"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)

J. K. BROWN

NRL Director

"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 )

Wilf SCOTT

Cliff Robinson reported to Wilf Scott, who protected us from the fearsome J.K.Brown."
(Mike Wetherfield)

Cliff ROBINSON - [EA]

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.

Mike ROBINSON - [EA]

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!

Mike WETHERFIELD - EA

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.

Brian MUNDAY

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)

Mike KINGSBURY

Deuce programmer, already at NRL when I joined. Extrovert and a keen hockey player.
(Mike Wetherfield)

Bill GILLOTT

Deuce programmer, the next to join after Jim Lucking.
(Mike Wetherfield)

Dennis BLACKWELL

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

Neville Hawkins wrote both the Interpreters (Mark 1 and Mark 2) which executed the code produced by the corresponding Alphacode Compilers.
(Mike Wetherfield)

Bernard CARRÉ

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)

Allan GILMOUR

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

John DENISON

"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)

Roger SMITH

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 BOOTHROYD - Mathematics Dept - NRL

"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.
[English Electric Co., DEUCE News No. 36, April 1959]

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://www.users.bigpond.com./robin_v/acm-290.inc ]
( Robin Vowels)

"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 HALEY

"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)

Reg ALLMARK

"A natural electronics engineer – absolutely brilliant - no one better in the team."
(John Boothroyd - Deuce Reflections )

Albert NICHOLSON

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)

David OZANNE

"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)

Tom BUTLER - [EA]

"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 ]

David LEIGH

"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

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)

Wally SUMMERFIELD

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 ROBINSON

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)

Janet MARSH

"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)

Joseph GALVIN

My younger brother - Joseph Galvin - was photographer at English Electric in the 1950's (Bill Galvin)

Brian ROWE

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)

Rod WHITWORTH - [EA]

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.

Mike DREW - [EA]

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.

Roger Bishop JONES - [WWW] - [EA]

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.

Eric JACOBS - David JENKINS - Tony LLEWELLYN - Neil DONOHOE - Don MONTAGUE

J. K. TODD

"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)

Alan ASBURY

"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)

HARDWARE DEVELOPMENT

Frank THOMPSON

"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)

Fortunato MARCIANO aka "Rocky"
"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) -

John RYAN - [EA]
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, as in ON TOP OF, as I had to climb on top one day to repair a failed address tree.

Tom ELLIOT - John NEWMAN - Derek SAVORY - Vic MATTHEWS - Eric THOMAS - Jim MORWOOD - Roy HARDING
Ray MORRIS - John HERBERT -Norman DOWELL - Peter WRIGHT

MAINTENANCE ENGINEERING

Derek ROYLE
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)

Susan ELLIOTT - [EA]
Derek Royle's secretary
Married Jeremy Walker

Jeremy WALKER - [EA]
Digital Computer Mobile Service Unit - North
" 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 .

Brian BISPHAM
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 out of 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)

Neil CHARLESWORTH - [EA]
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 !

Eric DOBSON
DEUCE Maintenance Engineer

Bill BECKETT
DEUCE Maintenance Engineer

Doug WALTERS
Deuce Spares Manager

DEUCE COMPUTER BUREAU

Noel WESSON - [EA]

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.

(Extract from "My Deuce Time", discover how Noel was lured from LEO by a bevy of 545's !)

Mike GAHERTY
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)

Derek RANYELL
DEUCE Bureau Maintenance Engineer

"Derek Ranyell (was one of) my first engineers, although (he) went back to Stafford after a short while."
(Noel Wesson)

Brian COLLING
DEUCE Bureau Maintenance Engineer

Ed PEDERSON
DEUCE Bureau Maintenance Engineer

Keith POWELL
DEUCE Maintenance Engineer

John MACFARLANE
DEUCE Maintenance Engineer

Mrs POLIKOFF
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)

PROGRAM DEVELOPMENT & TRAINING

David GREEN - [WWW] - Deuce Documentation - [EA]

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 ?. 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. 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. 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 !!! Ed)

David remembers his first few months with English Electric. "Up to English Electric - 9.50 Euston to Crewe"

Roger ALLWOOD

"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.

