizz off and fix the particularly difficult faults, I can't actually remember fixing one...I must have done ...I hope.
I do particularly remember one overhaul week at the National Engineering Lab's machine at East Kilbride with Bill Beckett. They had an intermittent fault on the machine, so I suggested that I should have 'a tap around' - tapping the valves to see if the fault could be brought on. After a minute or so tapping, the machine stopped completely and it took us most of the week to get it running at all. After that, there was no time left for maintenance ... Bill never failed to remind me of this for years after.
My best memories of those days, on DEUCE , KDF6, and KDF9, was the lunchtime boozing in the Lawton Arms .... and the Corner Pin. It was hard in those days, hard, but somebody had to do it....
After a few years in this support job, ICL was formed and I was promoted to a 'Supervisor' at Kidsgrove, mostly checking the expense sheets of the site engineers - not very interesting to a free thinking support engineer, so I got poached off to another company, and eventually started my own business in the computer field which is still going.
But DEUCE, Copthall House, and Kidsgrove were the happy days !
DEUCE Maintenance Engineer
DEUCE Maintenance Engineer
Deuce Spares Manager
Deuce Bureau Engineer. "Mike Gaherty, it is alleged, has a drum - maybe the only horizontal one which was made." (Jeremy Walker)
"Mike Gaherty who worked with me in the Kidsgrove Deuce Bureau, looked after later EE computers and then moved to Kode, he is now retired to the south coast." (Noel Wesson)
DEUCE Bureau Maintenance Engineer. Derek Ranyell (was one of) my first engineers, although (he) went back to Stafford after a short while. (Noel Wesson )
DEUCE Bureau Maintenance Engineer
DEUCE Bureau Maintenance Engineer
DEUCE Maintenance Engineer
DEUCE Maintenance Engineer
Clerk. "Whilst I only spent six months working at Kidsgrove, I still have happy memories of the people, although I cannot recall all of the names of those I worked with; but I am sure that anyone ever involved there would remember fondly Mrs POLIKOFF who acted more like a mother than a clerk to the engineers and was very well liked. " (John Ryan)
I was with EE from January 1962 to May 1963. I spent all my time at
Kidsgrove programming DEUCE.
I shared an office with Mike Wetherfield and Trevor Neal(?). Trevor was developing
something called "Trevor Code".
I never did find out what Mike was working on, something quite exotic I would
guess. He seemed to be the resident guru.
(In 1962 what I was working on was no doubt the Director and other software for
KDF9 - Mike Wetherfield)
As it was just a short walk round the corridor to the DEUCE I never
got to see much else of what was happening in the other huts.
I think of myself as fortunate to have worked on two first generation computers,
each having a different architecture - DEUCE and then the
SILLIAC at Sydney Uni.
And then, surprise, in 1998 I came across a third first-generation machine,
a surviving Bendix G15 here in Perth in Western Australia. This is a development
of the DEUCE machine using a magnetic drum for main memory and in many respects
very like the DEUCE. ( Except for the size !!! JB)
David remembers his first few months with English Electric. "Up to English Electric - 9.50 Euston to Crewe"
I shared an office with Mike Wetherfield and Trevor ?. (David Green)
(The "Trevor" mentioned was, I think, Trevor NEAL - Mike Wetherfield)
"12 April 'Conference with Roger and Richard.' I think this refers to Roger Allwood, my direct boss, and Richard Burrows, another programmer reporting to Roger. The conference was about our work loads. Richard was directed to start work on Partial Differential Equations, I assume for the KDF9. I was to finish Chop Sticks III (which I did - I still have the code!). Roger would have reported to Cliff Robinson." Extract from David Green's 1962 Diary.
Another programmer reporting to Roger (David Green)
"In the early 1970's while I was studying at Bristol University my tutor, Fraser Duncan, showed me the logic diagrams of the DEUCE, the first computer with which he had worked. The DEUCE's main memory consisted of mercury delay lines each holding 1024 bits; just at that time the first integrated circuits implementing 1024 bit shift registers appeared on the market. Thus was born the idea of rebuilding the DEUCE from then current technology. We never really got into the project seriously, academic studies took priority, but it is still my intention, someday, to realize that dream. Sadly, I have yet to find any surviving DEUCE software." ( Hans B PUFAL )
The Computer Journal , Volume 3, Issue 2. 1960
The DEUCE Alphacode translator
FG Duncan and DHR Huxtable
The English Electric Company Ltd., Kidsgrove, UK
A description is given of a recently completed program for translating from a
single-level pseudo-code (Alphacode) to a multi-level machine code (orthodox DEUCE
code). The chief point of interest is the allocation of the single-level addresses
among the three levels of the real computer to obtain an efficient final program.
Full text in TIFF form
here
I joined EE at Kidgsrove in 1962 as a Student Apprentice (or its equivalent) for one year between sixth form and university. This opportunity arose from a Short Works Course which I was enabled to experience from school (Alleyns) - a scheme used by a number of larger employers to recruit future staff from public schools.
I worked for Roger Allwood who was responsible for the section dealing with stress calculations. I went on the last DEUCE programming course which lasted 4 weeks, at the end of which one could barely add to numbers together and print the result on a punch card!
My first task related to a matrix tehnique which Roger wanted to experiment with, called Transfer Matrix (that's about all I can recall) but this was the time when KDF9 was emerging from its prototype stage. Soon I was given a role in a small group of 2 or 3 preparing to develop the Matrix Scheme for KDF9, based on ideas already implemented for DEUCE. I succeed in getting a number of routines to work, testing them on a simulator running on DEUCE. In the process I wrote a DEUCE program to convert data to 'bastardised octal' the code unique to KDF9.
DEUCE was unreliable and a check sum technique was incorporated to provide additional verification. The maths behind this depended on the fact that the calculations were fixed point (essentially integer). We attempted to do the same on KDF9 but this worked in floating point and we soon discovered that A+B+C is not necessarily the same as A+C+B. After some tortuous efforts to devise the maths we realsied that KDF9 either worked or it did not, so the check sum was redundant and we scrapped the idea.
In April 1964 I joined English Electric Leo in the Management Science Department at Kidsgrove. Time spent on Deuce was limited but I used to run a Network Analysis program and sort and tabulate the results with the ICT card sorter.
I was lucky enough to find a job with English Electric after
graduation, working for John Boothroyd. 1962 was of course near the end of the
DEUCE era, but my boss thought it would be good for me to experience the rigours
of computing with the DEUCE.
I still have my old 'DEUCE Lecture Notes' on how to program, and some old code.
After two years at English Electric in Kidsgrove, programming the DEUCE computer and giving KDF9 training courses, he joined Control Data Corporation and was immediately sent to CERN to support the CDC 6600 Serial No 3, the first such machine in Europe. - (University of Glasgow, Physics & Astronomy Research Colloquia)
Web site [dead link]
I worked at English Electric Kidsgrove, first as a student apprentice and then as an engineer in the Data Processing Division from 1961 until 1966. I was instructed in writing computer programs for the Deuce using Alphacode during my first year and I found that fascinating.
For my later years as an apprentice, I worked mainly on the LACE analogue machine in the Kidsgrove laboratory.
I think I learned as much at Kidsgrove as I did at College. Certainly when transistors were taking over the valve circuits that were taught at college, the Kidsgrove training was indispensable to me.
I do recall being told a story about a supervisor and secretary in the Deuce computer room, being caught in compromising circumstances inside the computer cabinet - I remember being shown inside the main cabinet one time and it did look very cosy with the two rows of rack mounted valve modules. Not sure if there was any truth in the story, probably not, but it would be interesting to have it confirmed.
I would just like to say how nice it was to see all the old names I
either knew, was friends with or worked with at Kidsgrove in the early 1960's.
I initially worked on KDP10 in the old bureau, then RCA Spectrum developing
the System 4 executive software as the site supervisor of PTD.
I remember Lawrence Blanch giving me a crash course on
Deuce. Stunning to remember somewhat primitive mercury rectifiers, delay line
memory and being able to display every bit in memory on two circular crt's. Oh
for a return to thermionic valves, life was so much more enjoyable then.
"Our head of product development was Bill Nash. I am standing next
to him in the photo"
(Dennis Hollins - See The Model
Deuce Letters to the Editor )
"I immediately recognised one man in the photograph because it was my
father, Bill Nash with his pipe.
He ran the main workshop where the model of the Deuce computer was made, along with
the metalwork for the production machine"
(Peter Nash - See The Model
Deuce Letters to the Editor )
"I was part of the English Electric team at Kidsgrove who made the model of the first computer, and from memory I think its scale was one-fifth."
See also The Model Deuce Letters to the Editor
"I took a look at the link to the news item on the model, &
recognised the seven faces on the right of the picture, but apart from two of
them I have forgotten the names.
Those two being Len Calvert & Stan Elkin, who is standing 7th from the right
& is wearing spectacles. See The Model
Deuce
I worked with Stan in later years." (Jack Merritt)
"It was taller than a doorway and large enough to fill an average sized kitchen, according to retired engineer Len Calvert, who was then employed at the factory opened by English Electric in 1952 specifically to build the first breed of computers for use by industry, the big banks and other major concerns."
"Although Len wasn't involved in the project to make North Staffordshire's historic first computer; he was a member of the team of engineers who made a scale model one eighth of the actual size, to show off the machine at an international exhibition in Geneva."
( From an article in the local Sentinel newspaper - 10/3/2007) See also The Model Deuce
My wife, Eileen, spent quite a few years wiring the chassis for this machine (from about 1954 to 1960) at English Electric. (Roy Rushton)
Eileen Rushton worked on Deuce and had a bench job building the plug-in chassis. (Jack Merrit)
I started work at English Electric, Kidsgrove, in August 1956, as a
technician apprentice.
My own direct involvement on Deuce consisted of fitting and assembly operations
associated with the Deuce units and the Mercury Delay Lines.
The years at English Electric were by far the happiest of my life, and in common
with many other people about that time, that's where I met my wife Pauline.
Geoff Mould was a young Engineer at the time and worked for Ray Binnion, consequently he was given jobs handed down by Ray, which required liaison with the shopfloor. In my job, I came into contact with Geoff quite a lot. (Jack Merrit)
Pauline worked in Goods Inward Inspection, where, amongst many other things, she used to run the burn-in rig for the high-reliability valves used in Deuce. If provoked, she could still tell you most of the valve types used and the relative failure rates.
Another job was calibrating the numerous Delay Networks for Deuce on a Marconi Q-Meter. (Geoff Mould)
Geoff's wife Pauline worked on the shopfloor & worked alongside Eileen. (Jack Merrit)
I started work at Kidsgrove in early 1958. The Deuce machines that I worked on were huge and one could walk through it to either plug-in or remove the chassis, some of which I built. I worked on the shopfloor section that built the units, plus in the Lab where they were being tested, doing jobs for the Engineers.
It always amused me to see that some wag Engineer had cut the caption from a packet of cigarettes of the day and stuck it above the entrance to the machine. It read 'It's the tobacco that counts'. Which I thought was very funny.
I was on the shopfloor myself for 6 years & then moved to a position in Manufacturing Engineering. I took a look at the list of people that you have as having worked on Deuce, & recognise quite a few. Geoff Mould & Pauline his wife, used to live by me.
I remember going on site with Derek Royle to attend to problems with a rolling mill at Ebbw Vale, in Wales. He would work on the live panels with one hand in his pocket. Not bothered at all about any danger. Again, a nice man.
Arthur Bailey made pointed comments a number of times during the course, "Don't change an AOT*" he said. "They are factory set to 1%. Changing an AOT is putting on a fault to clear a fault, and then there are two faults to fix."
Arthur also told us that if a program fails, and no test programs fail, then use that program as the test program. So when we had some problems with a certain GIP program that used matrix multiplication, I took Arthur's advice one step further. I made that program a permanent test program, to be run on high and low every day just before handing over for normal use.
While there were test programs for individual sections of the machine, there was none that tested everything altogether - that is, to test the machine as a whole. That GIP program seemed to fit the bill admirably. (Robin Vowels)
|
|
This Deuce Training Course Photo shows Arthur Bailey, at right, lecturing to a class of attentive students at Kidsgrove in August 1958. (Noel Wesson) |
I don't recall how long the Deuce Course was - a month? Arthur Bailey used to teach it and afterwards, Jim Richards. (Jeremy Walker)
Gordon, second from left, came from Purchasing and went back there after his course. (Noel Wesson)
The following people did the DEUCE training course with me at Kidsgrove in 1962. (Robin Vowels)
Data Controller. I was fortunate - if only for a few short months - (to) be involved with the DEUCE II computer that was at English Electric works at Kidsgrove.
One of my jobs - as a Data Controller - from July 1965 up to about early December 1965 - was to assemble and take trays of the 64 col cards from the punch girls at EE computer services (under Mrs Gater) for what, if I recall correctly, was the payroll and for the local rates for the local urban district council of Kidsgrove .
The cards were transported across the road from EE computer services to EE factory in battleship grey card trays and, on a very bitter winter morning in November and December, they could almost freeze to your fingers! We later had a small trolley to take the cards across with.
When DEUCE II was 'terminated' EE had an 'open day' and the public could attend and view the whole factory.
I'd forgotten all about Starcross St.! A few years ago - say 15 - I wandered round that area to see if our former premises were still there - they and the road had vanished. Jack Richardson and I 'opened' that Office - a grotty 3 storey tenement building - and when I left the Digital Computer Maintenance Service Unit and those premises later in whatever year it was - 1957? - it was to hand-over my role to newly-qualified Engineer Arthur Saville. (Jeremy Walker)
When I returned to Starcross Street (from Glasgow>) I was sometimes called back to Kidsgrove to help commission Deuces . I used to visit all the sites in the southern half of the country. Brian Bispham used to borrow me from Jack Richardson to help at Guildford, the first one I had worked on with 2 mushrooms.
Noel then moved to Kidsgrove to look after the Deuce Bureau Machines.
Eventually I was offered the job of looking after the Deuce Bureau at Kidsgrove and became a permanent member of Lyncroft instead of a part time one. We had a brand new 80 column Deuce, and a very old one from NRL. Mike Gaherty and Derek Ranyell were my first engineers.
(Extracts from "My Deuce Time" , discover how Noel was lured from LEO by a bevy of 545's !)
|
|
George program testing. |
George Davis started working on computers in September 1950, following Cambridge maths & wartime Radar. He joined the English Electric team helping NPL develop the Pilot ACE, which mainly comprised some wiremen led by a brilliant draughtsman called Arthur Roberts, and reported to Colin Haley who visited fortnightly.
GD helped with software & hardware development, and then got permission to set up and lead a dedicated maintenance team with systematic procedures, which eventually showed that Pilot ACE could work very reliably if treated kindly. Previously, the machine had lurched spasmodically between the Mathematicians who flogged it to death and the Engineers who redesigned it after every breakdown. During this time, GD also got permission from CH to go home for six weeks and write Pilot ACE Logic Design & Programming manuals, previously lacking; these, by their highly characteristic approach and style, are the very recognisable ancestors of the corresponding manuals for successive models of DEUCE, for years a mainstay of users & engineers, and now flaunted on various websites.
A little later, GD discovered that NPL had refused a request from Inland Revenue Department to calculate the PAYE Income Tax Tables on Budget night, visited the IR top brass and undertook to do the work himself. This caused ructions within NPL Admin, but ultimately proved spectacularly successful and NPL has been boasting about it ever since; in fact it is the only computer job given three mentions in the NPL official history (without mentioning that NPL had initially refused it!).
As the design of DEUCE accelerated, GD appointed himself a sort of Chief Technical Clerk, contributing a simple but significant invention, issuing successive drafts of the instruction code aimed at keeping the structure coherent, contributing various snippets of logic design, and embodying 25% of the multiplier-divider design team. In due course, the proposals & arguments died down, GD gave a lecture on the present position, nobody commented and DEUCE had been designed (logically, at least).
Eventually, GD was summoned to see the dreaded JK Brown at Stafford and told to set up and run a DEUCE Computing Service at English Electric HQ, Marconi House in the Strand, as much as anything so that EE Chairman Sir George Nelson could show it off to his important visitors. So lone wolf maverick GD had to become an organisation man! Staff were recruited, many becoming lifetime friends. Vic Price, Chris Woodall, Peter Landin, Ron Eitel, Jack Richardson & John Woolger initially, later Doug Flower, Audrey Birchmore, Anne Stower (now Woodall), Peter Docherty & others. The London DEUCE was a year late, and the team were kindly housed by NPL while Ron, Jack & GD maintained the NPL DEUCE; also, on Monday afternoons, the same three redesigned the card reader & punch control circuits, which had been botched by EE Stafford, while Chris W provided fiendish test programs to show that our new designs didn't work either!
Later, GD became involved in KDF9 specification & marketing, and left EE in 1963, but it was great while it lasted!
|
|
Vic is second from the left in discussion with John Woolger. |
I joined English Electric in 1952, from Cambridge, having graduated
with a BA in Mathematics and a Ph.D. in Theoretical Chemistry.
I had started computing on EDSAC1 as a research student.
I applied to English Electric and was interviewed at Rugby by W. E. Scott.
I was appointed to the Nelson Research Laboratories, which were based at
Blackheath near Stafford.
I met many interesting people, including George Davis and Colin Haley who was
designing and building Deuce but I do not think I ever met J. K. Brown the
Director of Research.
I also met there (possibly on subsequent visits) Cliff Robinson (manager of the
Stafford Computing Service) and Alan Gilmour (concerned with programming train
timetables)
The Pilot Ace had been built at NPL and a group of EE staff under
George Davis was based there.
I remember meeting another mathematician, John Dennison and was fortunate to meet
Jim Wilkinson (not EE staff) a noted numerical analyst and from him I learned how
to compute the eigenvalues of matrices.
The first problem I was given was concerned with flutter on aircraft. Having the
maths from RAE Farnborough, I wrote a progam for the Deuce, which was later used
for the design of the Lightning fighter under development at Warton.
A Deuce was installed in Marconi House and the London Computing
Service (LCS) established under George Davis.
I can remember Peter Landin (linear Programming), Chris Woodall (commercial
applications) and Doug Flower (operational research).
