Clive Sinclair formed the first company to bear his name in 1961 while he was a 21-year-old electronics journalist. He had received no formal education in the subject, but as a highly intelligent autodidact, he had developed a passion for electronics during his teens. He also had an entrepreneurial bent, and even in his youth spotted a business opportunity in designing and selling kits to fellow electronics enthusiasts.
Unfortunately, his initial attempt to do so — Sinclair Radionics’ first incarnation — exploded before it had even lifted off the pad; Sinclair’s finance provider unexpectedly pulled out during the countdown. Worse, Sinclair had just quit his well-paid job at Bernard’s Publishing, a London-based purveyor of electronics books and periodicals, in order to work on his new business full-time. No wonder the abrupt withdrawal of the seed capital left Sinclair, now unemployed and desperately trying to get freelance commissions from his former magazine colleagues, with an abiding unwillingness to be forced to rely on anyone else to fund his business activities. From now on, he insisted, Sinclair businesses would be funded only out of their own growth.
Undeterred by this first failure, Sinclair tried again the following year, this time going it alone and keeping his full-time job. He brought to market a micro-amplifier with the unique selling point of being the “smallest of its type in the world”, or so Radionics’ advertising claimed. Radionics’ early products were financially successful in large part because of Sinclair’s ability to find very cheap components. Thanks to the contacts he had made as a journalist, he located manufacturers who were happy to sell sub-specification parts at a substantial discount. The amplifier, for example, was built out of transistors which Semiconductors Ltd, a joint venture between Britain’s Plessey and America’s Philco, had rejected during manufacture. Sinclair discovered that these parts were nonetheless within the tolerances required by his gadget, or close enough, so he bought them on the cheap and added them to his kits.
Having quickly gone back to technical writing following Radionics’ failure to launch in 1961, Sinclair wisely remained a journalist and author while his second stab at a business began to take off. Continuing to write brought in new opportunities too: he could pen articles detailing projects which used the very low-cost components he was able to procure, ensuring a ready market among hobbyists keen to build the devices he had described in print. However, by 1963, Radionics had grown sufficiently for Sinclair to run the company full-time. He began hiring. Fifteen year old Jim Westwood, who was to become one of Sinclair’s most able and closest lieutenants, joined the company from a components supply shop, Technical Services Ltd., in September 1963. He would remain a Sinclair employee for the next four decades.
The 1960s: the Radionics decade
For the remainder of the 1960s, Radionics’ growth would continue on the back of a series of radio devices and hi-fi components conceived by Sinclair and realised by Westwood and a growing team of new recruits and allied contract engineers. Kits slowly gave way to more consumer-oriented ready to use hi-fi systems. In 1966, the company also announced it was moving into the visual realm with the Microvision, a pocket TV set, which it promptly but prematurely demonstrated in prototype form at a press reception. The assembled hacks were generally sceptical, but their dismissal of the miniature set turned out to be justified: Sinclair promised the Microvision would go on sale in 1967, but it would be a further ten years before his portable television was finally made available to the public. This was the first of many over-enthusiastic promises Sinclair-owned companies would make and then fail to deliver. For now, though, Radionics was on the up and up. In 1967, six years from its inception and five from its true beginning, the company reported a turnover of £100,000.
Sooner or calculator
Through the late 1960s and early 1970s, Radionics’ business continued to grow. The firm slowly began to move away from low-cost hi-fi products, a market by then slowly being dominated by Japanese companies able to make better sounding and more reliable devices than the locals — and Radionics in particular — could produce. We’ll see this happen again, in other markets, but Radionics was now about to enjoy its first true mainstream success on the back of what was a genuinely innovative product: the world’s first truly pocket-sized calculator. Sinclair’s passion for miniaturisation — recall the world’s smallest amplifier — was the driving force; the increasing complexity and reduce cost of silicon chips was what made it possible. So-called pocket calculators existed already, but they were essentially battery powered and scaled down versions of the bulky desktop calculators that had come into vogue in the mid-1960s to superseded the previous generation of even more massive electrically powered but mechanically based calculating machines. The current ‘pocket’ calculators were really just portable models.