Richard BURROWS
Another programmer reporting to Roger
(David Green)

Fraser DUNCAN

"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 )

David HUXTABLE

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

Mike TODD - [EA]

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.

Robin IRVINE - [WWW] - [EA]

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.

Eric McINTOSH - (EA)

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) "

Graham JULLIEN - [EA] - [WWW]

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.

Jim EDWARDS -Arthur TOMLINSON - Tony HANCOCK - R. ASHBROOK
S. BIGGINS - K. BLOOR - Alex BRUNT - Vic COLDHAM - G. GLASS - Reg GRAY
M. J. HAWKE - P. B. HOLLAND - C. HUDSON - D. PEDDER - M. SHELMERDINE
H. STOKES - John WALKER

PRODUCTION

Bill NASH

"Our head of product development was Bill Nash. I am standing next to him in the photo"
(Dennis Hollins - SeeThe 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 - SeeThe Model Deuce Letters to the Editor )

Dennis HOLLINS

"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

Stan ELKIN

"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. SeeThe Model Deuce
I worked with Stan in later years." (Jack Merritt)

Len CALVERT

"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. ) See also The Model Deuce

Eileen RUSHTON

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, & had a bench job building the plug-in chassis. (Jack Merrit)

Geoff MOULD- [EA]

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, & 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 MOULD

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)

Jack MERRITT- [EA]

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.

ENGINEER TRAINING

Arthur BAILEY
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, at right, lecturing to a class of attentive students at Kidsgrove in August 1958.
(Noel Wesson)

Jim RICHARDS
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 JONES
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)

Kåre Steira, (Norway)
Chris Laverty, Belfast (probably Queens Uni)
U. J. Amera - Singhe, Ministry of Agriculture, Fisheries & Food, Guildford, Surrey
R. McFarlane, St Helens, Lancs
John Horton, Blythe-Bridge, Staffs
Elizabeth Christopher, Staffs

Phillip PARKER - [EA]
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)

Noel WESSON - [EA]

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.
(Extract from "My Deuce Time", discover how Noel was lured from LEO by a bevy of 545's !)

Noel then moved to Kidsgrove to look after the Deuce Bureau Machines.

Bob COLLINS - Ron BRIGGS - John BISHOP - Arthur SAVILLE - Dennis LANE - Val OCCARDI

George DAVIS - [EA]

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!

Jack RICHARDSON
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)

Doug FLOWER - [EA]
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.

Jean DACE nee THOMAS
"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)

Peter DOCHERTY
"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)

Audrey BIRCHMORE

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."

Richard FORRESTER
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

Vic PRICE
Ron EITEL
Peter LANDIN
Chris WOODALL
Anne STOWER (Married Chris WOODALL)
John WOOLGER

Peter WAKELY -
Mathematician/Programmer
Director - English Electric Mechanical Engineering Laboratory.

Charles BROYDEN - [EA]
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 the quasi-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.

Fred FORD
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)

Charles SHEFFIELD - (WWW)
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) )

Sarah SHEFFIELD
Mathematician/Programmer [APD]

D HOLMAN
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]

Garry TEE - (EA)
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 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

Mike KELLY
With colleague Brian RANDELL developed EASYCODE at Whetstone.

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

Brian RANDELL - [WWW] - [EA]
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 !!

Mike JEAYS - [WWW] - [EA]

"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."

Programmers

Mick WILLIAMSON --- Lawford RUSSELL --- Kurt METZER --- Eric RICHARDS ---P RULE

Roger HOCKNEY - (WWW)
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)

Maurice BATEY - [EA]
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!

Deuce Operators

Seth HOLT - (EA)

"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

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)

Alan JONES --- Ron GENT ---Cyril ATKINSON ---Rob DUDLEY

DEUCE Maintenance Engineers

Bill WORTH
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)

Alf HORSLEY --- Tony RILEY - [EA] --- Ernie STURGESS --- Maurice DENDLE

Bob HAGERTY
Punch Room Manager

Winifred HACKETT
Head of GW Division

John O'BRIEN
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)

Programmers

Dick BOND --- Arthur MUSGROVE --- Vivian KELLY --- Michael le'GOODE --- David GIBBONS --- Ian ???