Later, I met Audrey Birchmore and Janet Porteous in Marconi House. There were 4
of us cramped into a dressing room of the Gaiety Theatre which was being
demolished to be replaced later by English Electric House.
I started work on solving numerically the integro-differential equations
modelling the control of a nuclear reactor.
I met Tony Hitchcock a mathematician on the staff of UKAEA a nuclear energy
group based at Risley near Warrington.
I paid many visits to Kidsrove and ultimately the Deuce program was working to
the satisfaction of Tony.
The base of the LCS was moved to Queens House in Kingsway.
There was Janet Porteous, Audrey BIrchmore and various other staff whose names I
can't remember.
The main problem I worked on was slip surface stability of slopes. This was in
collaboration with the consulting engineers BInnie Deacon & Gourlay ( later
renamed Binnie & Partners)
I was working with some staff of the Civil Engineering Dept of Imperial College,
primarily Nordie Morgenstern.
(We were jointly awarded a prize by the Institute of Civil Engineers for a paper
we had published.)
The Deuce programs which I had written for slip circle analysis and later for
more general slip surfaces, were used on the Daer slope, which caused a tragedy
in Wales when it slipped, the Wraysbury reservoir near Heathrow airport, some
motorway in Kent, but primarily on the Mangla Dam in Pakistan.
At this time, the program incorporated earthquake analysis, as specified by Dr.
Ambraseys of Imperial College.
I am pleased to be able to report that the dam is still standing in 2009
despite earthquakes in the region.
Janet was working on a Deuce GIP version of the program. In 1959 we both met
Angus Skinner of Sir William Halcrow & Partners.
They got married in August 1960 and are still very good friends of mine. In 1961
Angus joined the staff of the Civil Engineering Department of Imperial College.
Janet has helped me to produce this document by correcting some of my memories
of the days in Marconi House and English Electric House.
I left English Electric in 1962 and joined the staff of the
Mathematics Dept of the Northampton CAT later to become the City University
London.
I acknowledge assistance in preparing this document from Janet and Angus Skinner
and from my son Colin.
I have been very lucky in this life. God has certainly blessed me with a good
wife, 3 sons , 5 grandchildren and 4 great grandsons.
Writing this potted history has given me great pleasure as I think back to all
the people I met at English Electric.
My prayer is that it will trigger happy memories for you.
|
|
John is on the extreme right in discussion with Vic Price. |
I came into Deuce Operating in 1956 at Marconi House working under George Davis. Chris Woodall interviewed me subsequently having to go to NRL Blackheath to be further interviewed by Wilf Scott. He told me as I was coming from working at the London School of Economics I should not expect to be coming into the same academic atmosphere but into a commercial world. I did however find that life on Deuce and all who surrounded me to be as academic as the university, much to my delight.
My task was to supervise the Computer installation which consisted of a Hollerith Tabulator, sorter, reproducer and Deuce, along with its peripherals of Input, a modified Hollerith reproducer and a Gang Punch for output, as my experience was 8 years of Punch card operations I found little trouble in adapting to the set up.
The Computer room was a large showroom with marble flooring and wooden lined walls just on the left of the entrance to Marconi House, huge heavy curtains over the glass panelling at the entrance shielded our goings on when pulled across from casual viewing.
Marconi House was situated next to the shell of the Gaiety Theatre, burned out during the war. The programmers were housed in the stage door room, on the next floor up, accessed through a door which adjoined the buildings. George Davis's office was on the ground floor in this area.
At a later date EE Co. decided to construct the future English Electric House on this site. In preparing the site for the new building they had a huge crane with a giant concrete ball hanging from the arm and swung it around letting it crash to the ground to bring the remains of the old theatre to complete its destruction. I remember particularly that it drove Peter Docherty (who was our Deuce engineer at this time) to destruction, for when they crashed this ball down it shook the building and put the drum backing store on Deuce out of sync and so our operations came to halt for the time being until we put a stop to this operation while we were switched on.
When this building was finally completed they had two classical naked male statues over the front entrance, the buses which had a stop just across the road was called "cobblers corner" by conductors advising passengers of their destination.
Deuce in 1956 was a 32 word length binary programmed machine and used but 32 Columns on each row of an 80 columns of the Hollerith punched card but was later extended to 64 columns when its peripherals became along with rest of the installation to IBM equipment.
The high speed memories of the machine were ultra sonic mercury delay lines the higher capacity memory was held in the mushroom and were very long lines in the temperature controlled enclosure which gave it its name as it looked like a large field mushroom, there were single 32 digit and double word lines which you could see where the arithmetic was carried out.
The Drum which was a slower magnetic storage device manufactured by Napier, a company being a part of the EE group. It was a cylindrical device 11" high and about 6" in diameter with movable heads giving access to considerably more computing space than the delay line stores. It was the only part of the machine you could see something going on. It was the pride of the engineers and myself as it really showed off the considerable engineering which went into the Deuce although most of it was electronic.
The computer had 1700 valves which were held in numerous chassis along with resistances and condensers etc, these were held in racks in the machine which must have been 30 feet long x 12 feet wide and 9 feet high, you could walk inside with comfort and engineers had access to the valves. So much heat was dissipated we did not need to heat the room even in winter.
We were not a free gift to the companies of the EE group who used our facilities and all usage was charged out at £1 per minute, I believe that was correct, each user used to book their requirements and had a company number and a job number for the particular tasks which they then recorded in a log book showing starting time and finishing time.
It was a little while before I managed to set up a punched card processing system for analysing the usage and all the users thought that they were getting complimentary usage. So after about 3 months I produced a charging method which I sent to Wilf Scott for approval NRL stared sending out the bills. It came quite a shock to the user companies to see how much they used and Marconi at Chelmsford protested madly on the subject but they still had to cough up.
The users were, prior to getting their own Deuce, Marconi, Whetstone, Napiers, TSR 2 project at Luton, RAE Farnborough, MAFF at Guildford, numerous people when they suffered breakdowns on their machines would be directed to us by engineering to provide time for them. NPL and RAE in particular. They used to take over the computer overnight.
Ron Eitel was the engineer when I first arrived who trained Peter Docherty who took over from him when he went to Kidsgrove. Tom was our Lab Steward who helped Peter and gave me a hand with some of the heavy jobs around, like storing the boxes of cards we ordered from Hollerith.
We did program writing for customers and Vic Price wrote a program for calculating Slip circle analysis and Slip slope analysis for a company called Binney and Partners who were engaged in designing very large Dam projects in India. Janet Skinner nee Porteous also helped in cooperation with Imperial College London. This became a bread and butter job for us as further dam construction projects were taken on.
The Chairman Lord Nelson also brought in some of his learned friends to see the wonders of computing, his son Lord Caldecote, the managing Director of EE whom they nicknamed half Nelson also looked us up occasionally. For convenience to save ducking down to get in his car he bought a London Taxi but he got annoyed when people kept hailing him so he had it repainted in army camouflage which made the hailers realise it was not a normal Taxi.
At the top of the building was stored a lot of Marconi's original radio equipment gigantic Transmitting valves plus other large radio pieces. The other side of the entrance hall, although only accessable from the street, was Short's bar, the owners decided to move across the Strand in which EE House was sited which left the place available to us to accomodate our offices in order to vacate the Gaiety stage door we used behind the Computer room.
All the programmers moved into the relative luxury of this new accomodation which allowed the builders to begin clearing the site next door for English Electric House.
I always believed our Deuce was the second produced and the fact that it had "English Electric" on the front must have been put on after the Jack Richardson photograph. It's only move after that was to Queens House in the Kingsway in 1961 or so and then finally decommissioned in 1964.
Later - Digital Computer Mobile Service Unit - South
He impressed me as very knowledgeable about DEUCE and was ever ready to assist with technical support from the UK. He was head [I think] of EE's Field Engineering Services (FES).
One day on the DEUCE engineer's course, Jack was passing by, and Arthur Bailey invited him to tell us about sneak CMI. Jack told us about a magnetics instruction sometimes sliping through Control without being excecuted, and how he had discovered that a particular bias point was very close to letting this condition happen, which they observed with the CRO. They issued a circuit change that altered the bias point to give a higher safety margin. - (Robin Vowels)
I was at EE London, Marconi House in the Strand, and at Queen's House in Kingsway from 1956 until the London Computer Centre closed and moved to Hartree House in 1963. I was manager of the centre from 1960 to 1963. I then moved to Kidsgrove from 1963 to 1971 then back to Newman Street in London when the ICL merger took place.
I joined English Electric London Computing Service in 1957, and left in 1963 to start a family. Originally I was based in Marconi House, but then moved with the office to Queen's House, Kingsway. In Queen's House we had separate rooms, and I shared a room with Dr. Tony Goodbody. We remain good friends, as he married one of my school friends, Ann Mountford. I introduced them to each other in 1961, and they got married in 1962! They now have three sons and are still happily married. Tony left English Electric to take up a lectureship at Liverpool University in the Mathematics Department.
My time at English Electric was very happy, and I am still in touch with some of my ex-colleagues. I was mainly working on Slip Circle Stability analyses with Dr Vic Price, but did other bits of work as well in the ‘Engineering/Scientific Section’. It was while Vic and I were working on a ‘Berthing Beam analysis’ that I met my husband, Angus. ( We will be celebrating our Golden Wedding Anniversary in the summer – how time flies!)
The only "bad day" I had (while working in Marconi House) was when I tried to retrieve a Hollerith card, which was stuck in the Hollerith Card Reader, without turning off the power. I got quite a large electric shock! It was very stupid and completely against the rules! I was lucky that I suffered no permanent damage, and only got an awful fright! I was well and truly told off by my husband, Angus!
Janet also helped in cooperation with Imperial College London. This became a bread and butter job for us as further dam construction projects were taken on. - (John Woolger)
Janet was working on a Deuce GIP version of the program. In 1959 we
both met Angus Skinner of Sir William Halcrow & Partners.
They got
married in August 1960 and are still very good friends of mine. In 1961 Angus
joined the staff of the Civil Engineering Department of Imperial College.
- (Vic Price)
Jean previously worked for English Electric in Marconi House in London as an early Deuce Programmer, later as an Instructor and then a KDP10 programmer. - (Jeremy Walker)
Pete Docherty looked after the Deuce at English Electric House in the Strand, I met him at a reunion and he had been working on the new Jubilee Line of the underground. (Noel Wesson)
(Comment on Ed Thelen's Website
http://ed-thelen.org/comp-hist/english-electric-deuce.html#Photo
Ed, I noted
that you had no price for the Deuce. In 1956/57 the price was £50,000.
I worked on the DEUCE as an engineer in Marconi House, The Strand, London WC2 from
November 1956 until the end of 1961.
The DEUCE it was by then in Queen's House, Kingsway. I left there to go on a KDP10
course at Kidsgrove and helped to commission the machine which went to BOLSA
(Bank of London and South America) where I became the resident engineer .
I have one or two circuit diagrams of the DEUCE machine.
Peter Docherty
Author of :-
Principles of Programming
Briefly describes an approach to the job of writing a computer program which
might lead to more efficient working both on and off the computer.
Standard Operating Instructions for Deuce
Operating a computer on a production run requires attention to a number of
details to ensure successful operation.
Deuce Control Panel Manual
Gives a description of the DEUCE Console, Reader and Punch and describes the
function of all switches and lights.
My name is Richard Forrester and I started work for English Electric in 1963 at the age of 16, at Kingsway in Queens House which housed the DEUCE computer. I started as a junior working for John Woolger, progressed and started operating, under supervision, the computer. I remember taking lunch at the canteen in the EE building in the Strand. EE then merged with LEO and I moved to Hartree House and eventually ended up operating a LEO3
"Mum was working at English Electric, presumably either Whetstone or Marconi house as she lived in North London, when she worked on DEUCE." - (Mark Sewards)
I've just been watching the film 'Enigma' again on BBC this evening and it has prompted me to come and browse the Web for EE's DEUCE computer. I am astonished to find so much information.
I joined EE in 1957 aged 17 to work as one of the 'Mathematical Assistants'. I only stayed for two years, deciding that the work was not for me.
I remember myself being used as a model to sit at the computer for an advert that was used in the national press!
David Ford & I were married in 1960 and had four daughters. They are now all married with children of their own. Sadly David died suddenly eleven years ago.
Reading all the old names like George Davis, Vic Price, Peter Landin, Chris Woodall, Doug Flower etc. brings lots of memories back.
|
|
Daily confirmation of margins on thermionics. |
During 1954 I joined English Electric Nelson Research. Initially
seconded to NPL at Teddington working on Pilot ACE (Automatic Computing Engine).
To prepare for the introduction of the English Electric Deuce (Digital Electronic
Universal Computing Engine).
The attached pictures were taken at English Electric House at Aldwich in The Strand, London.
This DEUCE was the first production universal computer made available to the public for general applications.
In 1958 I transferred from Nelson Research to Computer Engineering at Kidsgrove.
My late husband David Ford was recruited by Ron Eitel to backup Peter Docherty as a maintenance engineer in 1958 when he left the RAF after 5 year's as a radar technician. He went to Kidsgrove for three months to train up on servicing the machine and was then based at Marconi House.
David worked on the computer for about two years himself and I well remember the trials and tribulations the maintenance team had every morning to make sure DEUCE worked properly for the day. When VIP's came, I remember they impressed them by getting the machine to play a tune with the Hollerith punched cards.
David eventually became a successful salesman of electronic components and later on computers, spending some years at Texas Instruments, Elliott Automation, Hewlett Packard and Data General - (Jennifer Ford nee Lees)
"In Queen's House we had separate rooms, and I shared a room with Dr. Tony Goodbody. We remain good friends, as he married one of my school friends, Ann Mountford. I introduced them to each other in 1961, and they got married in 1962! They now have three sons and are still happily married. Tony left English Electric to take up a lectureship at Liverpool >University in the Mathematics Department." - (Janet Skinner)
Mathematician/Programmer - Director English Electric Mechanical Engineering Laboratory.
Mathematician/Programmer (APD). I was employed in the APD at Whetstone from October 1955 to 1965 but with a two-year break in the middle when I worked in Birmingham. During this period the so-called Good (and bad) Broyden methods, the single-rank symmetric (SRS) method (also discovered independently by Phil Wolfe of IBM, Yorktown Heights) and the quasi-Newton methods all first saw the light of day. Although I did not write my paper on thequasi-Newton and SRS methods until after I had moved on from English Electric, I did most of the work on them while employed by the company and in the paper I thanked the Company's Directors for permission to publish. I also acknowledged the help of Fred Ford, who arrived in APD at the same time as Charles Sheffield and was in our group. It was probably the most productive period of my life. After EE I went into academic life, first at the University College of Wales, Aberystwyth and then at the Universities of Essex and Bologna, Italy. I retired from Bologna last year (2003) when I was seventy.
Mathematician/Programmer [APD]. "I also acknowledged the help of Fred
Ford, who arrived in APD at the same time as Charles Sheffield and was in our
group."
(Charles Broyden)
Mathematician/Programmer [APD] Scientist and prolific science fiction
writer.
His SF short story, "Georgia On My Mind" which was first published in Analog
Science Fiction in January 1993, opens with a reference to Deuce in 1958 and is
dedicated:
"This is for Garry Tee
- who is a professor of Computer Science at the University of Auckland:
- who is a mathematician, computer specialist and historian of science:
- who discovered parts of Babbage's Difference Engine in Dunedin, New Zealand:
- who programmed the DEUCE computer in the late 1950s, and has been a colleague and
friend since that time:
- who is no more Bill Rigley than I am the narrator of the story".
Click here to read the opening chapter.
"Dear Mr. Barrett,
You have my permission to reprint that material from "Georgia on my Mind" on your
website, as described above.
Charles would have liked that.
Best, Nancy Kress" (WWW)
Mathematician/Programmer [APD]
Programmer [APD] I never met him, however he deserves mention as the originator of the most brilliant piece of Deuce programming I ever saw. This used the multiplier purely as an autonomous shift register. Holman realised that if one placed a 32-bit value 'X' in the lower half of DL 21, cleared DL 13 and the upper half of DL 21, started the multiplier, and then executed 21-26 (l) followed by 22-26 (l) (each for 32 m.c., one of them possibly 33 m.c.) the sum of the non-zero bits of 'X' would be left in DL 13.
How it works: in the course of a 32-bit shift of DL 21, every non-zero bit in 'X' appears in every bit position from 1 to 32, and in that position is subtracted from DL 13. The result is to subtract "all 1s" from DL 13, i.e. to add 1, for every non-zero bit. The effect of the inevitable 2 m.c. hiatus between the two long instructions is cancelled by using source 22 in the second instruction. Marvellous!
Summing the '1' bits in a word is genuinely useful - I believe Turing himself felt the need for such an instruction, and we certainly included one in KDF 9. - (Mike Wetherfield)
The description of D Holman's method of summing the bits in a word needs modification.
The result in TS 13 has a missing P32, corresponding to the gap between the two long transfers to D26. Thus a test is required to eliminate this bit when TS13 is negative.
I used this sequence in STAC III in the storage allocation phase. In STAC it was necessary to keep track of the number of remaining available minor cycles in Delay Lines; also the number of available minor cycles modulo 0, 1, 2, and 3, and the number of minor cycles modulo 0 and 1. Thus, this instruction sequence had to be executed before each instruction or data was allocated a location. - [Robin Vowels]
Mathematician/Programmer MEL. I began programming in Deuce binary machine code, but by 1964 I was using ALGOL and GIP as my principal languages. J. H. Wilkinson [dead link]and his colleagues wrote the GIP subroutines, which formed the basis of the superb NAG Library of mathematical software. And GIP remains yet one of the most useful computer languages which I have used. (Sample GIP coding in PDF format)
"In 1958 he went to England to get into computing - quite literally, since on cold English winter days, he and 2 others used to huddle together inside the central processor of a DEUCE computer, basking in the radiation from 3500 glowing thermionic valves. He became a mathematician with English Electric Company" - University of Auckland Bulletin
Garry proposed an interesting modification to Deuce, the addition of
a knitting machine as an output device. !
DEUCE Bulletin No.
10, 13 June 1960, pp8-12
Mathematician/Programmer APD. I am the Mike Kelly who was an APD mathematician/programmer on the Whetstone DEUCE from Aug 1957 to May 1960.