Sinclair learned of a single, tiny chip which integrated all the circuits required to do arithmetic. The existing machines used collections of chips for the same tasks, so this one had the potential to shrink the entire device down to, say, the size of a cigarette packet. Sinclair handed the chip to his engineers who, with some very creative thinking got it not merely working but able to run off the tiniest batteries available. The upshot was the Executive, which was released in September 1972 for just under £79.95 — less than 80 per cent of the price of the nearest competition. It weighed a mere 71g. Radionics had its first true hit. Executives flew off the shelves; businesspeople bought them simply because they could and because Radionics sold the calculators through retail, not just off electronics magazine pages. The Executive’s black polycarbonate casing won its designer, Richard Torrens, a Design Council Award for Electronics.
A cheaper, more everyday calculator, the £24.95 Cambridge, followed in August 1973; three months after that came an upgraded Executive, the Memory, which was able to remember a calculated result and use it in the next sum. The Scientific, which offered some advanced maths functions, was released in March 1974, and a programmable versions debuted in August 1975 — right about the time the calculator market was starting to collapse. When the Scientific was released, it cost £14.95 in kit form or £21.55 pre-assembled. The Cambridge was then selling for just under a tenner as kit, or £13.99 if you wanted it ready to use. The kit price represents a drop of 60 per cent over a period of just seven months.
Trouble on Clive’s watch…
Radionics may have been first to market with a pocket calculator, and quick to release follow-up products, but it was so easy to see what it had done that it was almost immediately competing with companies who should churn them out too — hence those massive price reductions. By the time the Cambridge was launched, new entrants, including chip maker Texas Instruments, the source of the Executive’s breakthrough chip, had released product of their own. Established calculator brands, among them Britain’s Sumlock, America’s Commodore, and Japan’s Casio and Sharp, leapt into the fray. Soon every one of these manufacturers and more besides had a pocket calculator on the market. Some of them delivered more functionality or simply worked better than Radionics’ offerings did, which notoriously suffered from glitches and quirks. By 1974, world production was 20 million units. For a time, Radionics could ride the rising tide of new buyers and people replacing less functional devices with newer more capable models, and it quickly developed a very strong export business, but by Christmas 1975 the market was saturated. Radionics would continue to offer new models throughout the next few years, but never with the success it had enjoyed in 1972.
Radionics R&D was still not ready to hand over a marketable pocket TV, so the company had one glimmer of hope: an entry into the emerging digital watch market. Some hope. By 1975, there were more than a hundred digital watches available from dozens of manufacturers so Radionics’ offering would not be a first. The plan had been to come to market in 1973 or 1974, but problems getting its central chip ready — an attempt to integrate several chips into a single one — first with Mullard and then ITT proved difficult, as custom chip design almost always is. Getting the design right is only part of the process; next you have to iterate the production process sufficiently to get the yield that will ensure the working chips that you produce are cost-effective at the price you will retail your product for. No chip production now is capable of a 100% yield, and in early stages of production the yield can be very small indeed. In the mid-1970s, yields would have been significantly lower at each state in the development of a chip. Finally, the chips started to come through, and in November 1975, Radionics introduced, but did not ship, its new product: the Black Watch. It went on sale, in both ready-built and kit forms, in January 1976.
Within a very short space of time any Radionics staff anticipating that the digital watch would make up for evaporating profits on sales of calculators were given a rude awakening. The Black Watch was a car crash. Yes, it looked amazing, unlike anything currently on the market, and, at £24.95, not particularly expensive, but technical flaws rendered it a highly unreliable timepiece. Kit makers could probably cope with the quirks of the design, but consumers expected products to work reliably right out of the box. Radionics sold plenty of Black Watches — almost 20,000 in the first month, the company claimed — but so many came back for replacement. Ship two or even three products for the price of one — the initial sale and its replacements — and you can wave goodbye to any profits you expect to make. Dealing with disappointed customers cost Radionics hundreds of thousands of pounds and contributed to the firm’s first ever loss: despite sales of £5.6 million for the year to April 1976, Radionics was in the red to the tune of £335,000.
The company could no longer afford to continue developing the much promised but still unreleased pocket TV set or continue to fund Sinclair’s electric car dreams, manifested in a Cambridge University research project into which he was pumping Radionics money. Suddenly the company’s survival as a going concern was called into question. Neither the watch nor the calculators were going to contribute much in the way of financial buoyancy, and while the instrument business was very successful, it lacked the volume of sales to keep the ship afloat on its own. Unwilling to abandon his pet projects in order to focus on what could be delivered here and now, or to undertake any form of restructuring, Sinclair reluctantly accepted the stark choice offered him by the company’s financial advisor, merchant bank NM Rothschild: find some more money from somewhere or shut Radionics down.