Jim FISHER - [WWW] - [EA]

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.

Eric YOULE - [EA]

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.

Operators

Phyllis EVANS -- Leo ???

Ron STOKES
"My immediate boss then was Ron Stokes who had some DEUCE experience but I don't remember where." (Peter Stanley)

Peter STANLEY - [EA]
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

Joan FARMAN

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)

Deuce Maintenance Engineers

David SMITH- [EA]

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.

Roger BIRD

"The DEUCE did actually go to Stevenage after I moved to Kidsgrove. The engineer in charge then was Roger Bird." (Peter Stanley)

Fred LUNN

Peter BRADSHAW - [EA]

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!

John ROLLETT - [WWW]
"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)

MÜLLER-STOY - [EA]

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.

Pat ELLIS
Chief Programmer
Primarily responsible for the development of TIP

Brian MILLER - [EA]
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."

Paul HEMMING - [EA]

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.

Jim FLETCHER - [EA]
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."

Mike WARKE
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)

Bill COULSHED
"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 DUERDEN
"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)

John McDONNELL - Head of Maths Services
"John was one of life's great gentlemen, as well as being a first class mathematician." (Peter Dukes)

Peter DUKES - [EA]
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

David BOOKER - [EA]

Sydney KELSEY
"Was doing good work at Warton before he joined Argyris at Imperial College." (Peter Dukes)

Ian TAIG
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 HALLIDAY
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 ]

Richard YOUNG
"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 ]

Peter LANCASTER- [EA] - WWW

Extracts from an interview by Prof. Nicholas J. Higham at the University of Manchester, March 15, 2005 . Full text in PDF format available here.

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.

Maurice MARVIN
Programmer in Aerodynamics department who married Josephine LLOYD

Philip ROBERTS
"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)

Bill MOXHAM - [EA]
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!!

Mary JONES - [EA]

As I came across your site I thought I would add myself to you "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]

John MacFARLANE
"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

Steve ALLCOCK - [EA]

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.

Al BEEDON
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)

Terry HUGHES
DEUCE Maintenance Engineer BAC Warton

Fred DAVIES
DEUCE Maintenance Engineer BAC Warton

Eddie POOLE - [EA]
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)

Robin VOWELS
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 !Josephine MARVIN nee LLOYD
"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 ALLCOCK nee SALISBURY
"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)

Margot CAREY - nee CLAPHAM - [EA]
" 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)

Akim ADIWALI --- Marie BAMBER --- Hilda BERRY --- Sheila BREWSTER nee BRADLEY--- Muriel DENN
Pauline DENNETT --- Mike FLEMING --- Sylvia HALSALL --- Margaret HARDMAN --- Una McLAUGHLIN nee HORAN
Sylvia MILLER nee HOTHERSALL--- Shirley JACKSON --- Mike MALLINDER ---
Margaret SMITH nee MATTHEWS--- Margaret DOWNER nee MIDGLEY--- Nancy MARSHALL nee BOULTON
Brian PETRIE --- Barbara PILKINGTON --- Valerie ROOCROFT --- Leslie HOLDEN nee WITTAKER --- Dorothy BLACKWELL
June CHAMBERS --- Sheila BRADLEY

John HAHN
- Head of Maths Services

Peter FRANKS
- Head of the DEUCE programming group

Peter GROVES - [EA]
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!

Tony COOKES - [EA]
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 24x7.

Barry SWAINSTON - [EA]
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.

David WICKS
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]

Mrs WICKS

Programmers

Mike FASEY --- Tony RHODES --- Roger COLLIER ---Ron KERR ---- Lyn EDWARDS --- Eileen ELLIS --- David ROBINSON
John HAINES --- John BELL --- Mike DAVIS

Punch card girls

Mary WAKEFIELD --- Diana "Dinks" BOND

John KELLEHER - [EA]
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.

Alan GRAY - [EA]
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.