I would like to add an anecdote about our (Brian Randell and me)
mistreatment of a female fellow programmer named Enid.
She was noted for conversing with the machine when doing hands-on debugging.
Sometimes her tones were soothing (things were going well) and sometimes not! I
think I remember her slapping the console at times.
Well, our mistreatment consisted of modifying the synchronising cards
that started every program so that her program sychronised half the time randomly.
After a day or so we corrected the cards and the program ran properly again.
Of course Enid remained perplexed about what had happened and worried that it
might occur again. We never told her what we'd done.
To this day I think this 'bug' would be very hard to find, especially since
computer time was so precious.
I also remember an MEL programmer - I think it was Eric Richards - who could whistle two notes at the same time in harmony. Always useful to have a back-up skill.
There was an MEL programmer named Barry Clark. He moved to IBM, Hursley about the time I did and he wrote diagnostic microcode for the S360/40. Several years later we both transferred to IBM in America but not to the same location.
Anybody who knew me and would like to touch base is welcomed. I am now retired from IBM and, together with my wife of nearly 53 years, am a part-time antiques dealer in Virginia.
With colleague Brian RANDELL developed EASICODE at Whetstone.
And here's a story about Mike Kelly, who was I understand one of a
small number of people who essentially simultaneously found out ways of exploiting
the fact that one could "frig the mult" as we called it, i.e. interfere with the
multiplication operation, e.g. by changing operands.
He spent much of his last few weeks at Whetstone before he left to join IBM
Hursley (where he was involved in developing the microcode for one of the first
S360s) developing what he called the "ultimate DEUCE program". This was a single
64-column card which when read in, using "Initial input" (!), punched out an
*exact* copy of itself - a mind-boggling feat of DEUCE coding, that I doubt anyone
else could have achieved.
(Mike also made the crucial discovery that enabled EASICODE to be an amazingly
fast interpreter. Typical interpreters on DEUCE had an interpretation cycle that
took 50-100 instructions. EASICODE took only about 5, largely because Mike found
a way of exploiting instruction execution and modification straight out of the
quad stores, using the "17 - 0" instruction.) (Brian Randell )
Preliminary Report on EASICODE , Kelly, M.J. and Randell, B. W/AT 216 Atomic Power Division, English Electric, Whetstone, Leics., September 1958 - Complete document in PDF format here
Programmer
I was employed there (E E Whetstone) as an applications programmer (but was actually devoting all my time to compilers - or "automatic programming" as we then called it).
The Whetstone site housed the Mechanical Engineering Laboratories (MEL) and the Atomic Power Division (APD), and where there were two DEUCEs. I was in APD (1957-64), initially working on nuclear reactor calculations, but then with a colleague, Mike Kelly, developed Easicode. Mike left to join IBM, and I became head of a new Automatic Programming Section, working first on DEUCE and then on KDF9.
Thanks to Turing's design, DEUCE was typically much faster in operation than its rivals, albeit almost entirely at the expense of its programmers. Such was the innocence of youth that I and my colleagues actually enjoyed its intricacies, and the problem of finding ways of automating, at least partially, the programming task. Indeed, we felt that contemporary American computer developments, by IBM and others, such as the provision of what seemed to us to be huge memories, and of floating point arithmetic hardware, were in effect cheating. Certainly they were depriving compiler writers such as ourselves of interesting and (we thought) worthwhile challenges.
In my time at English Electric I and my colleagues learned the hard
way the importance of writing robust programs, though I cannot recall whether
the actual term 'robust' was used. However, whatever term we used, we meant
programs that could cope well with whatever strange data they were given,
whatever mistakes were made by the operators, etc. We took an active interest in
robust programming out of self-defence because we worked in close proximity to
the people who were mistreating the compilers that we were developing. In fact
we had a very effective, albeit ad hominem, 'formal' definition of compiler
robustness - the ability to cope with programs written by William White, and
key-punched by Barbara Black, running on a computer being operated by Gerald
Green - except that in what I’ve just said the names have been changed to
protect the guilty. This was, one might say, my first exposure to the need to
'face up to faults', albeit at this stage just those of other people.
[
Facing Up to Faults - 2nd Turing Lecture ]
Also check the drains here !!
"I worked on the DEUCE at Whetstone for about two years, having started my programming life on the analogue "SATURN" machine, and then discovering digital computers with the DEUCE, and later with the KDF9. Most of my work was in the Safety Section, studying the transient behaviour of the Advanced Gas-Cooled Reactor in fault conditions. Even then, we talked about the risk of an aircraft crashing on a reactor site, but we never thought about anyone doing it deliberately."
"Back in those days, when I studied mathematics at university, computers were thought undignified by mathematicians, and were the responsibility of the "dirty paws" in the engineering department.
Programmer [APD]. I would like to offer my memories of the Deuce
systems at EE Whetstone where I worked in the APD from Autumn 1963 to April
1967 as a reactor safety analyst, my first job after reading maths at St John’s
Oxford.
As a programmer I mainly wrote in the interpreters STEVE (with built in steam
tables) and BEVERLEY (with Bessel functions) as well as the vanilla Alphacode.
I learned DEUCE machine code and wrote a few oddments in this, testing the
routines myself using the single shot keys and the binary displays.
There was also a telephone dial which could be used to execute a small number of
cycles if you were feeling brave.
I remember particularly spending hours trying to programme the paper tape punch.
At the time we had three machines, one being a Mark IIA with extra delay lines.
I was an early user of the “Walgol” Algol 60 interpreter being developed in the
next office by Lawford Russell and the “Kalgol” Algol 60 compiler written at
Kidsgrove.
Towards the end of my period with APD (and its successor company NDC?)
I started to use the KDF9 initially by sending work on paper tape to Kidsgrove and
latterly by train on a nightly basis to Newcastle University, where we had the use
of their machine at night.
We maintained our own operations staff in Newcastle who commuted to the North east
on a weekly basis.
This was a hazardous procedure as the overnight turnover depended on our van
meeting a train at Peterborough as there was no direct train service from Leicester
to Newcastle.
After leaving Whetstone I moved to Newcastle University Maths Dept and ended up
operating the same KDF9.
I also remember an MEL programmer - I think it was Eric Richards - who
could whistle two notes at the same time in harmony.
Always useful to have a back-up skill. (Mike KELLY)
"I was an early user of the “Walgol” Algol 60 interpreter being developed in the next office by Lawford Russell" ( Rob THIRLBY )
Mathematician/Programmer [APD]. "After graduating in the Natural Science Tripos (Part II in Physics) from Cambridge University (UK) Roger spent three years as a Teaching Fellow at the University of Michigan. Then he joined English Electric as a nuclear physicist using early analogue and the first generation of UK electronic computers (EE DEUCE) for the design of nuclear power stations. This is the period when Roger fell in love with computers. From then on he would always have a computer in his hands for the rest of his life." - ( Vladimir Getov ) [dead link]
Website http://www.maurice.icuknet.co.uk -
Operator/Programmer. I worked at the EE MEL at Whetstone (Leic.) from
1959 until 1964, latterly in the Atomic Power Division (Mk.2 DEUCE).
My first job was as DEUCE Mk 1 operator, with fellow operator Alan Jones.
Dr. Peter Wakely was head of DEUCE programming; one of his guys was Mick
Williamson (whom I bumped into last Sept. on a walking holiday in Provence!).
At Xmas it was the custom for the staff to gather round DEUCE to sing carols, for which we had a pack of cards which played tunes on the punch relays. One Xmas someone dropped the box of cards, and as there was no time to get them sorted out the engineers slipped home to fetch their penny whistles and flutes, which they took with them inside DEUCE, where they crouched behind the console, whose lights were flickering away, thanks to a small program I had cobbled together. Very few there were aware of this first hardware emulation!
Another memory: One of Peter Wakely's programmers (who shall be
nameless) was notorious for making fundamental programming errors, and resisted
our attempts to enlighten him.
One day he asked us what the telephone dial was on the control panel (the one
used to signal 'n' Single Shots to a program under test that had 'stoppers' on
strategic instructions).
We kidded him that it was for calling out the engineer, so he dialled in Bill
Worth's number, and we were astonished to see Bill Worth appear shortly afterwards!
I don't know if he ever discovered the truth...
Later on at Whetsone I joined Brian Randell (later Prof. of Computing at Newcastle Uni) and Lawford Russell in writing the first ALGOL compiler on DEUCE, in conjunction with the Oceanographic Dept. at Liverpool University, who wrote the runtime package. Copies were requested from all over the world, e.g. Sydney Uni. Our chief there was Dr. David Parkyn. Randell & Russell wrote a book on 'ALGOL 60 Implementation', which I helped to proof-read."
One incident I remember vividly is the episode of the "drains", which Brian Randell's article mentions. On that day, when we realised that the sewer had interfaced with DEUCE, one of us (one of the engineers, I think) hung a toilet roll on a piece of string on the rear door handle. Seemed appropriate!
"In reading through your website I came across Monte Carlo. I'm sure I remember this software because it was the only time when operators could take a long rest. It calculated for about one hour and then punched one card. Sometimes the program was allowed to run for four hours and we were very happy. There was a great temptation to have a kip."
"I hope to keep you informed of anything interesting. It seems you are heavily involved with Deuce and I am very glad that you are continuing to keep it up. I enjoyed my time there and only left because I could not get into the engineering. I eventually went back to EE in Kidsgrove (as an engineer) and I have remained in that area, now retired."
"There were eight operators working a 3-shift system and we had about twelve or so programmers. If any more memories come to the surface I will relay them to you."
Pete Berridge was a constant problem for Bill Worth. One day he brought his Vespa scooter into the computer room to clean it. This was on a late shift so not many people saw it. Somebody snitched however and Bill went a little wild. It was not a clean room of course because we had no mag tape at that time. Pete stayed but didn't repeat that trick. He eventually went to South Africa. (Seth Holt)
DEUCE Maintenance Engineers
Chief DEUCE Maintenance Engineer EE Whetstone. All I remember of Whetstone was Bill Worth, who was in charge and very hands on. He made some gold tipped wiper arms for the drum that lasted much longer than usual. (Noel Wesson)
Noel Wesson mentioned Bill's drum wiper arms. He did not mention Bill's clock. The potentiometer on the drum wore down with use, so Bill designed and, I think, fitted a slow motion clock which turned once every 24 hours or so. This spread the wear out around the pot wiring. I don't know what happened to this modification but it sounded interesting. Bill also was designing what I thought was another output for Deuce. This was an ICT 030 punch I think. (Seth Holt)
With referenc to the chief engineer, Bill Worth, another memory is of him bustling in when called out on a problem, bringing with him a rubber hammer, with which he would bang on the mercury memory 'mushroom' to check for an intermittent connection problem. (Maurice Batey)
I can add to the story of Bill Worth's slow motion clock. Though this spread the wear on the potentiometer, after a while it was recognised as the culprit behind a mysterious transient drum fault which occurred for a brief period at the same time each day, when a worn area of the potentiometer came under the wiper arms. The clock's use was then abandoned. (Brian Randell)
Another Bill Worth story - he used to rail against programmers who repaired cards by sticking chads back into holes. (You were supposed to do this just in order to immediately produce a new card using the reproducing punch - and to mark any reinserted chad with a pencilled cross, so that you could see if the chad had dropped out.) He objected violently when he found chads in the bottom of the card reader - and always pointed out that he could tell that they were the result of programmers fitting them back into cards, and ignoring the rule about using such cards only on the reproducing punch, since the chads had crosses on them. I challenged him to prove that any were mine, because my practice was to draw a circle around any reinserted chads! (Brian Randell)
Punch Room Manager
I was just talking to a friend about the early days of computing and thought I would see if I could find an image of the DEUCE computer. That is how I found your excellent website - and have been enjoying a trip down memeory lane.
In September 1956, having completed my National Service as a Radar Fitter, I went to English Electric, Whetsone to see if they had a job available to suit my experience (I also had A level in Mathematics and Physics). I was offered a job as a 'Mathematical Assistant' in the Mechanical Engineering Laboratories. MEL, along with the Atomic Power Division, was awaiting the delivery of a Deuce Computer; when it arrived I was supposed to be part of the maintenance team (working under Bill Worth - who was still at Kidsgrove at the time, where he was learning about the Deuce computer and gaining hands-on experience; I believe his background was in TV repair).
I recall the decision to purchase a Deuce computer was based on a survey of staff in the several laboratories within MEL, combined with the expected needs of the Atomic Power Division. The conclusion was that perhaps the computer would be used from 9 to 5, 5 days a week! Of course, hardly anyone had any idea just what you could do with such a machine; this was uncharted territory. The reality was, of course, that shortly after it did eventually arrive it was in use all day, every day.
I say 'eventually arrive', because by the time it did, I had got fed up waiting for it. I had been hired with an arrangement of working part time in the Control and Servomechanism Laboratory of MEL, alongside the DEUCE work; this part-time arrangement became a full-time one in CSL. Before DEUCE arrived, Bill Worth came back from Kidsgrove to get things set up; I can identify with the things said about Bill elsewhere on your site. During that time I learned some machine code programming. I was sent on a programming course at Cambridge University - along with some notable engineers, and scientists who were learning about computer programming for the first time. We learned to program on EDSAC I; EDSAC II was in the process of construction. I can still see the room where the ferrite core matrix was being hand wired! I was very envious of Ferranti people on that course; they talked of 'pseudocodes' which saved an awful lot of effort in programming. ALPHACODE was still in its infancy; we had to use machine code - which was, of course, less wasteful of precious memory.
Because of the delay mentioned above therefore, I never actually worked on DEUCE - but I knew several people mentioned on your website. I was particularly interested in the piece on David Holman, who died a few years ago. David joined about the same time as I did and we became friends. I am not sure it is fair to all mathematicians to call David a 'typical' mathematician, but alongside his undoubted brilliance, he was a man of very few words.
Head of GW Division
Senior Programmer. "I devised the first "Read eight 8-digit integers" subroutine, R24T - and later, after John O'Brien (Marconi) produced an improved version (R24T/1), I inevitably had to go one better and produced R24T/2, which used even fewer instructions." (Mike Wetherfield)
My name is Harold Fineberg and I worked at English Electric, Luton,
from September 1956 (immediately after graduating from Bristol University with a
degree in Physics and Mathematics) until I took leave of absence in August 1959.
I was engaged initially as a Graduate Apprentice, which meant that I spent about
3 months at a time in different departments to enable me (and management, I
suppose) to determine where I might fit best.
One of those periods was in the computing department of which Winifred Hackett
was the head.
I learned how to use the mechanical calculators (Friden and Marchant come to
mind) and it was while there that I heard about DEUCE.
They even had a manual that described how to programme it and I knew that I had
found my place. This would be sometime about 1957.
I decided that was where I wanted to be and so officially joined the department,
working under John O'Brien.
Read on to find out how Harold became known as "The Music Man"
The manual itself didn't bring back memories (except by reminding of the syntax!), because I never saw one when I was programming DEUCE. The department at what was then EE (GW Division), Luton, which later moved to Stevenage, (or, rather, its head Dr. Winifred Hackett) considered that to be a programmer it was essential to have a maths honours degree, while I was a mere physicist turned system engineer and user. In consequence, I never had any formal training nor ever saw a manual. I made use of a set of lecture notes given to me by someone who had been on the proper training course, and taught myself from that. It came as quite a shock to the chief programmer when he discovered (several years later!) that I had written quite a few programs, including some quite large, complex ones. It was much faster than waiting for the official programming team to produce what I needed for my engineering calculations.
I was at English Electric Luton from 1956 until I moved to Stevenage, around 1964. Stayed at Stevenage, until we left for OZ in 67.
I was an Electronics Engineer, then Systems Engineer. I did not work directly on DEUCE, but did some programming. As I recall we had some evening courses in Digital Computing, which at the time seemed total mumbo jumbo to me. So I did nothing with it, I think initially the coding was in binary, then in a pseudo code (the name escapes me [Alphacode JB]). At that stage I got involved and wrote a few programs or maybe adapted some. I recall the card punching was done by some girls on keyboards, to coding sheets, but I do recall the almost impossible task of hand punching cards - I found it almost impossible to punch a whole card without making a mistake.
One thing I do recall about those early days was the preoccupation with with computing precision - reversion to double length arithmetic, manipulating equations to ensure that there was no embedded precision errors - all a thing of the past, no one gives it a thought now.
"My immediate boss then was Ron Stokes who had some DEUCE experience but I don't remember where." (Peter Stanley)
Deuce Maintenace Engineer. Came across your web page by accident. Very interesting! Certainly stirred some memories.
The Luton DEUCE was delivered in 1958 and commissioned by Frank Thompson. I looked after the machine from 1958 until Luton closed in 1962, when I moved to Kidsgrove, did the KDF9 course and was in charge of Systems test. The engineer in charge then was Roger Bird. I have no idea what happened to him.
I remember Jeremy well. He was the man who started the build of the Luton DEUCE on 1 April 1958. His first comment in the log was "What a stupid day to start building this machine!" He was also the man who entered the complete "buzz and go" maintenance program by hand through the front panel keys! At Luton I reported to Ron Stokes (I think that was the surname), but the computer was "owned" by Doctor Winifred Hackett who I see gets a mention on your web site.
Regarding the Luton LACE. As you probably know this was designed at Luton and I believe it was at the time the largest Analog(ue?) computer system in Europe. I had a little to do with that but the main men were Mike Brown and our joint head of department Roger Grimsdick.
More memories of EE Luton from Peter & Joan Stanley here
Married Peter Stanley
"My wife also worked at Luton, initially as an operator for DEUCE, but later as a
programmer. When Luton folded she went on to Elliot's and then Leo and finally
Kidsgrove."(Peter Stanley)
I was an Engineer at both Luton and Stevenage together with Peter Stanley and Roger Bird. I well remember Kidsgrove / Arthur Bailey and many of the other names in your listings.
"The DEUCE did actually go to Stevenage after I moved to Kidsgrove. The engineer in charge then was Roger Bird." (Peter Stanley)
The first computer I "used" at the National Physical Laboratory was Deuce, a later re-engineering of the 'Pilot model' of the first machine, the Automatic Computing Engine, ACE (essentially Alan Turing's design): Deuce was simpler and entered service before the full-size ACE.