Sinclair was eventually able to win government funding for the development of a flat television tube. This brought in £1 million of R&D grants from the National Research Development Corporation (NRDC) to be provided over a four-year period from October 1976. But other sources of finance that could be used to operate the business itself proved elusive. Time was running out. So in June or July 1976, Sinclair called the NEB.
The Sinclair side-hustle
But let’s step away from the deeply troubled Sinclair Radionics for a moment, because it was not Clive Sinclair’s only business. Three years previously, in March 1973, a company called Ablesdeal Limited was founded by Business Economy Products, one of the many British firms that specialised in registering new concerns on behalf of company founders. Ablesdeal was formally incorporated on 19 September 1973. Some 16 months later, on 15 January 1975, the company’s name was changed to Westminster Mail Order Ltd., the move being formally recorded on 11 February. Whatever the reason for the change of name, the company’s designation was revised again 17 months later. Westminster Mail Order became Sinclair Instruments Limited on 9 June 1976. If Sinclair wasn’t behind Ablesdeal at its foundation — the only names listed in Companies House records are those of the company registration agency — he certainly was now. The name change was the result of an extraordinary general meeting held at the same address as Radionics’ HQ: Enderby’s Mill, usually referred to simply as ‘The Mill’, in St Ives, Cambridgeshire. Formal recognition of the change of name was granted on 12 July 1976.
The most recent change of name occurred precisely at the time that Sinclair was beginning to negotiate a Radionics survival plan with the NEB. Perhaps Sinclair viewed the new company as his next venture if the NEB either refused to rescue Radionics and the business collapsed or, in return for a substantial cash injection, the agency demanded greater control of the company than he was willing to cede. That was certainly how the move was portrayed long after the event. Whatever the reason for the activation of Sinclair Instruments at this time, Sinclair’s own worst fears were not immediately realised. After a three-month investigation into the business, the NEB agreed to throw Radionics a £650,000 lifeline in return for 43 per cent of the company. Sinclair would remain in charge. The rescue deal was made public on 23 November 1976. We’ll come back to the effect the NEB intervention had on Radionics, and its implications for Clive Sinclair’s subsequent career in the microcomputer business later.
With the NEB now on board, and Radionics’ near-term future secure, what to do with Sinclair Instruments? Enter into our story one Christopher John Curry. At this point, Curry had been working for Radionics for a decade. He joined the company in the Summer of 1966 when he accompanied a chum, Lyndsey Lloyd to an interview with the firm. Curry may have just been tagging along, or so he said, but he nonetheless got a job too.
By 1976, and the arrival of the NEB, Curry was very close to Sinclair. He had first worked on some of the Radionics’ hi-fi products, but it was his construction, with Sinclair’s long-standing sidekick Jim Westwood, of the prototype of the Sinclair Executive calculator and the source of Radionics’ huge success in the early 1970s, for which he won kudos from Sinclair. “Clive returned from the States with the first single-chip calculator. He gave it to me with a wedge of paper and said ‘get that working’. It was completely new to me. We built a breadboard around the chip and built a keyboard from bent wire. After a little fiddling, the thing worked. It really was like magic to see those numbers appearing on the display; and then when you used one of the functions and the result flew across the screen — it was incredible. To see this happening with this little piece of electronics was really exciting.”
Curry, like Westwood, was now one of those people whom Sinclair deeply trusted — Rodney Dale, in his official biography of Sinclair, calls Curry “a Sinclair confidant”. Curry was someone with whom Sinclair could thrash out his ideas and get exactly the right mixture of challenge and affirmation that his ego required. It should be no surprise, then, that Curry was in on Sinclair’s brainstorming sessions over what to do if Radionics went bust.
Whatever the nature or extent of their discussions, the outcome was that Sinclair told Curry to set up an office well away from Radionics and establish Sinclair Instruments as a business there. The name of the new venture, chosen in the same month that the NEB first came to all, June 1976, perhaps signals where Sinclair had decided his new business’ future lay. Radionics launched its first instrument product, a digital multimeter, in 1972. Designed by AIM, a consultancy also based at The Mill, the DM1 was novel, inexpensive and popular. So popular, in fact, that Radionics couldn’t initially supply enough to meet demand. By the time it could, however, it found that potential buyers had cottoned on to the DM1’s limitations and were no longer interested. But the company persevered, and Radionics engineer John Nicholls came up with a superior version, the DM2. This was not only popular but, rarely for a Radionics product of that era, also able to be assembled easily and to operate reliably — a boon for both kit makers and Radionics’ in-house production facility, not to mention customer services and finance: fewer replacements demanded by unhappy customers. The DM2 sold well and profitably; Radionics’ instrument business grew and grew, always generating sales — it accounted for 20 per cent of Radionics’ revenue toward the end — and income for the business. No wonder Sinclair might have reasoned that if ever Radionics collapsed, instruments would be a good business to begin again with.