John ELLISON
EE DEUCE Maintenance Engineer

David BACK
EE DEUCE Maintenance Engineer

Peter BENSTEAD
DEUCE Maintenance Engineer
He worked at CEGB Deuce and was on my Deuce course (Noel Wesson)

Eric EASTWOOD
"Director of Research at Great Baddow - later Sir Eric Eastwood and President of the IEE." (Bernard de Neumann)

Peter BRANDON
"Led the Mathematics Group, went on to become Prof of Electrical Engineering at Cambridge." (Bernard de Neumann)

Norman HUTLEY
"Led the Programming Section under Peter Brandon." (Bernard de Neumann)

Josef SKWIRZYNSKI
"Led the the Mathematical Physics and Circuitry Section under Peter Brandon." (Bernard de Neumann)

Doug SHINN
"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)

Bernard de NEUMANN - (EA)
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 SADLER
"Marjorie worked on the DEUCE at English Electric House on the Strand and moved te Great Baddow because of her prior experience with the EEH Deuce. Wrote a PIP ( Polynomial Interpretive Program) interpreter.' (Bernard de Neumann)

Ed PACELLO
" Ed worked on antenna and circuitry software and developed general tools." (Bernard de Neumann)

Denis BROCKINGTON
"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)

Neville VINCENT
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.

Mike ADLER
Worked under NH on finite element techniques (Bernard de Neumann)

Ted COCKLE
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 SEABROOK
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)

Julia DAIN - [EA]
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.

Programmers

Daphne GALPIN --- Dudley & Mrs HULL ---Drayton PALMER ---Don GILL---Rosalind FARTHING ---Ed DUNLOP
Roger OBRAY ---Ken BROWN ---David WARDEN ---Alan BOYCE --- Mabel CLARKE ---John THACKERAY
Digby WORTHY ---Fred McKEE ---Brian WESTCOTT ---Judy ???

John COOPER - (EA)
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!

Doug FAWCETT
Operator

Sid BENTALL
Operator

Geoff WARDELL
Chief DEUCE Maintenance Engineer

David LEE - (EA)
DEUCE Maintenance Engineer

"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"

"John Cooper (almost) remembers some names of the people who worked in the off line activities associated with the Deuce." (JB)

Linda BALLS -- Jenny ??? --- Carol ??? --- Barbara

Yvonne ???
Head Punch Girl and IC Tabulator Room (Bernard de Neumann)

Alex ROBINSON
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)

John WRIGHT
English Electric DEUCE Maintenance Engineer

John BARRETT
English Electric DEUCE Maintenance Engineer - [WWW]

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."

Wyn THOMAS
Operator MAFF Guildford

"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)

David LEIGH - [EA] - Lecturer

"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."

Melvyn CHAPMAN - [EA] - 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."

Sam VALENTINE - [EA] - 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."

George WILLIAMSON - 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)

David BROWN - [WWW] - [EA] - 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)!"

Rod GREALISH - [EA] - 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."

Ron HETHERINGTON - [EA]

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!

Richard TAYLOR

" '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."

"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 )

Mike MANSFIELD- [EA]

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.

Peter JOHNSON - [EA]

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

Clinton BOURNE

"I was told about your web site by Clinton Bourne (he worked on DEUCE at Kidsgrove from 1962 until 1965 or thereabouts)."
(Peter Johnson)

Michael WEATHERILL - Lecturer

"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)

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:

R. G. SMART -
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)

George KAROLY
Lecturer in Electrical Engineering (Programming)

"George Karoly did some programming work too, including for the Soda project." (Robin Vowels)

Miss M. OATES
Technical Officer (Programing and Secretarial duties)

Miss G. GASKIN
Technical Assistant (Card Preparation)

Keith FORD
Laboratory Assistant (Machine Maintenance)

W. SCROGGIE
Laboratory Assistant (Machine Maintenance)

L PARKES
Technical Officer - replaced Miss M Oates

Margaret FOSTER
Technical Assistant - replaced Miss G Gaskin" - (UTECOM 1st Annual Report)

Barry THORNTON
Team Leader - (Trevor Pearcey)

"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)

Larry PARK
Senior Programmer - (Trevor Pearcey)

Gordon BELL - (EA) - [WWW]

"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.

"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."

Bob BRIGHAM - (EA)

"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

David ELLIOTT
"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

Charles L. HAMBLIN - [WWW