"Used" is in quotes because Deuce was protected from its customers by a young lady who would punch the cards from handwritten input (experimental results in our case, recorded by writing meter readings on a clipboard). At one point we were Fourier-transforming some correlation data for turbulent flow (SLOW Fourier transforms in those days, the early 1960s). One set of results, instead of looking somewhat like a 'bell curve' (Gaussian) was dominated by a single cosine wave. It was obvious that a card had been mispunched with a large error, amounting to a spike in the input (FT of spike is cosine wave). We worked out from the wavelength that it was the 41st card, so we trudged over to Deuce, which was of course on the other side of the laboratory grounds, and accused the lady operator of a misplaced digit. However the 41st card was OK, so we trudged back to the office and found a slight error in our arithmetic - the sine wave corresponded to a mispunch in the 82nd card, not the 41st. So we trudged, etc ..... The operator was a cheerful and personable young woman: I had no social (or physical) contact with her whatsoever, but she was the heroine of by far the most detailed erotic dream I have ever had. Perhaps I was really lusting after the computer.
Towards the end of my time at NPL we got dial-up Teletype terminals linked to a commercial 'time-sharing' service (no screen, just paper or paper tape output) and we did an successive-approximation calculation of the turbulent flow over a swept-wing leading edge. This involved using the "shooting" method on two coupled nonlinear ordinary differential equations (one for spanwise motion, one for streamwise). The hunt for converged solutions for both equations was the nearest thing to a blood sport that I have ever experienced!
Website Dead link.
"After a preliminary venture on the prototype Ferranti Pegasus computer, he turned his attention in January 1956 to the English Electric DEUCE computer which had been installed at the National Physical Laboratory. Very quickly he produced a string of programmes for complex crystallographic calculations." (Durward Cruickshank)
I did some programming for the DEUCE computer at NPL in 1959. Does
anyone remember the DEUCE Maintenance Engineers name?
I still remember the operators phone number, it was (obviously?) 248.
ian@huth Website
"I played my first computer game some 45 years ago when I worked for the DSIR. When they had open days at the National Physical Laboratory they used to run a driving game on the ACE computer (fastest in the world at the time with its 1 MHz processor). You should have seen the size of it with not a transistor in sight but thousands of valves. The software for it was all on punched cards with a couple of switches rigged up to move the car (a single lit lightbulb). The idea was to keep the car on the road (between two parallel lines of lit lightbulbs in a giant grid). We also had DEUCE which was a production version of ACE with a lot of the components in cabinets rather than the open frames of ACE."
RAE members are listed on the Royal Aircraft Establishment page
Head of Maths Services Department
Chief Programmer. Primarily responsible for the development of TIP
Programmer. "I started out with BS in 1960 in the Maths Services department. The term 'Computer Department' was reserved for those lesser mortals who did uninteresting things like payroll on uninteresting machines by IBM."
"TIP (Tabular Interpretive Program) was the brainchild of our little cell and used extensively in scientific calculations. "
"One of the most interesting projects I worked on was to support a paper written by Dennis Boston on artificial intelligence. We applied an iterative adaptive process to the design of turbine discs where human judgement was input after each design iteration. And all this using TIP on DEUCE! "
"And one of the least successful, a program to calculate stress and creep in turbine discs using a very complex set of equations that had just been published. I remember writing and debugging for a long long time until it appeared we'd got it right. Alas, the whole theory was debunked and the program turned out to be little more than a complicated random number generator!"
"Looking through your site certainly indeuces (sic) a deep sense of nostalgia and I can once again imagine myself punching out binary cards and single shotting my way through programs where the displays of binary patterns were instantly translated into something meaningful. God knows how we did it."
"When I've tried to explain to latter day programmers the notion of NIS, wait and timing numbers, multipliers etc, I seem only to induce strange stares of incomprehension."
Programmer. "Thank you for a fascinating website."
"I programmed the DEUCE at Bristol Siddeley Engines in Patchway in the period 1962 to 1963. I was a recently graduated engineer working to create "Scheme T", the first Bristol Siddeley true specific heat part load engine performance program."
"As I recall, there was doubt about whether the DEUCE was large enough. However, I was young and fearless and somehow I managed to cram everything into a program that would handle all the engine types that Bristol Siddeley then had or planned to have (days of the Olympus, and the BS 100)."
As an engineer, I was not supposed to program or gain access to the machine and initially needed to be escorted by Dickie Burden (a wonderful man). Eventually, Maths Services figured out that it was easier to adopt me than to keep me out, so I moved into the Maths Services area with folk like Brian Miller. They were a lively bunch and I had a lot of fun with them."
"I left in 1963 for graduate school in the USA and retired ten years ago from Northrop Grumman. I look back with affection and pride on my work with the DEUCE. I am as proud of what I did then as I am of the things that I have done since."
Programmer. Dickie Burden was a DEUCE programmer who worked in Maths
Services (I think) but was closely linked with folk like Prem Gupta, Brokie
Brokenshaw, and John Uden in engineering analysis. Dickie was a paraplegic from
polio contracted in India during RAF service.
He was one of the most amusing and positive people I have ever met and a great
inspiration and help to me. (Brian HUNT)
Gosh, surprised to be sent a link to your site.
Having joined Maths Services at Bristol Siddeley Engines in 1964 I spent a good deal of time, initially punching cards and I do remember running the 'man-hours' programme.
Administration was looked after by Steve Mieville, who also did some operating and then there were Mike Oakes, Wendy Breward (Higby) and me as trainee operators.
Other staff I remember at the time were
Diana Sprague - Punch Room
Heather Francombe - Punch Room Supervisor
Jenny Wilson (Reason) - Punch Room
Bob Sinclair - Programmer and later to be our very popular boss as operations
manager.
Eric Griffiths - Ex Ansty (Near Coventry) - Operations Manager
Joyce Fey - Some programming and operating as I remember
Liz Hand (Ball) - Operator
Jeanette Grant (Forgotten her married name) - Programmer and Operator
Mike Warke - Operator - who you already have listed.
Bob Griffin - Operator
Mike Warke, Bob Griffin and Liz were a bit of a whizz with the
printing side as well.
Those plug boards were a nightmare as I remember. I was only young then however
and easily impressed.
Deuce Operator. "I have just read your description of the DUECE computer, I operated one of those in1958 when I was employed by what was then called Bristol Siddeley Engines which then became Rolls Royce. There were two of us on a night shift, four 12 hour shifts a week. The biggest chore was the feeding in of all the cards to start the thing off. Also a very long restore control routine should the m/c go into a loop, doing the same thing over and over again. One benefit of the m/c was that if you felt like a nap during the night you could open the door at the back and you could hear the sensors moving up and down, if they stopped you woke up and sorted it out. Should it break down it was a trip round the m/c tapping all the pull-out trays and hopfully getting the valves to come to life again."
Deuce Operator. A friend of mine who worked as an operator at Bristol Siddeley on Deuce and then went on through Kidsgrove to higher things. (Seth Holt)
I was Mike Warkes Best Man! (Paul Hemming)
"When I joined the company in 1953 it was known as The English Electric Company (Aircraft Division) but by the time DEUCE came along it was English Electric Aviation Ltd. At the time of the TSR 2 contract we merged with Vickers and others to become British Aircraft Corporation. I think that was before the DEUCEs were phased out." (John Halliday)
"Although it was before my time (I joined in Jan 67, from GEC in
Coventry) DEUCE was still being talked about when I started there. .
I never saw DEUCE or knew where the DEUCE room was!" (Cliff Elliott)
Cliff has a website for Warton Windtunnelonians here
"Though he was Head of Electronics, he was never personally involved with DEUCE, other than giving us his blessing, approving expenditure, and conducting VIPs into the room to gawp on their tour of the site facilities." (Steve Allcock)
"Tom was recruited by Bill Coulshed and had as his mission, bringing the benefits of automation to English Electric. Tom later introduced numerically controlled machine tools to the factory at Strand Road, Preston." (Steve Allcock)
Head of Maths Services. "John was one of life's great gentlemen, as well as being a first class mathematician." (Peter Dukes)
I was at Warton in the period 1957-1963. English Electric had three principal technical groups using the DEUCE, -- Mathematical Services, Aerodynamics and Stress Analysis. I was in the first and David (Booker) was in the second. However, I think we would both agree that when it came to pushing out the envelope, the Stress Office were at the forefront of the application of matrix methods and the early finite element developments.
Read Peter's fascinating memories of mathematical manipulations and midnight mercury hunts to keep the Deuce running to complete Concorde's fin stress analysis. DEUCE Recollections - BAC Warton - 1962-63
"Was doing good work at Warton before he joined Argyris at Imperial College." (Peter Dukes)
Probably did some of the first structural optimisation FE analysis (Lightning Fin) by computer in the world, and that was on the DEUCE. (Peter Dukes)
John recalls his early days of programming, before BAC Warton had their own Deuce, when a few minutes "hands on" use of the computer at NRL was very precious.
" We used to travel down to Stafford in the morning, have lunch at the
Swan hotel,
and then go out to Blackheath Lane. If you were lucky you might get three
ten-minute hands on sessions on the computer during the afternoon and early
evening. In this time you had to get as much information as you could from your
program testing runs in order to debug your program during the rest of the week
when you were back at Warton"
Read John's memoirs
"DEUCE to KDF9 at BAC Warton"
"John H had been involved in matrix analysis work for a long time and wanted to explore the usefulness of approximate methods for the solution of large numbers (e.g. about 100!) of simultaneous equations." [ See Peter Dukes - DEUCE Recollections ]
"One of the team, Dick Young, was a keen musician and talked of entering an international competition for computer music to be run in Monte Carlo in 1963 (?). I recall he programmed one of the Warton DEUCE's to play tunes and accompanied it on his trombone at Christmas 1962." [ Peter Dukes - DEUCE Recollections ]
Website [dead link]
Extracts from an interview by Prof. Nicholas J. Higham at the University of Manchester, March 15, 2005. Full text in PDF format available here [dead link].
NJH: And I know that after your undergraduate degree in mathematics at the University of Liverpool you worked in the aircraft industry in the north west of England.
PL: I was in the aero-structures group at what was then called Warton Aerodrome and was the research arm of the English Electric Company, which later became British Aerospace.
NJH: What computing facilities were available?
PL: Desk-top calculators (Monroes and one or two Brunsvigas) were the standard equipment to begin with. However, the English Electric Company had its aircraft division in Warton (near Preston) and a digital computer division in Stafford. So we would occasionally travel down to Stafford to do our calculations on budding highspeed digital computers (the DEUCE). Of course, at that time, we could use only machine language. There was nothing else — no high level language. This probably turned me away from writing programmes for life. Nowadays I am happy to use a high level language, MATLAB in particular, but I lost my taste for programming in those early difficult years.
"I joined Ian Kerr's Stress Team which was located in the so-called
Fish Tank in the Stress Office in September 1965. After being taught programming
methodologies at Imperial College I was amazed to find that Ian's method was to
sit with a hand punch and a pile of cards, writing another program for the Deuce!
The flow diagram was kept in his head!
We still used the Deuce for some problems. I remember in particular running the
Resisting Moment program on Deuce to calculate the failure bending moments for the
Concorde fuselage. This must have been one on the last uses made of the Deuce.
I greatly enjoyed working for Ian Taig, who was so encouraging to his stress team.
I left in 1967, following in the path of Sydney Kelsey, to work at Imperial under
Argyris."
Programmer in Aerodynamics department who married Josephine LLOYD
"Phil Roberts was a very important person at Warton, being the expert on Scheme B. He was located in Stress Office, not directly with John McDonnell." (Steve Allcock)
I started in Maths Services in July 1961, was a programmer initially
on the Deuces for 4 years then became a programmer for the IBM 7090 in London
before becoming the Computer Manager for the IBM installation at Warton in 1967.
Will be in touch further when I've reloaded my memory file!!
|
View of the DEUCE room with me at the console. |
My Name is Michael Caine and I worked for English Electric Aviation at Wharton from 1960 to 1968.
I started as a Deuce computer operator, graduated to programming the KDF6 when it arrived around 1965, based at Strand Road Preston, ending up as the KDF6 Operations Manager.
We operators were not mathematicians, but we did learn how to apply logic to problems, work to strict standards, simple programming, be self reliant and at the same time have a lot of fun.
The operators at Wharton worked 24/7 alternating day shift and night shift and there was always plenty of weekend work. Some of the names were Ron Broxham, Brian Petrie, Michael Mallinder, Pete Riley, Bas Drummond, Mike Flemming and others I cannot now remember. I do remember we had a great time, with real gentlemen as bosses such as Gordon Pitt, John McDonald and Tom Duerdon, who gave me my first management job.
I remember playing cricket at 3.00 am just outside the computer room and someone putting the ball through the MD's (Freddy Paige) window. The following morning he came down to ask who did it. The culprit, I forget who, took one pace forward and owned up, expecting the sack. Mr Paige said "If you cannot bat straighter than that you shouldn't be playing cricket" threw the ball to him and walked out.
Just been looking at the photograph of part of the DEUCE console you
attached. It reminded me of the first time I ever read the DEUCE user guide. At
Wharton if there were any jobs which needed little attention from operators,
these were always left for the nightshift, to make their task easier. If fact
there were some jobs which only required a few punched cards as input, but the
resultant processing could take hours, sometimes all night. This meant that we
had nothing to do so we slept. (Or played cricket! JB)
Our DEUCE had a small toggle switch. When you looked this up in the users guide
it said "Switch for alarm bell" For use by dormant Operators at Wharton. If this
bell was set, a very loud alarm would ring if the computer stopped for any
reason, waking us from our slumbers. As far as we know, it was only available on
the DEUCE at Wharton.
However, what I learned at Wharton working on the Deuce was a boon for the rest of my career. We operators had to learn to program in machine code, logical flowcharting, GIP and correct any programs that went wrong on the night shift as there was no one to turn to for help.
Operators at Wharton had to be able to do some limited programming, nothing special. At the time, I was after a Programming Team Leaders job at Elliott Automation Ltd. My CV at that time showed that I had been successful in programming the KDF6 and had done some programming of the DEUCE. One of the three people interviewing me said to his colleagues "If this guy's programmed a DEUCE, he can programme anything", and I was offered the job there and then.
I had a great time at Wharton and will always remember it.
As I came across your site I thought I would add myself to your
"people" index.
My maiden name was Mary Jones and I joined Maths Services at Warton straight
from Manchester University in 1958.
Just after I joined there were seven people in the group. I was there for seven
years.
It was headed by John McDonnell. I joined Dick Keyser, Roy Smith,
Jim Cruickshank, Bob Galloway and I think the final person
was Joyce ?
I regret to say that at the moment I cannot recall her last name. I have enjoyed
reading the articles on the site. It has taken me back down memory lane. When I
have time I may be able to put a few notes together myself and send them to you.
[Mary Morris nee Jones]
"Handled the output." (Steve Allcock)
John Halliday remembers many of staff at BAC Warton:
Ron GREEN --- Roy SMITH --- Ron BRADLEY --- Philip
TATTERSALL --- Alan PEACOCK
Gordon PITT --- Alan JENNINGS --- Eddie GREEN --- Colin BARKER --- Barry (Bas)
DRUMMOND
Ronald GRAHAM --- David BECK --- Peter LEAKEY --- Oliver WHITAKER --- Barry
O'NEILL
Brian WARDMAN --- Philip BRIGHTLING --- David COPSON --- Bill COLES --- Philip
COATES --- Ron THOMAS --- Gordon SUMNER Richard KEYSER --- James MALLOCH
I was the DEUCE Maintenance Engineer at English Electric Aviation (name subsequently changed to British Aircraft Corporation, British Aerospace, etc.) at Warton Aerodrome, near Preston.
I was alerted by John Halliday to the existence of your website, I knew a remarkable number of the people you list.
It was evocative to see the pictures of the twin machines at Farnborough, and to read again what was called a programming manual.
Read Steve's extensive Memoirs of Maintenance Modifications and Marriage at Warton.
DEUCE Maintenance Engineer BAC Warton
DEUCE Maintenance Engineer BAC Warton
DEUCE Maintenance Engineer BAC Warton
"Steven Allcock also had a Canadian assistant whose first name was Alan but I cannot recall his surname" (John Halliday)
DEUCE Maintenance Engineer BAC Warton 1961-1964.
"I was recommended to check with the maintenance engineer, each time I wanted to
do a three hour matrix manipulation job, whether he thought the machine would run
for that length of time! " (Peter Dukes)
As Robin's machine in Sydney did not have magnetic tape. He spent three weeks of the Deuce training course at Warton where two Mark I DEUCE were operating side-by-side.
"The engineers had installed a facility so that words in the monitor(s) could be conveniently viewed with a space after every fifth bit, and with spaces after each field of the DEUCE instruction. I forget whether this was achieved with an additional switch, or whether it was an additional position of an existing switch.'
"The site engineers wanted ICT to make a modification to the card reader (October 1962). For this purpose, I had put DEUCE on Request Stop (9-24, clear read) and DEUCE was stopped on that instruction. The cover of the card reader had been removed. An ICT engineer was inspecting the reader, and had his hand clasping the drive belt from the 1/8th HP motor."
"One of the site engineers noticed that DEUCE was on Request Stop. For some unaccountable reason, and despite at least three people standing behind the reader, he decided to release Request Stop. The card reader started immediately (the belt moves virtually to full speed instantaneously)."
"I never saw anyone move as quickly as did that ICT engineer. He was just so lucky !"
The DEUCE a user's view
"Jo was a Stafford employee and was seconded to Warton to help set up and run the punch card team. She later transferred and later still married Maurice Marvin who was a programmer in Aerodynamics department at the time." (John Halliday)
|
|
Barbara Salisbury operating the 32 Column Deuce at BAC Warton in 1958 |
"Barbara was Jo's No. 2. having transferred in from Flight Test where she had been a Mathematical Assistant -i.e- a computer, doing much calculation by turning the handle of a Brunsviga machine. These two girls led the half dozen who operated the DEUCE for much of its production work." (Steve Allcock)
"Strayed on the site whilst reminiscing about the 50's Deuce room which I worked on from time to time and was amazed to recall names I had forgotten. I was there from leaving school in 1957 till around 1965."