In any case, once Sinclair was certain Radionics would survive its woes, whatever plans he may have had for Sinclair Instruments would have seemed irrelevant, so we should consider Sinclair Instruments’ direction from August 1976 onwards as driven primarily by Curry’s own agenda. Sinclair was the owner, but Curry had a stake in the new business; he held just under 11 per cent at the time of his departure in the Spring on 1979, though within a year he had sold the shares back to Sinclair for reasons we shall discover in due course. Sinclair Instruments quickly focused on electronics products that could be sold in kit form to enthusiasts, a market that Sinclair had himself largely left behind.
Equipped with £500 borrowed from his father, Curry went in search of premises; he found and rented office space above a shop at 6 King’s Parade, Cambridge, opposite King’s College. Sinclair Instruments was initially a one-man band: “I used to do everything — placing the ads, packing kits, taking them to the post office, mending some of them, answering technical queries — absolutely everything. Later on there was a secretary, then more and more people were involved,” said Curry in 1982.
The office now set up, Curry set to work finding products to sell. One notion he had was to scour Cambridge’s University community for smart people with smart inventions but no desire — or expertise — to bring them to market themselves. Enthusiastic and with a quick smile and wide eyes that lent him a youthfulness despite his 30-odd years, Curry would have certainly found it easier to engage with students and contemporary electronics hobbyists than would his ‘egg head’ boss.
Inside Sinclair Instruments
The first three Sinclair Instruments projects, however, were classic Radionics offerings: a calculator, a digital watch and an amplifier chip, though only one of them ever went on sale. Two of them were found by nosing around the engineering departments of Radionics and Cambridge Consultants. The first product, the one that made it to market, was a clunky looking miniature calculator — “a ghastly thing”, Curry would later say — worn like a wristwatch and conceived by one of Radionics’ industrial designers, John Pemberton, who would shortly win a Design Council Award for Radionics’ slim Sovereign calculator. Dubbed the Wrist Calculator, it went on sale in January 1977 as a kit priced at £9.95 plus eight per cent VAT. The ads that Curry wrote — oddly naming the company ‘Sinclair Instrument’ — described the Wrist Calculator as “the ultimate in common-sense portable calculating power”. The device, the promotional material claimed, “offers the full range of arithmetic functions” and uses “normal algebraic logic (‘enter it as you write it’).” It had “a % key plus the convenience functions” — square roots, reciprocals and squares — and “a full five-function memory. All this from just ten keys!”
An attempt to sell some of the Wrist Calculators in the US proved a disaster, perhaps not surprisingly given how tricky the tiny devices were to put together and how unappealing the Wrist Calculator looked when it was finished, whether it worked or not. And many of them didn’t work, thanks to the cheap, bin-end, “minimal tolerance” electronic components Sinclair Instruments included in the kit, and the ordering of circuit boards from two suppliers, only one of which made a decent job of it. Yet Sinclair Instruments sold at least 5,000 kits, possibly as many as 10,000.
Clunky though it was, at least the Wrist Calculator made it to market. A futuristic digital watch designed by Cambridge Consultants’ Allen Boothroyd didn’t get that far because it was too slick. He offered Chris Curry an elegant digital watch design that look like a bracelet: it comprised a set of interconnected segments, some of which contained the watch’s workings — display, chips, batteries and controls. Each active segment required a flexible printed circuit board, but these were not cheap to make. Sinclair loved the unquestionably stylish design but everyone else pointed out that with the addition of physical controls and indicators to make it clear which parts did what, the watch was useless. Making those compromises killed the slickness of the design. Maybe too the Black Watch farrago was just too readily recalled by the principals. Whatever the reason, Boothroyd’s incredible design was never realised. The designer would eventually do better by Curry: he was later hired to design the casing for first the Acorn Atom and, shortly afterward, what became the BBC Micro. Right now, though, Boothroyd left Cambridge Consultants to focus what was now called Meridian Audio, a high-end hi-fi company he had founded a few years earlier with electronics and acoustics specialist Bob Stuart, and which is still in business today.
