This article was contributed by Sqn Ldr Derek C Tilford.
In 1979, having been flying as a navigator in the RAF, I was posted to my first 'desk job' - as a planner in the Air to Air Refuelling Cell at HQ No 1 Group.
This task meant that I would be living away from home during the week, living at the HQ. Here there were basically only two things to do in the evenings, watch TV, or drink in the bar - neither appealed to me to any degree. I had always had an interest in electronics -1 had made a couple of games for my children using relatively new digital electronic kits as opposed to wiggly amp projects and thought I might venture a little further into this new technology. At this time there was a project in the Practical Electronics magazine to build a 'home computer' from a kit based on a project that had been developed in the US. Nothing had ever existed like this before except very expensive commercial computers.
Date : Unknown
I thought this might be something for me to attempt, with all this spare time in the evenings. From the article I thought it was within my capabilities and reckoned that it might take a month to build, a month to debug, as it was bound to have some faults initially, and then we could see what a 'UK 101 computer' with 8K of memory could do. So I ordered my kit and when it arrived I realised what I had taken on - a host of components to be soldered to a circuit board, each with eight or so pins to be fixed firmly in place, each with its own designated position, mapped out on the board. Godfathers I thought, that's going to take a lot longer that a month to build.
So I started on assembling the various components into their slots on the board and working after dinner until late into the night I had the whole lot assembled within one week. To my amazement when I switch it on, up came a message 'Ready' - IT WORKED.
So what could this thing do. First of all I had to learn the basics of the Basic language, a new Bill Gates version developed especially for this machine. It was not difficult and very soon I was able to write small programs which were interesting and gave an idea of what might be achieved, but this was only scratching on the surface - what could it really do!
Working during the day, our task in the office was to plan air-air refuelling deployments of RAF Tanker and Fighter aircraft to various venues around the world. A regular task was the deployment of aircraft to Cyprus where they carried out exercises in the less crowded skies of the Eastern Mediterranean. To accomplish this we had to work out from first principles the fuel plan for each of the aircraft using hand held calculators - a 'small' task that only took over three days to calculate, then another two days to check and verify. It was not an easy task as each aircraft had its own rate of fuel consumption, and transfers of fuel from Tanker to Receiver could only take place within designated airspace along set routes around the globe. This seemed like it could be a reasonable test to evaluate the capabilities of this new machine.
So I started on the project - deployment of two tankers plus four chicks from UK to Cyprus. One tanker taking the four chicks from UK to an area in the Mediterranean between Malta and Sicily where the second tanker would rendevous with the formation and take over the four chicks and proceed on to Cyprus with them.. Things one had to take into considerationwere rates of fuel usage in the climb, during flight the rate of burn by the fighters was fairly constant, but the tankers rate of bum changed considerably as it used fuel itself/gave fuel to the fighters en route, and then the final descent leaving sufficient safety fuel for a diversion capability if the weather was inclement at the destination.
Using a set of tables within the program I persevered with the program and found that I could just squeeze this program using 7.5K of memory. All that was need as input into the program was the type of receiver aircraft (fuel burn rate), together with the distances along track and the length of refuelling brackets and a wind component that we obtained from an ICL 1900 computer that languished in the corner of our office - known as the 'Green eyed Goddess'. With this information and inputting it into my program we could obtain a refuelling plan for the whole deployment in half a day - WOW. We compared it against our hand calculated plans over a couple of exercises to prove whether the program produced reliable results, which was duly confirmed.
Thus proven, I took my machine in to the office and demonstrated it to my commander. "Great" he said having witnessed an impressive demonstration, there's only one thing wrong with it - when you're posted from here you'll be taking that with you, won't you! Right Boss! Right, seeing as you have developed that program on your machine, I want you to transpose it onto that machine in the corner - 'The Green eyed Goddess'.
1 was allocated 8K of memory on the ICL and set to work. The Basic interpreter was not as efficient as that on the UK101 and having used up the 8K of allocated memory, was barely halfway through my program. I was allocated a further 8K and still when this was used up I needed a further 4K of memory to complete the program. Once this was completed the three of us in the office all used the program to assist with the planning of our detachments - a significant improvement in the efficiency of our office. It was still in use when I finally left the office in 1982 for my next posting.