"Margot Carey provided many of the names of the people who worked in the off line activities associated with the Deuce." (John Barrett)
Head of Maths Services
Head of the DEUCE programming group
I worked as a programmer on DEUCE at British Aircraft (or whatever it's called now) at Filton, Bristol, from 1958 until I left to join IBM in 1962. The programming sheets on your site bring back fond (?) memories, and will enable me to show my son what real programming was about! I was also familiar with Gert and Daisy at Farnborough as we occasionally used them through the night on a Friday when we had a lot of work to process. We were mainly running simulations of the homing phase of the "Bloodhound" missile and had to run it many times with different noise patterns. I am always boring young people today and telling them how computers used to be so big the cleaner used to keep her buckets and mop inside!
From 1958 to 1965 I worked as a programmer in Mathematical Services Dept of BAC Filton, UK. Both the Aircraft Company (at Filton) and the Engine Company (at Patchway) started with just 1 DEUCE but later acquired a 2nd machine each, to help keep up with the huge amount of work we were trying to do. All 4 machines were run 4x7.
I worked on DEUCE between 1960 and 1966 although most of my time at Filton was taken up as Head of the Analogue Computing Group. I have fond memories of my time wrestling with DEUCE machine code. It was like being paid to do difficult crosswords. After leaving BAC I worked for computer companies before finishing up in the flight simulation industry.
I joined the BAC, Filton, in 1960 as a DEUCE programmer, having learned a smidgeon of Alphacode and machine code from the late Jack Cole, then of Queen's College, Dundee, in what must have been one of the earliest undergraduate computing courses in the UK.
In 1968 I moved to Oslo to join the SIMULA compiler implementation
teams at the Norwegian Computing Centre.
My knowledge of ALGOL 60, gained at the BAC, was a solid foundation for my work
on SIMULA which is generally credited with being the first of the breed of
object-oriented programming languages.
DEUCE was a vehicle for great ingenuity in programming. I recollect having fun at the expense of John Kelleher, our resident engineer. We devised a loop consisting of a sequence of instructions designed to illuminate for the minimum time the minimum number of the console lights. This was planted into a genuine program and, when entered, the apparent total visual absence of activity gave the impression of a drop-out which left the engineers scratching their heads until the host program unexpectedly sprang into life again.
In the 1980s, there was an essay about quiche eaters using Pascal while real programmers worked in FORTRAN. In the 1960s, quiche eaters would have used (slow) floating point subroutines while real programmers performed (fast) calculations using intricate scaling to preserve precision. One of the pleasures was to watch the convergence of iterative routines in the binary displays of the CRT.
In 1972, I joined Newcastle University, working for several years with Brian Randell, ex-DEUCE programmer and implementor of the Whetsone ALGOL compiler for the KDF9.
His claim to fame was his initials, DCW, which are, of course, the
three central fields of the instruction word.
His wife also worked there, but I can't recall her first name. [Peter Groves]
Programmers
Punch card girls
>
EE DEUCE Maintenance Engineer BAC Filton. I joined E.E. in 1958, and went to Kidsgrove to do my course under Arthur Bailey.
Finishing the course I was moved to the commissioning lab where there was a row of machines waiting to be commissioned. Behind us, at ceiling height, was a long rectangular trunking with vents at intervals and a fan at one end to blow cooling air on us. A favourite trick at the time was filling a bag with chads from the punch, switching off the fan, tipping the chads into the top end of the trunk and restarting the fan! Whereupon all the engineers would get showered with chads!
My first real assignment was the single DEUCE at Bristol Siddeley Engines at Filton. Soon after arriving I was left on my own. The machine was in a single storey room with a window in the roof. Summoning up courage I decided to stay late one night and do some work. Having switched off the machine I boldly approached the chassis with a hot soldering iron. Suddenly there was a flash and bang which made me soil my pants. Then I realised this was caused by a thunderstorm outside!
The BAC machines were working on airframe design leading up to Concord. The BSE [later Rolls Royce] machines were working on the engines for the jump jet.
Incidentally we also had responsibility for four EE LACE analogue machines used for Concord cockpit layout and control simulation.
There was another machine at BAC up the road and later both were doubled up and all began working full shifts 7 days a week. I became site supervisor and had a staff of 17 engineers. They were all EE engineers shared between BAC and Bristol Siddely Engines.
|
|
Bob Mc Call at the Console with me (Alan Gray) standing by the m/c intelligently inspecting a waveform on the typical oscilloscope provided for maintenance purposes. The photographer thought it would help the composition so Bob is wearing my sports jacket whilst I am shirt-sleeved by the 'scope. |
EE DEUCE Maintenance Engineer. I have just discovered your DEUCE
site. You have put in a lot of work and produced a valuable resource. There is
one omission that I spot and that is ME: Alan Gray [not to be confused with Reg
Gray whom I knew].
You have a photo of the August 1958 Deuce course. I attended mine in September
1958 and in due course ended up with John K at BAE Filton.
From there to Bristol Siddely Engines but you do not seem to have listed the EE engineers for that site except for David BACK. Dave was there for six weeks, I believe, before leaving for a KDP10 course prior to Exeter
For a number or reasons I am very grateful to my DEUCE years; one reason being that it got me and my family a three year tour of Buenos Aires at EE expense installing KDP10, KDF8 and recruiting and training local maintenance engineers for the Bank of London and South America.
EE DEUCE Maintenance Engineer
EE DEUCE Maintenance Engineer
DEUCE Maintenance Engineer. He worked at CEGB Deuce and was on my Deuce course (Noel Wesson)
"Director of Research at Great Baddow - later Sir Eric Eastwood and President of the IEE." (Bernard de Neumann)
"Led the Mathematics Group, went on to become Prof of Electrical Engineering at Cambridge." (Bernard de Neumann)
"Led the Programming Section under Peter Brandon." (Bernard de Neumann)
"Led the the Mathematical Physics and Circuitry Section under Peter Brandon." (Bernard de Neumann)
"Dr Doug Shinn was a Radio Astronomer from Cambridge who was an
expert on reflector antennas (like Jodrell Bank). At Cambridge he programmed
and used EDSAC to solve reflector antenna problems. He joined Marconi at Great
Baddow, and programmed DEUCE to handle reflector antenna problems."
(Bernard de Neumann)
When I joined Marconi in 1961 there was a Mathematics Group led by
Peter Brandon.
I was in Josef Skwirzynski's section as a mathematician. I used to design and
program algorithms for evaluating special mathematical functions, and also
wrote programs to evaluate multichannel telephone systems, antennas, and
electrical filters.
"Marjorie worked on the DEUCE at English Electric House on the Strand and moved to Great Baddow because of her prior experience with the EEH Deuce. Wrote a PIP ( Polynomial Interpretive Program) interpreter." (Bernard de Neumann)
" Ed worked on antenna and circuitry software and developed general tools." (Bernard de Neumann)
"Denis wrote musical and games software for the DEUCE that was brought out at Xmas time. Usually involved in 3D analysis software." (Bernard de Neumann)
After graduating in general science, he began his working life programming the English Electric DEUCE computer for Marconi Ltd. There he researched in radar propagation and various defence systems as well as computing. (From http://www.health-informatics.org/ but since removed.)
Worked under NH on finite element techniques (Bernard de Neumann)
Ted did a lot of work on inventory control and auto-stock ordering. His first program, which had been 'thoroughly' tested and was in live service, ordered 100 times as many transistors as were necessary. At the time a very expensive mistake. (Bernard de Neumann)
Elizabeth led a team of young women programmers in NH's section. She acted as a kind of chaperone until the girls were wise to the ways of the world (Bernard de Neumann)
I have been looking at your Deuce People page. I have very fond
memories of the Deuce. I worked at Marconi at Great Baddow, Chelmsford, Essex in
1966 as a "pre-university trainee", during which time I wrote programs for the
Deuce in Alphacode for processing radar data, producing histograms and the like.
It was my first experience of programming (and operating) a real computer. It
was great to go into the machine room to watch your program running, or step it
through to debug it.
I worked in a room with three women graduates. I've forgotten their names, except
one of them was called Judy. We had to clock in and out using a machine that
punched your card. Our section was headed up by a man called Neville, I think.
During my time at Baddow, a KDF9 was installed there, running something new
called an Operating System. This was POST. I remember attending a talk introducing
us to the concept of an Operating System. I got to write programs in Algol 60 for
the KDF9.
Thank you for the website. It brought back happy memories.
Operator. I have enjoyed reading your Deuce page, it brings back old memories.
I worked from 1962 until 1964 as a Computer Operator on the Deuce Mark II installation at Marconi's, Great Baddow, Chelmsford. There were three operators who worked in shifts from 8.30 am until 10.00 pm, and we would run the programs that the programmers left with us, presumably to speed up the work. However there were some hiccups, particular in warm weather. The card reader would not feed when it was hot, and had to be manually restarted to feed the data in.
Once, the engineers smelt burning inside the main unit, and it was several days before they found that the chimney of a recently installed sanitary incinerator was rather close to the air intake for the cooling fan - hence the smell of smoke!
The programmers got a bit lazy at times, and would jam the punched pieces of card (chads) back into the holes to try variations of data, the pieces would fall out and the results could be useless! Computers have come a long way since then!
Operator. My name is Doug Fawcett and as someone who was an operator
on the the Deuce I was very interested to see my name mentioned on the site.
I worked on the Deuce first as a part time operator and then became full time,
the names mentioned brought back a lot of memories and jogged it a little.
I started about 1962 and was there until it was replaced by the KDF 9, when I
then became an engineer and worked on the tape stations.
I got to know most of the people associated with the department very well, I
considered many of them as good friends and it was good to see that my fellow
operator John Cooper has added a piece to the site.
I enjoyed my time on the Deuce very much and it set me on the way to a long
career in computing, for on leaving Marconi I joined Kode Ltd, the Company that
supplied the card machines and paper tape punches for the research computers.
I also ended up marrying the Yvonne Solly, also mentioned in the article, she
was the punch room supervisor.
When I think of the size of the Deuce with all the valves and mercury delay
lines for memory, you come to realise how far we have come, the mind boggles
but I was very proud to have worked on Deuce and to have met the people
mentioned associated with it.
I have since been long retired and now live in West Wales with Yvonne but on my
regular returns to Chelmsford I do occasionaly meet the odd one or two people
that I worked with at Gt. Baddow.
Operator. I used to be an operator on the Deuce at the Marconi Co, Ltd., Great Baddow in 1963 at the age of 16 years and worked with many of the names listed on your site.(Peter Brandon, Mike Adler, Doug Fawcett, Geoff Wardell, Linda Balls etc.)
Operating this machine was a great experience at the age of 16 years. I also went on to be an Operator / Shift Leader on their English Electric KDF9 and to his day still have my Operating Manual.
One incident I recall is when the Deuce room over heated during winter one day, the room reached a temperature in excess of 100 degrees farenheit, whilst the snow was coming in through the open windows. It took Geoff Wardell some three days to get the system back up and running.
I was pleased to see your site, what a great job keeping the history going.
Operator
Chief DEUCE Maintenance Engineer
On right
DEUCE Maintenance Engineer
On left
"In early 1959 a DEUCE Mk II computer was installed in the Mathematics Department of the Marconi Company at their research facility at Great Baddow, near Chelmsford, Essex.
The computer was managed by Geoff Wardell, who was assisted by
myself as maintenance engineer. During the initial commissioning period both
Geoff and I separately went to the Kidsgrove factory for training. After
commissioning, the first two hours of each day were spent conducting
preventative maintenance before handing over to the mathematicians and
programmers. Reliability was such that some programmers would immediately
blame the hardware for program execution problems that they encountered. In
order to "defend" the hardware, I learnt how to program the computer, which
enabled me to also improve the coding of some of the provided library
subroutines, making them faster and/or smaller, thus saving precious memory. I
also modified the hardware to assist program testing:
(1) An option to store the penultimate instruction in the output staticisor.
(visible on the control panel)
(2) An option to split the cathode ray screen display into instruction oriented
groups.
The Marconi DEUCE was still operating in early 1964 when I left to migrate to Australia"
I ended up marrying Yvonne Solly, she was the Punch Room Supervisor.
(Doug Fawcett)
Head Punch Girl and IC Tabulator Room (Bernard de Neumann)
"John Cooper (almost) remembers some names of the people who worked in the off line activities associated with the Deuce." (JB)
Chief English Electric DEUCE Maintenance Engineer. "Alex Robinson looked after the Min of Ag & Fish Deuce at Guildford, he moved back to Kidsgrove as a lecturer, he reckoned he had 6 weeks to gen up on KDF9 from the engineers before giving his first lecture! He has the distinction of being the only engineeer I know who stayed with the company till retiring age." (Noel Wesson)
English Electric DEUCE Maintenance Engineer
English Electric DEUCE Maintenance Engineer - [ Website ]
In 1958 I left RAE to join the Government Section of ICT based in London as a Field Service Engineer working on the first electronic multipliers, calculators and computers that ICT produced. I transferred to MAFF (A Block) when the 555 machine was installed, as resident engineer for all 5 series machines there. After a couple of disappointing years with ICT I joined English Electric and became one of the three engineers maintaining the Deuce Mk II at MAFF (C Block). This machine operated on a two shift system, five days a week, calculating, verifying and cross checking the multitude of subsidies available to farmers in the UK as well as generating reams of statistical information for the agricultural bureaucrats in London.
In 1961 I married Jackie Rawkins, who was a machine operator in the room adjacent to the Deuce computer room at MAFF and we moved to Kidsgrove straight from our honeymoon in Torquay. From November 1961 I was working on the development and production of the KDF9 computer and at the end of 1963 we left Kidsgrove to install a KDF9 at Sydney University, train local staff and maintain the machine for the first year. We arrived in Sydney on New Years Eve 1963 on a 3 year contract but never returned to the UK."
Operator MAFF Guildford
Visit "The Staffordshire University Computing Futures Museum" "
Here
[dead link]
Which has photos of Nelson Research Labs where the early DEUCE were built and
tested.
"The Liverpool University (DEUCE Mark I with paper tape equipment) was acquired by the Stafford Technical College in 1964" (Robin Vowels)
"Came in via crane and a second floor window space, the window had
to be removed, frame and all. Stayed with us 'til 1969 (or was it '70, or '71,
or ...?)"
"The only room in the place with aircon! Warm in winter - so very popular for
winter practicals. Long DLs ran at 44.7C; if the power went off, the first job
was to wrap them in blankets, no matter what the season."
(David Leigh)
"In the late 1960's the college had outgrown its original building
and was using quite a lot of prefabricated huts as extra classrooms. These huts
had three series of numbers to identify them, namely n1, n2, n3 ..., x1, x2, x3
... and t1, t2, t3 ... This naming scheme was in imitation of the storage
locations as used in Alphacode on the DEUCE."
(Sam Valentine)
"I was poached (from NRL) in 1966 to lecture at
the college where I stayed until I retired. I also helped as deputy engineer
while I was there.
(The Deuce) came from Liverpool in 1964. Came in via crane and a second floor
window space (the window had to be removed, frame and all). Stayed with us 'til
1969 (or was it '70, or '71, or ...?).
The only room in the place with aircon! Warm in winter - so very popular for
winter practicals. Long DLs ran at 44.7C; if the power went off, the first job
was to wrap them in blankets, no matter what the season.
As I look to my left, I can see two trays of DEUCE cards, and over two dozen
manuals (DEUCE News, programme documents, subroutine documents, and "How to
build a DEUCE" documents), among others. [I kept hold of quite a few items when
we "retired" the machine.] behind me I have a (short) delay line, with some
associated electronics. Others got the front panel, the drum, and so on. Damn
it."
Lecturer. "I cut my teeth on a Ferranti Pegasus in 1961. The obvious difference between the two machines was that you couldn't walk inside a Pegasus. My association with DEUCE was more pedestrian but it kept me out of mischief until its replacement - when some small insignificant parts found their way into the control circuitry of my domestic central heating system."
Lecturer. "At the beginning of 1968 I went to work as a lecturer at
the Staffordshire College of Technology, which later became part of the North
Staffordshire Polytechnic, and is now the University of Staffordshire.
They had just taken delivery of an English Electric System 4, which was an IBM
360 lookalike. The DEUCE was still in use, however, and I did use it for the
teaching of Alphacode programming to first-year students.
It was a bit odd for me to be using a first-generation machine after four years'
experience of more modern machines, ( ICT 1301, ICT 1500, FP6000 and IBM 360)
but Alphacode was easy enough to use."
Maintenance Engineer. George Williamson is probably the 'key' person. He was the engineer as well as being a full-time (mature) Computing Student. He was regularly called out of class to fix the Deuce. (Rod Grealish)
Website - Student
" The first computer I did serious work on was an English Electric Deuce computer -- a commercial version of the famous (?) Ace computer that Turing helped to design. It had vacuum tubes, a drum (retrofitted), and mercury delay lines. User input to give signals to the program was via a rotary telephone dial."
"I was in the second year of the Computer Science B.Sc. degree when it was still North Staffs Poly, in Stafford, England. They had a Deuce that had been retired from somewhere -- an insurance company I think. It took up a whole room. I remember the delay lines being very temperature sensitive, and part of the operation being adjusting how open the windows were to control the room temp. We learned Deuce autocode (Alphacode - JB) to start with, moving on to other languages later on other computers. The joys of 5 hole paper tape and "chinese binary" (most sig bit on right)!"
Student, Graduated 1970 and joined the college staff. "My memory of the Deuce was that it occupied (but did not fill ) a room. It was the size of a Horse-box. The Internal circuits could be accessed by a 'corridor' running inside the box. The 'corridor' was about 2 feet wide. I remember it had a parquet-style floor. There was a control desk attached to the horse-box. Programs were entered using papertape prepared in another room using teletypes."
I stumbled across your web page by (sort of) accident.
I was a (Liverpool Based) English Electric Student Apprentice from 1962 to 1967. As such I went to North Staffs Poly (Semister B) starting February 1963; to study Electrical Engineering.
I remember the Deuce - and the Alpha Code and the 2000 or so cards of the Algol compiler.
Two things come to mind.
1. I wrote a machine code program to generate random numbers for the college newspaper - the "Picture House" cinama (as I recall) gave us two complementary tickets each issue, and we needed an unbiased way of finding the winner. It was done by looping around with the arithmetic unit, then selecting the random number via a 'key' on the console, with all attempts before a certain time being dismissed (so that early selection was not possible) and then the number was truncated to the range of the 'sales'. All very fair (I think !).
2) The operator (can't remember her name) asked me one day if I could get the Algol Compiler copied - No Problem! I wrote a machine code program to do, in a loop, start the Card Reader, Read a card into a delay line, stop the card reader, start the card punch, write the card from the delay line, stop the card punch, repeat until no more cards. Simple!. Test it, copies a few cards perfectly.
Next I get a message to go and see Head of Department urgently. "Are you responsible for copying cards on the computer?". "Well probably - I gave (forgotten) my card copy program!". "Don't ever do such a stupid thing again - You don't read just one card, you fill up the delay lines, then stop the card reader - you have just burned out the clutch on the card reader - you *** *** etc. etc.".
So if anyone remembers the card reader in bits - I have to admit, it was my fault.
As it turned out I did very little Electrical Engineering (I left English Electric Liverpool in 1970) - I have spent most of my working life since then designing and/or writing computer software - all because I caught the bug from a Deuce!.
One little aside - When I first went to Stafford in February 1963, the (sponsored student) payroll was still done using punched cards (I am not sure which E.E. computer was used by Stafford Works at the time). They only allowed 10 characters for the surname - one of my old college friends to this day persists in calling me "HETHERINGT" - with an emphasis on the "T" - Ah! those were the days!
'Ah, Deuce!' Now your talking. When I went to college in Stafford (1964-1969) they had a second hand Deuce there. I learned all my initial programming on that machine. First in a language called Alphacode, and then in the machines own low level code. I wrote a simple machine code compiler for it, and also an algol like high level language that I called Simplecode. I can still remember a large number of the command codes and programming rules for this machine.
Read Richard's more detailed memories of those early days The first meeting with DEUCE
"I sometimes helped George Williamson repair Deuce when it was
sick and on one occasion he was on a half day's leave and the machine went
down in the middle of the afternoon. When I went to use the computer
everybody was standing round looking lost. I said I would try and fix it
if I could run my programs before the rest of the queue. About 10 minutes
later the machine was working and I had my results - it was just a valve in
one of the delay line refresh circuits that had gone down. Considering its
age, the diagnostics on the system were excellent."
(Extract from a letter to Jeremy Walker in the mid-80s )
Just come across your web site on Deuce computer. I actually programmed the last working version of this computer at Stafford Polytech in 1971. We used Alphacode and sent and received the programs via teletype paper tape.
I remember the heat from the computer very well, as well as its rest period when the sun shone !. I later went on to become a computer operator on IBM 360/30, 370´s and later in to telecoms at TSB. I have spent 35 odd years in IT and its nice to see that somebody remembers these first generation computers.
People find it hard to believe that I actually worked these beasties. Such fond memories, thanks.
Like others, I stumbled upon your DEUCE pages by accident, when my eldest son pointed me at some information about the LEO.
Great work, thanks for all that you and others have done!
I studied at Staffordshire College of Technology in 1962, as part of a "thick sandwich" apprenticeship with English Electric in Stafford.
Programming the DEUCE was part of the course. The most exciting part, I recall!
We programmed "Pythagoras' equation" using about 22 punched cards, and I clearly remember the delay line memory but not the drum - perhaps that came later?
I first met DEUCE in 1955, worked with many of the people on your
list, and switched off the DEUCE at SCOT in 1968 possibly 1969.
Will put together a more detailed chat at some point
"I was told about your web site by Clinton Bourne (he worked on DEUCE at Kidsgrove from 1962 until 1965 or thereabouts)." (Peter Johnson)
renamed, on 7th October 1958.
U niversity of T echnology E lectronic COM puter : UTECOM
"The third academic computer was commissioned at the then recently established New South Wales University of Technology by the Premier of New South Wales, the Hon. J. J. Cahill, while opening a Symposium on 'Automation and Australia' on 11 September 1956. At this laboratory the main installation was one of the early vacuum tube machines produced commercially by the British-based English Electric Company, DEUCE, called UTECOM for University of Technology Computer.
The DEUCE was a manufactured version of the first realisation of the system originally conceived by Alan Turing in 1946 (1945 - ed) at the British National Physical Laboratory, Teddington, United Kingdom. It used delay sonic storage, similar to the EDSAC and the Mark 1, but at the higher pulse rate of 1 Megahertz. It had a precision and instruction length of forty bits (32 bits - ed), the latter having a multi-segmented format which made programming rather difficult. But, carefully used, it provided much faster speeds than the Mark 1, although the total delay of each sonic channel was much the same ." - ( Australian Computing, the Second Generation )[dead link]
The UTECOM m/c was the first one I worked on - during it's commissioning at Kidsgrove. Tom Elliott and Eric Thomas took the drum by air - in it's own seat - to Australia. (Jeremy Walker)
Thanks to Robin Vowels for following extracts from Utecom's first annual report and most of the information about Utecom staff.
"Ron Smart and Keith Ford assisted installing UTECOM, with T Elliott and E Thomas from EE."
"The initial staff consisted of:
Lecturer in Electrical Engineering (Machine operation and supervision)
"After being awarded the University Medal, Ron Smart was snapped up to become the inaugural director of the UTECOM Laboratory, and sent to England to learn about the DEUCE that was being built for the University. At that time, there was only one computer in Australia, Trevor Pearcey's CSIRAC built in 1949.
Ron would have given programming courses etc, drummed up business from the outside world, etc. The laboratory did outside work (both programming and selling computer time)." (Robin Vowels)
Lecturer in Electrical Engineering (Programming)
"George Karoly did some programming work too, including for the Soda project." (Robin Vowels)
Technical Officer (Programing and Secretarial duties)
Technical Assistant (Card Preparation)
Laboratory Assistant (Machine Maintenance)
Laboratory Assistant (Machine Maintenance)
Technical Officer - replaced Miss M Oates
Technical Assistant - replaced Miss G Gaskin" - (UTECOM 1st Annual Report)
Team Leader - ( Trevor Pearcey )[dead link]
"Barry Thornton was a mathematician at the university at the time. He was not involved with the computer, but he was very much involved with making sure Gordon and I had a wonderful stay in Australia." (Bob Brigham)
Senior Programmer - ( Trevor Pearcey )[dead link]
"Bob Brigham, my roommate, and I went to Australia as Fulbright
scholars, taught a graduate course, and built a pretty impressive compiler for
their computer. (See Bob Brigham entry) It was the English Electric Deuce, a
follow on to the NPL (National Physical Laboratory) Ace that Turing designed.
It was a very hard machine to program because its main memory was delay lines
with 192, 32-bit words and programs resided on an 8 K word drum. It had card
input, and you signed up to use the computer for short periods of time – it was
used as a personal computer, albeit one you could walk into. We wrote a compiler
to optimize programs and make it easier to use. It’s 32 word, 32-bit memories
could be displayed on a CRT, so you could interact with it."
Extract from 'An Oral History Interview with Gordon Bell - April 1995' - Full
text available
here
[dead link].
"In 1958, when I met Wilkinson at NPL (that Deuce came from) to give a talk, he said 'who needs a symbolic, optimum assembly program' -- (in essence) real men program in binary. "
"I wrote a program that proposed to my wife, Gwen, using the Deuce switches and the displayed delay lines where you scrolled messages on the 32 x 32 dot grid."
"I remember going to work each day wondering whether we'd get much time on the computer. It seemed to go down all the time. Ron Smart was a miracle worker who always could get it running. He was a workaholic, a tremendously nice guy, and a real joy to be around. So was Gordon fun to be with. As he mentioned, we had known each other several years at MIT and roomed together during our graduate year there."
"I remember the punched card input where we punched in the instructions in binary. If we made a mistake and we wanted a 1 where we originally had a 0, it was no problem - we simply punched another hole. However, if we had a 1 and it should have been a 0, we went to the trash can, picked out a punched out piece of card, inserted it in the hole, rubbed it with our finger to make it smooth, and reproduced the card. Needless to say, the entire process was somewhat time consuming."
"We went to the Snowy Mountains on a trip and Gordon and I took along some sheets and did some of the programming for our compiler then, between freezing and enjoying some of the most beautiful scenery I have ever beheld."
The Computer Journal , Volume 2, Issue 2, 1959 - A translation
routine for the DEUCE computer
RC Brigham and CG Bell
New South Wales University of Technology, Sydney, Australia
Most computer in operation today have supplementary programs which do automatic
coding or program assembling. These programs either translate, automatically
code, or interpret pseudo instructions which in themselves may cause the
enaction of hundreds of actual machine instructions. The outstanding feature of
such routines is that programming time and effort is cut to a minimum. This
paper deals generally with translation and interpretive schemes, and
specifically with a suitable translation routing for use with the DEUCE
computer. The translation program is called SODA, or Symbolic Optimum Deuce
Assembly Program. - Full text available
here
"After graduating from London and Princeton Universities, Elliott's
interest in computing began in 1955 while working in the Mathematical Division
of the National Physical Laboratory in England, where he gained experience on
the English Electric computer DEUCE. He came to Australia in 1957, working for
twelve months on the same machine, now termed UTECOM, at the NSW University of
Technology."
Extract from 'Leading Edge' A.C.S. SA Branch News Nov/Dec 2000
[ Website][dead link]
Lecturer and Professor in the School of Philosophy" Charles Hamblin
conceived the address-less language based on reverse polish notation and
implemented a compiler for it called GEORGE
(GE neral OR der GE nerator) by May 1957. This compiler
became important in the teaching of programming to students and staff, and was
used in preference to the English Electric Alphacode scheme."
Geoff Roper of the Chemistry Department used it extensively, running it all weekend."
GEORGE included subroutines, something that FORTRAN did not have until June 1958. Charles wrote a program requiring 160-bit integer arithmetic to produce Babylonian tables. On DEUCE, five words were required. This would have been straightforward on account of the 32-bit and 32/64-bit adders, which gave convenient access to the carry bit. (Robin Vowels)
"from Melbourne used the machine for weather forecasting, as the machine was more reliable than CSIRAC."
John Webster was employed initially as trainee programmer from about the beginning of 1961. However, he also became interested in and was involved in the maintanance (mainly logic) of UTECOM. As a programmer, he was par excellence, and his command of DEUCE logic and circuit diagrams was outstanding.
John wrote several useful programs. One was NSW 71, a GIP brick that read six 9-digit numbers or eight 8-digit numbers, automatically choosing one or the other format depending on the first decimally-punched card. This was intended to replace an EE GIP brick -- namely LR21BT that read six 9-digit numbers -- that was described by DEUCE Librarian R. A. Smith as "ham fisted" because it randomly stopped the card reader while reading the data. The author of that program had thought that magnetic interlocks did not exist, for he used waste instructions to delay up to 15ms following a track write.
The other problem with it was that there were two blank card columns between the third and fourth numbers -- a gap that was incompatible with general decimal output routines. John's brick did not stop the reader between cards. After the cards had been read, track reading and writing were overlapped during the scaling process so as to eliminate delays owing to magnetics operations. John's brick occupied 15 tracks. LR21BT occupied 22 tracks. It seems impossible that John's brick could combine the tasks of two bricks, and for the result to be smaller than one of them! In short, a very professionally-written brick.
However, John's major achievement was his brick-changing program, WIP. This was intended to replace DEUCE program ZC14T/1. The latter required detailed knowledge of DEUCE to set up the parameters to use it. John's WIP did not. It used a subset of GIP codewords. So anyone who knew GIP could use WIP. Where WIP excelled were in its speed (faster than ZC14T/1 and GIP) and its small size. WIP occupied 5 tracks, whereas GIP occupied 23 tracks. John writes: "[WIP] seemed to satisfy quite a few people's needs. Don Craig was one guy who was happy to have the freed-up tracks for his work - he heard from someone about WIP's smaller drum occupancy and came to "beg" to be able to use it (to my surprise, and great self-esteem!) because he needed that extra space."
Both of John's programs (NSW 71 and WIP) were submitted to English Electric for publication, but they suffered the same fate as Hamblin's GEORGE, which also was never published. (We don't know whether SODA was ever submitted.)
John also wrote a double-precision version of Hamblin's GEORGE, plus an enhancement to print on the teleprinter (in algebraic source form) the user's complete GEORGE program, along with a message at the very place where the program had halted.
John's other contributions included a test program, TNSW02, to
rigorously test early and late sources. We included it in the general engineer's
(Robin Vowels)
Website
[dead link] See also
The
DEUCE -- a user's view Robin A. Vowels
Technical Officer UTECOM Machine Maintenance
I was initially employed as programmer from about March 1961, and was responsible for the maintenance from December 1962. My main task was to rehabilitate UTECOM and to bring its reliability up to that of DEUCE machines overseas.
The DEUCE engineers' course was 6 months, and in 1962, Arthur Bailey was the instructor. The course included logic and circuits for all of DEUCE, not just the basic machine. Peripherals covered were 64-column I/O, paper tape I/O, 80-column I/O, and magnetic tape.
As our site did not have magnetic tape, English Electric decided that some on-site experience would be more appropriate than sitting though the magnetic tape component. They therefore sent me to Warton where two Mark I DEUCE were operating side-by-side.
In the 3 weeks that I was there, not a single breakdown occurred while I was present; there was, however, one overnight. EE also gave me hands-on experience on the magnetic drum, among other things, setting the head gap with the aid of a hi-tech device, to wit, a cigarette paper.
Being the last 64-column DEUCE to get it, I installed Rationalized Magnetics on UTECOM, by building the unit from scratch, after stripping down the old unit. This was done in July 1963, and took about three days' wiring work.
I also installed ME/M modifications for the pseudo parity check on the drum. This came as a piggy-back unit from English Electric, and it was shipped out to us with a refurbished magnetic drum unit, received in about March 1963.
In about August 1963 I installed a Siemens M100 teleprinter to take some load off the ailing card punch. Initially this was a primitive affair requiring programmed I/O, as the simple interface (reading or writing a single bit at a time) was non-standard. The first teleprinter unit at 50 Baud was on loan from the PMG until we received the 75 Baud machine (10 cps) that was on order from Siemens in Melbourne (at that time, they were assembled in Australia).
I had hoped to add cams to the unit, so as to duplicate the requirements of the standard DEUCE paper tape punch, but this proved impossible owing to lack of space. I therefore designed an electronic parallel-to-serial interface, and built it up on a spare blank DEUCE chassis. This permitted standard programming, as it generated the TIL signal when ready to punch; a Destination 29 instruction caused the output of one character, also standard. As the interface required no modifications to the teleprinter, it looked like it might have applications beyond our site, so I sent a copy of the circuit to Jack Richardson, who replied that "It looks delightfully simple. I'll see if there is any use for it here."
One day I noticed an external user open a cabinet door of the machine and fiddle with a unit. Curious, I asked her what she was doing. She said that she was turning A.I.M. off. And sure enough, there it was: a two-pin jumper plug, on Unit AIM, and the spot was engraved AIM OFF. This was in 1963, some 4 years after A.I.M. was installed. It was a non-standard modification, and did not appear on the circuit diagram.
Read Robin's memories of bringing UTECOM reliability up from 60% to around 95%. UTECOM - Major Machine Malaise
I also completed STAC (a symbolic assembler) in my spare time.
We didn't have to resort to Royle's reported method of "switching off" UTECOM. It switched itself off periodically when one of the many feed-through capacitors short-circuited. The lads found that it was quicker to locate the offender by removing the covers in the internal walkway, switching off all the lights, and "powering up". The culprit revealed itself with a brilliant flash. The power stayed on only momentarily, and it was usually necessary to power up more than once before the source of the flash could be identified. Needless to say that the internal covers were permanently removed, as this event occurred relatively often. By 1963, I ordered a few hundred of the different brand of capacitors (other DEUCEs in the UK were already fitted with them) and during a weekend working bee, we replaced the remaining ones.
When we were having problems with the multiplier, we ran the DEUCE test program that uses the hardware multiplier and which compared it with the product obtained with software.
When a fault was detected, the content of the multiplier were punched out during each of the 64 minor cycles of the multiplication, followed by a copy of the correct version as simulated by software.
Unfortunately, all the output was punched in 32 column, which wasn't convenient as all the values were 64 bits.
So I modified the progam to punch the results in 64 columns. Then it was easy to see the results as the bits were shifted up one-shift-at-a-time (recall how easy it was to compare results? hold pairs of cards up to the light to see which bits were different).
I sent this to EE, expecting that it would be a welcome addition to the library, but like many other programs from this part of the world, it was ignored. NIH (Not Invented Here) syndrome.
Technical Officer UTECOM Machine Maintenance
Technical Officers UTECOM Machine Maintenance
Bob Reid and Rodney Bell installed the 64-column modifications for the card reader and card punch in 1959. This involved extensive cabling in the main frame (cables with 64 multi-colored wires for each device), and the flexible cables connecting the machine to the card reader and punch. Modifications were also required to Units C, D, and E for the reader.
These modifications were probably done in the days leading up to the re-wiring, as the alterations would not have affected input. A new roller had to be fitted to the card reader. This was a "split roller", with an insulating spacer between card columns 48 and 49, as the two parts of the roller had to be kept at different voltages.
Standard units MV/1 and MV/2 were probably supplied for the conversion of the output. The conversion was carried out over 8 consecutive days in August-September 1959 with the machine switched off, and required 89 hours' labor.
Bob related an incident during the cabling. At one end of the cable in one part of the machine, Bob would call out the color of the wire that he was connecting, and Rod at the other end of the cable would connect the corresponding wire at his end. Nearing the end of the job, Bob called out a color, but Rod did not have a matching one. It turned out that one of them was color blind !