Having learnt a new skill, the ability to program both my UK101 and use the Basic interpreter for programs on the ICL 1900, and later a 2900 I was asked to solve some of the other probiems within the HQ, by producing suitable programs on the ICL machines. The Green eyed Goddess was in fact a remote terminal from a computer located in HQ Strike Command. Two particular programmes that I managed to produce were one to manage the Groups Fuel Allocation for different aircraft/airfields during the 1981 fuel crisis and another to speed up the producing of competition results for the annual bombing competition between the US and RAF crews. The problem was that computer programmers were in short supply within the RAF and their expertise was restricted to solving problems at the higher HQ, rather than out in the sticks at Group level, so I was in demand despite this being a secondary activity from my prime task.
In 1982 when I was due for my next posting -1 had been involved in the planning of the Black Buck raids in getting Vulcan aircraft down to the Falkands, an effort to dislodge the Argentinian forces who had invaded the Islands on the ground, I thought that for the first time in my career, I was in the right place at the right time. It had been demonstrated that we were drastically short of Air-Air refueling assets, we were receiving VC10 aircraft which had been obtained from one of the colonial airlines and were receiving new tanker aircraft from Boeing. In the HQ we were drawing up safety procedures for the operation of these new aircraft and I felt I was ideally placed to be posted back onto flying with one of the Squadrons that would be formed as the new aircraft came on stream.
But no - it wasn't to be. Sitting in my office one day my phone rang and a voice on the end said" Hello Derek, you don't know me, I'm your desk officer, so I know a little about you and I think you can help us out with a problem we have". Uhuh I thought, then he started to spell out the specification of the job he needed to fill - Must have 'Command and Control' experience (OK that's been my job for the last 3 years), Must have on-line experience of computers, (OK, that's the Green eyed Goddess), Must have a formal ADP training - that you don't have but we can rush you through the requisite course - (Help I thought, he's digging a hole for me in the new weapons range up north) and lastly, must be able to speak German -1 gather you may need a refresher and that's no problem - (What on earth was he lining me up for?) - this would be for a new post with the German Air Force at a computer centre near Bonn. You'll be the first of three officers to work on a Command and Control Computer system run by the GAP. I want you to go home tonight, talk it over with your wife and let me know tomorrow whether you will accept this posting.
Discussing it with management, we had enjoyed our first tour in Germany in 1962 and as this appeared to be a three year tour, it should be different working with another Air Force. So the next day I called my desk officer and confirmed that we would accept this posting. True to his word I was rushed off to do a System Analysis course with the Civil Service, a data communications course with the National Computer Centre (NCC) and a manufacturers course on the Siemens BS 1000 before finally spending a month at the RAF language school to brush up my German (which I had learnt on the ski slopes down in Bavaria on Escape and Evasion training course and ski fitness expeditions back in the 1960s)
So I duly arrived at my new unit, the System Centre for the EIFEL (Elektrinische un Informations Fuhrungs System der Luftwaffe) Command and Control System. As the first British Officer, because of my background I was assigned to the Test and Integration section. Our responsibility was to test the on-line system for any errors, check out software developed to rectify any faults and new software developed for new functions and when we had given it the seal of approval, it was our responsibility to go out to each of the four computers located around the NATO Centra! Region and install the new software. Quite a responsibility to go to these computer centres and enter the main computer room, full of equipment but all silent, switched off ready for you. By the end of the day everything would be up and running with the latest software running having been tested in situ on the new system you had just generated. Some change form starting up and loading my UK101.
At the end of my three years in post I was asked if I would extend for another three years. Located in the Hunsruck mountains, south of the Moselle River, we didn't need much persuading. Halfway through this second tour it was decided that the BS 1000 machines were creaking at the seams, they were now nearly ten years old and needed to be upgraded to the new Siemens BS 2000 machines, and at the same time redevelop the Command and Control system into a C3 system (Command Control and Communications). This part of the upgrade was awarded to Dormer who had a name for developing satellite software in German Industry. It was decided that we needed someone from the British contingent to assist with this new development and keep an eye on the software development to ensure that it provided the facilities which we were helping to pay for. It was decided that I should be our man on the EIFEL II project. As a result I had to leave the beautiful Hunsruck and go and work with the German Procurement Executive (BWB - Bundesamt fur Wehrtechnik und Beschaffung) at their offices at St Augustin - just across the river Rhine from Bonn. Here my task was somewhat similar in that we had to monitor the development of the system by Dormer who were developing the application software and CSID (Computer Systems International Deutschland) who were tasked as subcontractors to develop the system software to work on the Siemens mainframes. At the end of 5 years in this post I had to leave as I only had some six months to serve with the RAF before retiring. We needed to resettle back in England having been away for nearly 10 years - all because of my original project - to build a UK 101.
Sqn Ldr Derek C Tilford
This exhibit has a reference ID of CH16778. Please quote this reference ID in any communication with the Centre for Computing History.
Click on the Images For Detail