Bob and Rod would also have installed the Automatic Instruction Modification unit (A.I.M.) at around the same time. This was a standard unit supplied by English Electric. (Robin Vowels)
Technical Officer UTECOM Machine Maintenance
Technical Officer UTECOM Machine Maintenance
Technical Officer UTECOM Machine Maintenance
Technical Officer UTECOM Machine Maintenance
"After Ron Smart left at the beginning of 1961, the late Les Hill was (acting?) director until 1966." (Robin Vowels)
Brian was UTECOM's scientific programmer, wrote any programs for (paying) outside users requiring mathematical expertise.
Brian amended the Crystallographic programs originally written by J. S. Rollett for DEUCE. These programs gave wrong results whenever fixed-point arithmetic overflowed, as Rollett did not include checks for overflow. He considered them unnecessary.
Users could not tell whether results were correct or not when this happened, and if not correct, they could not tell whether or not there was something wrong with their experiment. Brian also incorporated additional checks on arithmetic, for until that time, UTECOM was not reliable.
One PhD student wrote in his thesis that he had to run the crystallographic programs up to five times before getting two sets of results that agreed. Rollet's original programs punched out results one card at a time -- this was the principal cause of repeated mechanical failure of the card punch. Brian amended the output section to punch batches of 16 cards.
Regarding his Crystallography programs, Rollet wrote that he was
able to keep the multiplier running three-quarters of the time (about 14 Mcs
in 17½ Mcs) with concurrent programming, as the multiplier was asynchronous.[1]
(Robin Vowels)
[1] J. S. Rollett, "General Programs for Crystal Structure Analysis on the English Electric DEUCE Computer", pp. 87-101.
John worked for a short time as programmer in about 1964. He picked
up machine code with ease.
In one program where time was not critical, he dispensed with optimum coding,
and coded instructions sequentially in consecutive minor cycles in delay
lines.(Robin Vowels)
Jean wrote programs occasionally, and did card punching for programmers.(Robin Vowels)
Gay ran the Anderson Analysis, a long-running machine-language
program for analysis of TV viewing habits.
This program was run after hours one night each week. At other times, she was
secretary.(Robin Vowels)
( Website ) [dead link]
"While I did not know Alan Turing, I may be one of the select few here who
used a computer designed by him. In the late 50's, I had the pleasure of using
a mercury delay-line computer (English Electric Deuce Mark II) based on
Turing's original design of ACE."
( To Dream The Possible Dream
) [dead link]
"When the University of Glasgow switched on its computer (an English Electric Deuce) in 1958 it was the first at any university in Scotland and only the second at any site north of the Border."
"Only seven years before, Professor Douglas R Hartree of Cambridge University had stated, 'We have a computer here at Cambridge, there is one in Manchester and another at the National Physical Laboratory. I suppose we should have one in Scotland, but that is all.' "
DEUCE: Glasgow's first computer: Down memory lane (Now only accessible on the Wayback Machine: here)
"The Computing Laboratory set up in the University by Dr, later Professor, Dennis Gilles was home to the first university computer in Scotland."
Maths Post-graduate Student
"Ann was there when we delivered a machine to Glasgow University. She was appointed by Dr Gilles to adjudicate on the Customer Acceptance Test. The Acceptance Test at that time required three days of totally fault-free running - remember, there was no automatic error-recovery and precious little diagnostic ability. We learned in the end that the only way of getting these machines through such Tests was to subject every valve to a test for Microphony - very heavy vibration, i.e. hit it - and many's the time that one has banged too hard and shattered the glass. The blow also shorted out the internal electrodes and that naturally took the entire machine off with a bang and a crash. However, it was the only way you could get it to run for about four days. The valves started to become microphonic again in due course but that is another story." (Jeremy Walker "DEUCE - I'ts Life and Times")
DEUCE Maintenance Engineer
After the middle month of the Deuce course Derek Royle sent me to Glasgow to look after the Deuce at the University, newly commissioned by Jeremy Walker and Brian Bispham. That was where I met Ann who acted as my interpreter, the Clydebank accent being unintelligible to a Londoner. (Extract from "My Deuce Time", discover how Noel was lured from LEO by a bevy of 545's !)
Noel then joined the Deuce Mobile Service Unit - South at Starcross Street in London.
Doug worked under Dennis Gilles at Glasgow, starting in October 1958. At that point he was probably the most knowledgeable DEUCE staff member, having come straight from the dual installation at RAE. (Kathleen Williams)
"At Glasgow University, he studied Numerical Analysis in the early 1960s, using the DEUCE and KDF9 machines."
( www.dcs.gla.ac.uk/people/personal/jwp/)
When John Patterson came to Glasgow in 1961 he wanted to be a physicist. "It was only about 1962 I discovered the University had a computer", he says. 'They didn't exactly admit students to it because it was the University's only computer - but I got to use it anyway! "
"I never programmed the London machine but I did get onto the Deuce at Glasgow almost as soon as I discovered its existence. My first programming language was Deuce Alphacode, and I was exposed to TIP and GIP without knowing why I would need them."
"Glasgow still has bits of its old Deuce in the University's museum."
DEUCE Maintenance Engineer
"The University's site engineer was Tony White, a Brummie who lived for the mountains and had taken the job so he could spend every weekend in the mountains. Tony is now retired to the Isle of Arran, off Scotland." (Noel Wesson)
DEUCE Maintenance Engineer
DEUCE Maintenance Engineer
I owe many happy days and memories to DEUCE and the fact that The Central Bureau of Statistics in Oslo 1958 had decided to buy this computer in cooperation with 'Norsk Regnesentral' (Norwegian centre of Computing, part of the Central Institution of Industrial Research, known as 'SI').
Find out how Thin acquired the nicknames "The Whaler" and "The Cradle Snatcher" when he came to Kidsgrove to learn all about DEUCE. Deuce - Happy Days - Happy Memories
This Deuce Training Course photo shows Thin, forth from left.
"The name Thin Nenseth rings bells. I'm fairly sure I met him in Oslo, but I think he must have moved on to other things by my arrival." (Jerry Fraser)
"Thin Nenseth, a Norwegian, was learning English as well as Deuce and looked after the Oslo DEUCE." in 1958. (Noel Wesson)
DEUCE Maintenance Engineer - (While I was at Kidsgrove my surname was Smith. I changed it to Fraser - Old Scottish roots - in 64/65
I just came across your DEUCE web site and was very happy to see
that information about 'my first computer' still exists!
I joined English Electric back in the early 60's, and after a year or so in
Kidsgrove, was posted to Oslo to look after the DEUCE MKII at the Central
Bureau of Statistics for the last three years of its life... after that I
joined Digital Equipment. Quite a change from DEUCE to the PDP-8!
Must say, looking back on life, I'm very happy Derek Royle offered
me the job at Kidsgrove. I understand he was in two minds, because I turned up
for the interview, recently returned from the Persian Gulf, deeply tanned, in
what must, for the Midlands, have been a 'flashy' tropical suit.
However, I did know my flip-flops, and Arthur was pretty happy with my results.
"I remember Gerry Smith with whom I used to ski when I visited Regnecentralen in Oslo to do the Deuce Annual Maintenance." (Jeremy Walker)
DEUCE Engineer
|
Bob Mc Call at the Console with me (Alan Gray) standing by
the m/c intelligently inspecting a waveform |
>"Bob McCall installed that machine and spent a year or two there." (Jeremy Walker)
DEUCE Maintenance Engineer
DEUCE Maintenance Engineer
Kåre Steira did the DEUCE training course with me at Kidsgrove in
1962. (Robin Vowels)
I knew Kåre well. He was the only engineer on site when I shipped in to Oslo,
but he moved on to the bureau's IBM machine after 6-months or so.
(Gerry Fraser)
Demand for computer time grew and so in preparation for further machines a new computer building was completed in 1956. The ground floor was occupied with punched card accounting machines that were used for Monte Carlo neutron calculations whilst the top floor comprised offices.
An English Electric Deuce was installed in a separate room adjacent to the Ferranti Mark 1*. Deuce was also a valve machine, its fast access store consisting of mercury delay lines (12 stores of 32 32-bit words) and drum (8K words) as backing store but with punched card as the medium for input/output.
Both Deuce and the Mark 1* were run for 24 hours a day, 7 days a week by scientific staff: there were no operators! Programming was still carried out in machine code.
(Jim Taylor - History of Scientific Computing at AWE - Part 1 )[dead link]
"I commissioned the Capenhurst Atomic Energy one with Frank Thompson
and the acceptance tests were dragging on and on because one hole in one card in
a huge pack was sometimes missing, always the same hole in same card.
Jeremy breezes in one day in his "Northern Service Manager" capacity and having
listened to our tale of woe asked "Is the Hollerith earth separated from the
main earth". "Of course it is" snarls Frank, where upon Jeremy stands on a box
and snips a wire in the roof, whereupon all tests ran OK and Frank went home,
leaving me to look after it until their engineer finished his course."
"I worked on a Deuce in the summer of 1966 as a student: the site was the UKAEA uranium hexafluoride processing plant at Capenhurst, near Chester, UK. I remember little about it, except programming in Alphacode, and the mercury delay lines all over the place!!" (I recall the) "Monday morning routine of one of the technicians coming round with a large cardboard box (with Persil Soap written on the side) full of thermionic diodes (sic), and him walking into the machine to replace duff ones (or is my memory/imagination playing tricks here??)"
"Developed the paper tape version of STAC and rewrote the analysis phase of STAC. Their DEUCE Mark I had paper tape input and output peripherals, and also ran time-sharing programs" - ( Robin Vowels )
"Another machine being assembled and tested whilst I was there (at Kidsgrove), was the Liverpool University machine, in the hands of their first two maintenance Engineers, Danny GREENHALG and Brian MOONEY" (John Ryan )
"I was interviewed by Derek Royle at Liverpool University who were advertising for a site engineer. Derek poached off the applicants he fancied for his DEUCE support section at Kidsgrove." ( Neil Charlesworth )
"Later on at Whetsone I joined Brian Randell (later Prof. of Computing at Newcastle Uni) and Lawford Russell in writing the first ALGOL compiler on DEUCE, in conjunction with the Oceanographic Dept. at Liverpool University." ( Maurice Batey )
Worked on the DEUCE in the early sixties, before he emigrated to Australia
Queens University Belfast - DEUCE Maintenance Engineer
He worked on the National Engineering Laboratory DEUCE in East Kilbride. (Peter Dukes)
DEUCE Maintenance Engineer
One name I remember on my course was Dennis Bloor. (John Kelleher)
"The English Electric DEUCE was programmed in binary (one 32-bit word to each row of a punched card). Each instruction had to specify the location of the next instruction to be obeyed, and the way to get a fast program was to place instructions in the mercury delay lines such that there was no unnecessary waiting between instructions. An interesting technical challenge was to write a bootstrap program of twelve instructions on a single punched card. During my time as a pre-university student at the English Electric Company, a staff member was in the process of writing an assembler which would, inter alia, look after instruction placement."
"My first programming language was DEUCE Alphacode. The language provided a set of floating point variables (X1, X2, ..., X2200) and a smaller number of counting (integer) variables (N1, N2, ..., N63). One line of code could perform a single operation, for example "X1 = X2 + X3" or "X4 = ROOT X5". The statements of Alphacode were usually interpreted, not compiled. Writing and using an Alphacode program was an improvement on performing pre-specified calculations on an electro-mechanical calculating machine, my previous activity at English Electric." ( Programming Languages over the Years - Julian Blake)
To return to the DEUCE homepage click the image below.
ADIWALI - Akim - BAC Warton -
ADLER - Mike - Marconi Ltd -
ALBOROUGH - Rosemary - RAE Farnborough -
ALLCOCK - Barbara - BAC Warton -
ALLCOCK - Steve - BAC Warton
ALLMARK - Reg - EE NRL Stafford -
ALLWOOD - Roger - EE Kidsgrove -
AMERA - SINGHE - U J - EE Kidsgrove -
ASBURY - Alan - EE Kidsgrove -
ASHBROOK - R - EE Kidsgrove -
ATKINSON - Cyril - EE Whetstone -
BACK - David - BAC Filton -
BAILEY - Arthur - EE Kidsgrove -
BAKER - Joan - RAE Farnborough -
BAKER - Richard - RAE Farnborough -
BALLS - Linda - Marconi Ltd -
BAMBER - Marie - BAC Warton -
Barbara - ??? - Marconi Ltd -
BARKER - Colin - BAC Warton -
BARRETT - John - About the Author -
BARRETT - John - MAFF Guildford -
BARRETT - John - RAE Farnborough -
BARRITT - Marjorie M - RAE Farnborough -
BATEY - Maurice - EE Whetstone -
BECK - David - BAC Warton -
BECKETT - Bill - EE Kidsgrove -
BEEDON - Al - BAC Warton -
BELL - Gordon - UTECOM - Sydney -
BELL - John - BAC Filton -
BELL - Rodney - UTECOM - Sydney -
BENSTEAD - Peter - Central Electricity Generating Board -
BENTALL - Sid - Marconi Ltd -
BERRIDGE - Pete - EE Whetstone -
BERRY - Hilda - BAC Warton -
BETTS - Arthur - RAE Farnborough -
BIGGINS - S - EE Kidsgrove -
BIRCHALL - Paul - RAE Farnborough -
BIRCHMORE - Audrey - EE Marconi House London -
BIRD - Roger - EE Stevenage -
BISHOP - John - Mobile Unit South -
BISPHAM - Brian - EE Kidsgrove -
BLACKWELL - Dennis - EE NRL Stafford -
BLACKWELL - Dorothy - BAC Warton -
BLAKE - Julian - -
BLANCH - Lawrence - -
BLOOR - Dennis - EE Kidsgrove -
BLOOR - K - EE Kidsgrove -
BOND - Diana "Dinks" - BAC Filton -
BOND - Dick - EE Luton -
BOOKER - David - BAC Warton -
BOOTHROYD - John - EE NRL Stafford -
BOSTON - Dennis - Bristol Siddeley Engines -
BOULTON - Nancy - BAC Warton -
BOURNE - Clinton - North Staffs Poly -
BOXALL - Di - RAE Farnborough -
BOYCE - Alan - Marconi Ltd -
BRADLEY - Ron - BAC Warton -
BRADLEY - Sheila - BAC Warton -
BRADLEY - Sheila - BAC Warton -
BRADSHAW - Peter - NPL -
BRANDON - Peter - Marconi Ltd -
BRAY - Lilian - EE Marconi House London -
BREWARD - Wendy - Bristol Siddeley Engines -
BRIGGS - Ron - Mobile Unit South -
BRIGHAM - Bob - UTECOM - Sydney -
BRIGHTLING - Philip - BAC Warton -
BROCKINGTON - Denis - Marconi Ltd -
BROWN - David - North Staffs Poly -
BROWN - J. K. - EE NRL Stafford -
BROWN - Ken - Marconi Ltd -
BROYDEN - Charles - EE Whetstone -
BRUFFELL - Peter - MAFF Guildford -
BRUNT - Alex - EE Kidsgrove -
BUNNEY - Peter - RAE Farnborough -
BURDEN - Dickie - Bristol Siddeley Engines -
BURNETT-HALL - David - RAE Farnborough -
BURROWS - Richard - EE Kidsgrove -
BUTLER - Tom - EE NRL Stafford -
BYTHEWAY - Andy - North Staffs Poly -
CAINE - Michael - BAC Warton -
CALVERT - Len - EE Kidsgrove -
CAMPBELL - John - -
CAREY - Margot - BAC Warton -
Carol - ??? - Marconi Ltd -
CARRE - Bernard - EE NRL Stafford -
CHAMBERS - June - BAC Warton -
CHAPMAN - Melvyn - North Staffs Poly -
CHAPMAN - Rosemarie - RAE Farnborough -
CHARLESWORTH - Neil - EE Kidsgrove -
CHRISTOPHER - Elizabeth - EE Kidsgrove -
CLAPHAM - Margot - BAC Warton -
CLARKE - Mabel - Marconi Ltd -
COATES - Philip - BAC Warton -
COCKLE - Ted - Marconi Ltd -
CODY - Paul - RAE Farnborough -
COLCLOUGH - Fay - EE NRL Stafford -
COLDHAM - Vic - EE Kidsgrove -
COLE - Jenny - EE NRL Stafford -
COLES - Bill - BAC Warton -
COLLIER - Roger - BAC Filton -
COLLING - Brian - EE Kidsgrove -
COLLINS - Bob - Mobile Unit South -
COOKES - Tony - BAC Filton -
COOPER - John - Marconi Ltd -
COPSON - David - BAC Warton -
CORK - George - RAE Farnborough -
COULSHED - Bill - BAC Warton -
CRUICKSHANK - Jim - BAC Warton -
DACE - Jean - EE Marconi House London -
DAIN - Julia - Marconi Ltd -
DAVIES - Fred - BAC Warton -
DAVIS - George - EE NRL Stafford -
DAVIS - Mike - BAC Filton -
DAY - Ken - EE Kidsgrove -
de BOURCIER - Enid - RAE Farnborough -
de NEUMANN - Bernard - Marconi Ltd -
DENDLE - Maurice - EE Whetstone -
DENISON - John - EE NRL Stafford -
DENN - Muriel - BAC Warton -
DENNETT - Pauline - BAC Warton -
DICK - Ian - University of Glasgow -
DIPROSE - Kenneth - RAE Farnborough -
DOBSON - Eric - EE Kidsgrove -
DOCHERTY - Peter - EE Marconi House London -
DODD - Ken - RAE Farnborough -
DONOHOE - Neil - EE NRL Stafford -
DOWELL - Norman - EE Kidsgrove -
DREW - Mike - EE NRL Stafford -
DRUMMOND - Barry - BAC Warton -
DUDLEY - Rob - EE Whetstone -
DUERDEN - Tom - BAC Warton -
DUKES - Peter - BAC Warton -
DUNCAN - Fraser - EE Kidsgrove -
DUNLOP - Ed - Marconi Ltd -
EASTWOOD - Eric - Marconi Ltd -
EDWARDS - Jim - EE Kidsgrove -
EDWARDS - Lyn - BAC Filton / OSLO -
EITEL - Ron - EE Marconi House London -
ELKIN - Stan - EE Kidsgrove -
ELLIOT - Tom - EE Kidsgrove -
ELLIOTT - David - UTECOM - Sydney -
ELLIOTT - Susan - EE Kidsgrove -
ELLIS - Eileen - BAC Filton -
ELLIS - Gay - UTECOM - Sydney -
ELLIS - Pat - Bristol Siddeley Engines -
ELLISON - John - BAC Filton -
ELLISON - Ray - EE NRL Stafford -
EVANS - Phyllis - EE Luton -
FAIRTHORNE - Robert A - RAE Farnborough -
FARMAN - Joan - EE Luton / Stevenage -
FARTHING - Rosalind - Marconi Ltd -
FASEY - Mike - BAC Filton -
FAWCETT - Doug - Marconi Ltd -
FEY - Joyce - Bristol Siddeley Engines -
FISHER - Jim - EE Luton / Stevenage -
FINEBERG - Harold - EE Luton -
FLEMING - Mike - BAC Warton -
FLETCHER - Jim - Bristol Siddeley Engines -
FLOWER - Doug - EE Marconi House London -
FORD - David - EE Marconi House London -
FORD - Fred - EE Whetstone -
FORD - Jennifer - EE Marconi House London -
FORD - Keith - UTECOM - Sydney -
FORRESTER - Richard - EE Marconi House London -
FOSTER - Margaret - UTECOM - Sydney -
FOWLER - David - RAE Farnborough -
FRANCOMBE - Heather - Bristol Siddeley Engines -
Frank - ??? - EE NRL Stafford -
FRANKS - Peter - BAC Filton -
FRASER - Gerry - Central Bureau of Statistics - Oslo -
FUDGE - Jane - -
GAHERTY - Mike - EE Kidsgrove -
GALLOWAY - Bob - BAC Warton -
GALPIN - Daphne - Marconi Ltd -
GALVIN - Joseph - EE NRL Stafford -
GASKIN - G - UTECOM - Sydney -
GEARING - Gus - RAE Farnborough -
GENT - Ron - EE Whetstone -
GIBBONS - David - EE Luton -
GILL - Don - Marconi Ltd -
GILLES - Dennis - University of Glasgow -
GILLOTT - Bill - EE NRL Stafford -
GILMOUR - Allan - EE NRL Stafford -
GLASS - G - EE Kidsgrove -
GOLDEN - Armand - UTECOM - Sydney -
GOODBODY - Tony - EE Marconi House London -
GRAHAM - Ronald - BAC Warton -
GRANT - Jeanette - Bristol Siddeley Engines -
GRAY - Alan - BAC Filton -
GRAY - Reg - EE Kidsgrove -
GREALISH - Rod - North Staffs Poly -
GREEN - David - EE Kidsgrove -
GREEN - Eddie - BAC Warton -
GREEN - Ronald - BAC Warton -
GREENHALG - Danny - Liverpool University -
GRIFFIN - Bob - Bristol Siddeley Engines -
GRIFFITH - Joan - RAE Farnborough -
GRIFFITHS - Eric - Bristol Siddeley Engines -
GRINDAL - Ola - Central Bureau of Statistics - Oslo -
GROVES - Peter - BAC Filton -
HACKETT - Winifred - EE Luton / Stevenage -
HAGERTY - Bob - EE Whetstone -
HAHN - John - BAC Filton -
HAINES - John - BAC Filton -
HALEY - Colin - EE Kidsgrove -
HALLIDAY - John - BAC Warton -
HALSALL - Sylvia - BAC Warton -
HAMBLIN - Charles - UTECOM - Sydney -
HANCOCK - Tony - EE Kidsgrove -
HAND - Liz - Bristol Siddeley Engines -
HARDING - Roy - EE Kidsgrove -
HARDMAN - Margaret - BAC Warton -
HARVEY - Roy - RAE Farnborough -
HATLEY - Ruth - EE Marconi House London -
HAWKE - M J - EE Kidsgrove -
HAWKINS - Neville - EE NRL Stafford -
HEMMING - Paul - Bristol Siddeley Engines -
HERBERT - John - EE Kidsgrove -
HETHERINGTON - Ron - North Staffs Poly -
HIGBY - Wendy - Bristol Siddeley Engines -
HILL - Les - UTECOM - Sydney -
HOCKNEY - Roger - EE Whetstone -
HOLE - Bill - RAE Farnborough -
HOLLAND - P B - EE Kidsgrove -
HOLLINGDALE - Stuart - RAE Farnborough -
HOLLINS - Dennis - EE Kidsgrove -
HOLMAN - David - EE Whetstone -
HOLT - Seth - EE Whetstone -
HORAN - Una - BAC Warton -
HORSLEY - Alf - EE Whetstone -
HORTON - John - EE Kidsgrove -
HOTHERSALL - Sylvia - BAC Warton -
HOUSTON - Iain - -
HUDSON - C - EE Kidsgrove -
HUGHES - Stella - RAE Farnborough -
HUGHES - Terry - BAC Warton -
HULL - Dudley - Marconi Ltd -
HUNT - Brian - Bristol Siddeley Engines -
HUTLEY - Norman - Marconi Ltd -
HUXTABLE - David - EE Kidsgrove -
Ian - ??? - EE Luton -
IDDON - Jack - EE Kidsgrove -
IRELAND - Ralph E - EE Whetstone -
IRVINE - Robin - EE Kidsgrove -
JACKSON - Stella - BAC Warton -
JACOBS - Eric - EE NRL Stafford -
JEAYS - Mike - EE Whetstone -
JENKINS - David - EE NRL Stafford -
JENKINS - Gwilym - RAE Farnborough -
JENNINGS - Alan - BAC Warton -
Jenny - ??? - Marconi Ltd -
JENSSEN - Dick - UTECOM - Sydney -
JOHNSON - Angus - University of Glasgow -
JOHNSON - Peter - North Staffs Poly -
JONES - Alan - EE Whetstone -
JONES - Gordon - EE Kidsgrove -
JONES - Mary - BAC Warton -
JONES - Roger - EE NRL Stafford -
Joyce - ??? - BAC Warton -
Judy - ??? - Marconi Ltd -
JULLIEN - Graham - EE Kidsgrove -
KALDOR - Tom - UTECOM - Sydney -
KAROLY - George - UTECOM - Sydney -
KELLEHER - John - BAC Filton -
KELLY - Mike - EE Whetstone -
KELLY - Vivian - EE Luton -
KELSEY - Sydney - BAC Warton -
KEMSHELL - Peter - RAE Farnborough -
KERR - Ron - BAC Filton / OSLO -
KEYSER - Richard - BAC Warton -
KILTY - Paul - BAC Warton -
KINGSBURY - Mike - EE NRL Stafford -
KIRK - Alan - Liverpool University -
KNUTSEN - Odd - Central Bureau of Statistics - Oslo -
LANCASTER - Peter - BAC Warton -
LANDIN - Peter - EE Marconi House London -
LANE - Dennis - Mobile Unit South -
LAVERTY - Chris - EE Kidsgrove -
LEAKEY - Peter - BAC Warton -
LEE - David - Marconi Ltd -
LEES - Jennifer - EE Marconi House London -
LE'GOODE - Michael - EE Luton -
LEIGH - David - EE NRL Stafford -
LEIGH - David - North Staffs Poly -
Leo - ??? - EE Luton -
LINDEN - Carl - RAE Farnborough -
LLEWELLYN - Tom - EE NRL Stafford -
LLEWELLYN - Tony - EE Marconi House London -
LLOYD - Josephine - BAC Warton -
LUCKING - Jim - EE NRL Stafford -
LUNN - Fred - EE Luton -
MacFARLANE - John - BAC Warton -
MACFARLANE - John - EE Kidsgrove -
MALLINDER - Mike - BAC Warton -
MALLOCH - James - BAC Warton -
MANSFIELD - Mike - North Staffs Poly -
MARCIANO - Fortunato - EE Kidsgrove -
MARSH - Janet - EE NRL Stafford -
MARSHALL - Nancy - BAC Warton -
MARVIN - Maurice - BAC Warton -
MATTHEWS - Margaret - BAC Warton -
MATTHEWS - Vic - EE Kidsgrove -
McCALL - Bob - Central Bureau of Statistics - Oslo -
McDONNELL - John - BAC Warton -
McFARLANE - R - EE Kidsgrove -
McHUGH - Brian - UTECOM - Sydney -
McINTOSH - Eric - EE Kidsgrove -
McKAY - Billy - Queens University Belfast -
McKEE - Fred - Marconi Ltd -
MENTJOX - John - UTECOM - Sydney -
MERRITT - Jack - EE Kidsgrove -
METZER - Kurt - EE Whetstone -
MIDGLEY - Margaret - BAC Warton -
MIEVELLE - Steve - Bristol Siddeley Engines -
MILLER - Brian - Bristol Siddeley Engines -
MILLS - Kathleen - RAE Farnborough -
MONTAGUE - Don - EE NRL Stafford -
MOONEY - Brian - Liverpool University -
MORRIS - Ray - EE Kidsgrove -
MORWOOD - Jim - EE Kidsgrove -
MOULD - Geoff - EE Kidsgrove -
MOULD - Pauline - EE Kidsgrove -
MOXHAM - Bill - BAC Warton -
MULLER-STOY - Peter - NPL -
MUNDAY - Brian - EE NRL Stafford -
MUSGROVE - Arthur - EE Luton -
NASH - Bill - EE Kidsgrove -
NEAL - Trevor - EE Kidsgrove -
NELSON - Alan - UTECOM - Sydney -
NENSETH - Thin - Central Bureau of Statistics - Oslo -
NEWMAN - John - EE Kidsgrove -
NICHOLSON - Albert - EE NRL Stafford -
OAKES - Mike - Bristol Siddeley Engines -
OATES - M - UTECOM - Sydney -
OBRAY - Roger - Marconi Ltd -
O'BRIEN - John - EE Luton -
OCCARDI - Val - Mobile Unit South -
ONEILL - Barry - BAC Warton -
OZANNE - David - EE NRL Stafford -
PACELLO - Ed - Marconi Ltd -
PALMER - Chris - RAE Farnborough -
PALMER - Drayton - Marconi Ltd -
PARK - Larry - UTECOM - Sydney -
PARKER - Phillip - EE Kidsgrove -
PARKES - L - UTECOM - Sydney -
PATTERSON - John - University of Glasgow -
PEACOCK - Alan - BAC Warton -
PEARCE - Heather - RAE Farnborough -
PEARSALL - Mr - RAE Farnborough -
PEDDER - D G - EE Kidsgrove -
PEDERSON - Ed - EE Kidsgrove -
Pete - ??? - EE NRL Stafford -
PETHERICK - Edward - RAE Farnborough -
PETRIE - Brian - BAC Warton -
PIGGOT - Ian - RAE Farnborough -
PILKINGTON - Barbara - BAC Warton -
PITT - Gordon - BAC Warton -
POLIKOFF - Mrs - EE Kidsgrove -
POOLE - Eddie - BAC Warton -
PORTEOUS - Janet - EE Marconi House London -
POWELL - Keith - EE Kidsgrove -
POWELL - Richard - Marconi Ltd -
PRICE - Vic - EE Marconi House London -
PRIESTLEY - Maurice - RAE Farnborough -
PRINCESS - Margaret - RAE Farnborough -
RANDELL - Brian - EE Whetstone -
RANYELL - Derek - EE Kidsgrove - EE Stafford -
RAWKINS - Jackie - MAFF Guildford -
REDDY - Raj - UTECOM - Sydney -
REDFERN - Philip - RAE Farnborough -
REID - Bob - UTECOM - Sydney -
RHODES - Tony - BAC Filton -
RICHARDS - Eric - EE Whetstone -
RICHARDS - Jim - EE Kidsgrove -
RICHARDSON - Jack - EE Marconi House London -
RILEY - Tony - EE Whetstone -
ROBB - Jean - UTECOM - Sydney -
ROBERTS - John - UTECOM - Sydney -
ROBERTS - Philip - BAC Warton -
ROBINSON - Alex - MAFF Guildford -
ROBINSON - Cliff - EE NRL Stafford -
ROBINSON - David - BAC Filton -
ROBINSON - David - EE NRL Stafford -
ROBINSON - Mike - EE NRL Stafford -
ROLLETT - John - NPL -
ROOCROFT - Valerie - BAC Warton -
ROSCOE - Dave - UKAEA Capenhurst -
ROUTLEDGE - Norman - RAE Farnborough -
ROWE - Brian - EE NRL Stafford -
ROWLEY - Geoff - RAE Farnborough -
ROYLE - Derek - EE Kidsgrove -
RULE - P - EE Whetstone -
RUNCIMAN - Phil - EE Whetstone -
RUSHTON - Eileen - EE Kidsgrove -
RUSSELL - Lawford - EE Whetstone -
RYAN - John - EE Kidsgrove -
SADLER - Marjorie - Marconi Ltd -
SALISBURY - Barbara - BAC Warton -
SAMET - Paul - RAE Farnborough -
SAVILLE - Arthur - Mobile Unit South -
SAVORY - Derek - EE Kidsgrove -
SCOTT - Wilf - EE NRL Stafford -
SCROGGIE - W - UTECOM - Sydney -
SEABROOK - Elizabeth - Marconi Ltd -
SHEFFIELD - Charles - EE Whetstone -
SHEFFIELD - Sarah - EE Whetstone -
SHELMERDINE - Malcolm - EE Kidsgrove -
SHINN - Doug - Marconi Ltd -
SINCLAIR - Bob - Bristol Siddeley Engines -
SKINNER - Janet - EE Marconi House London -
SKWIRZYNSKI - Josef - Marconi Ltd -
SMART - Ron - UTECOM - Sydney -
SMITH - David - EE Luton / Stevenage -
SMITH - Gerry - Central Bureau of Statistics - Oslo -
SMITH - Roger - EE NRL Stafford -
SMITH - Roy - BAC Warton -
SOLLY - Yvonne - Marconi Ltd -
SPRAGUE - Diana - Bristol Siddeley Engines -
STANLEY - Peter - EE Luton -
STEIRA - Kare - EE Kidsgrove -
STEWART - Karl - -
STOKES - H - EE Kidsgrove -
STOKES - Ron - EE Luton -
STOWER - Anne - EE Marconi House London -
STURGESS - Ernie - EE Whetstone -
SUMMERFIELD - Wally - EE Kidsgrove -
SUMNER - Gordon - BAC Warton -
SWAINSTON - Barry - BAC Filton -
TAIG - Ian - BAC Warton -
TATTERSALL - Philip - BAC Warton -
TAYLOR - Richard - North Staffs Poly -
TEE - Garry - EE Whetstone -
THACKERAY - John - Marconi Ltd -
THIRLBY - Rob - EE Whetstone -
THOMAS - Eric - EE Kidsgrove -
THOMAS - Jean - EE Marconi House London -
THOMAS - Ron - BAC Warton -
THOMAS - Wyn - MAFF Guildford -
THOMPSON - Frank - EE Kidsgrove -
THOMSON - Anne - University of Glasgow -
THORNTON - Barry - UTECOM - Sydney -
THORPE - John - RAE Farnborough -
TILBURY - Alastair - DSIR East Kilbride -
TITMUSS - Keith - UTECOM - Sydney -
TODD - J. K. - EE Kidsgrove -
TODD - Mike - EE Kidsgrove -
Tom - ??? - EE Marconi House London -
TOMLINSON - Arthur - EE Kidsgrove -
TOOTILL - Geoff - RAE Farnborough -
VALENTINE - Sam - North Staffs Poly -
VINCENT - Neville - Marconi Ltd -
VOWELS - Robin - UTECOM - Sydney -
VOWELS - Robin - UTECOM - Sydney -
WAKEFIELD - Mary - BAC Filton -
WAKELY - Peter - EE Whetstone -
WALKER - Jeremy - EE Kidsgrove -
WALKER - John - EE Kidsgrove -
WALTER - Pat - RAE Farnborough -
WALTERS - Doug - EE Kidsgrove -
WARDELL - Geoff - Marconi Ltd -
WARDEN - David - Marconi Ltd -
WARDMAN - Brian - BAC Warton -
WARKE - Mike - Bristol Siddeley Engines -
WATT - Jim - RAE Farnborough -
WEATHERILL - Michael - North Staffs Poly -
WEBSTER - John - UTECOM - Sydney -
WESSON - Anne - University of Glasgow -
WESSON - Noel - University of Glasgow -
WESSON - Noel - University of Glasgow -
WESTCOTT - Brian - Marconi Ltd -
WETHERFIELD - Mike - EE NRL Stafford -
WHITAKER - Oliver - BAC Warton -
WHITE - Tony - University of Glasgow -
WHITTAKER - Leslie - BAC Warton -
WHITWORTH - Rod - EE NRL Stafford -
WICKS - David - BAC Filton -
WICKS (Mrs) - ? - BAC Filton -
WILLIAMS - David - RAE Farnborough -
WILLIAMS - Douglas - RAE Farnborough -
WILLIAMS - Douglas - University of Glasgow -
WILLIAMSON - George - North Staffs Poly -
WILLIAMSON - Mick - EE Whetstone -
WILSON - Jenny - Bristol Siddeley Engines -
WINTER - Ruth - RAE Farnborough -
WOODALL - Chris - EE Marconi House London -
WOOLGER - John - EE Marconi House London -
WOOLLETT - Mike - RAE Farnborough -
WORTH - Bill - EE Whetstone -
WORTHY - Digby - Marconi Ltd -
WRIGHT - John - MAFF Guildford -
WRIGHT - Peter - EE Kidsgrove -
YORK - Ted - RAE Farnborough -
YOULE - Eric - EE Luton / Stevenage -
YOUNG - Bill - RAE Farnborough -
YOUNG - Richard - BAC Warton -
AWRE - Aldermaston
Bristol Siddeley Engines
British Aircraft Corporation - Filton
British Aircraft Corporation - Warton
Central Bureau of Statistics - Oslo
Central Electricity Generating Board
English Electric - Kidsgrove
English Electric - Luton
English Electric - Marconi House - London
English Electric - Stafford
English Electric - Whetstone
English Electric DMSU-South - London
Glasgow University
Liverpool University
Marconi Ltd
Ministry of Agriculture Fisheries & Food
National Physical Laboratory
North Staffs Poly
Queens University
Royal Aircraft Establishment
The New South Wales University of Technology
UKAEA Capenhurst
To return to the DEUCE homepage click the image below.