Sinclair’s input into the product is interesting. Even though he was working for Radionics, Sinclair was clearly involved in what was taking place in King’s Parade. John Pemberton, the Wrist Calculator’s designer, said in 1985 that Sinclair Instruments was “a hobby for Clive… Clive was operating it, and Chris Curry was dealing with the problems of it”. It’s not hard to imagine Sinclair, always the broad ideas, big picture man, providing the general direction and insisting upon the final say while Curry got on with the day-to-day business of finding products, realising them and then selling them by mail order. In other words, almost exactly the way Radionics was run in the 1960s. Yet Rodney Dale, who had known Sinclair since Radionics’ early days and by this time was working for him, insisted in 1985 that “while Clive Sinclair was so busy at The Mill, arguing, fighting, cajoling, wearing himself and everyone else to a shadow, he took surprisingly little interest in [Sinclair Instruments]… He trusted Chris Curry to develop the company”.
So Clive Sinclair was around and taking an interest in Curry’s activities, even if he wasn’t involved in the day-to-day operations. On 17 June 1977, Sinclair, now named on the paperwork as Sinclair Instruments’ Chairman, renamed the company Science of Cambridge Ltd. Again, the change was formally incorporated into the record a month later, on 5 July 1977. Sinclair would go on to change the name of the company two more times: first on 15 October 1980 — formally registered on 10 November 1980 — to Sinclair Computers Ltd, and then, on 2 January 1981, to Sinclair Research Ltd, the name by which the firm would be forever remembered.
In December 1977, Curry announced the SoC20, described in the company advertising as “the most powerful monolithic [integrated circuit] amplifier in the world” thanks to its “20W output [and] less than 0.2 per cent harmonic distortion at all powers”. This was a classic early Radionics product, though an odd choice for the mid to late 1970s, especially for a company whose modus operandi was selling kits rather than standalone components. Punters could buy one SoC20 for £4.95 or two for £7.95. If they did, stocks were soon depleted; Curry didn’t promote the SoC20 for much more than a month; he soon had a better product to offer.
Enter Ian Willamson
By the beginning of 1977, Ian Williamson knew as much about the inner workings and application of microprocessors as anyone in Britain. As a young electronics engineer working for technology R&D hot-house Cambridge Consultants, it was his job, like that of all his co-workers, to “think the unthinkable and then find someone who wanted the problem solved”. It was a mission he was accustomed to. As a member of Cambridge Consultants’ Digital Systems Group, he had already tried to interest various electronics and pinball machine companies in the groundbreaking idea of creating arcade video games based on microprocessor technology, but in the mid-1970s, when he and a colleague went pitching, no one in the UK was yet willing to take a chance on computerised slot-machines changing the way the nation’s youngsters got their kicks.
Microprocessor technology was simply too novel for established electronics firms to grasp fully its implications for their businesses. In a bid to help them, the British Government would, in December 1978, launch the Microprocessor Application Project, a £15 million scheme to encourages UK industry “to apply microprocessor techniques to a wide range of products and production processes”. For now, though, Williamson knew that though these companies’ senior managers had little appreciation of the potential of the microprocessor, their engineers eagerly wanted to understand what the new silicon chips could achieve. So did hundreds of electronics enthusiasts. To gain that understanding meant first learning how chips worked and were programmed. Williamson realised that here was a market he could exploit with a combination of a simple, low-cost microprocessor kit and a step-by-step guide explaining how the hardware operated and, through programs, could be made to perform almost any kind of electronic trick.
“What I could see was this incredible thirst for knowledge about microprocessors. Certainly anyone who was an electronics engineer felt threatened by the technology and wanted to learn about it. The UK also had a well established electronics hobby sector and there was incredible interest in what the technology could do. But it was very expensive technology to play with. And you couldn’t just buy a chip, you needed everything that went with it — memory, interfaces, software — to program and use it.
“Some US companies had started to produce training kits, and there were some professional evaluation kits around, but they cost hundreds of pounds, beyond the range of any hobbyist. So I figured it would be nice to put something together that was affordable. My objective was to break price barriers.”
In hardware terms, that was the very same notion that had prompted Bywood Electronics’ John Miller-Kirkpatrick to devise the Scrumpi microprocessor kit and to bring it to market the previous year. Bywood’s Scrumpi cost around £56, right in the middle of the £50-60 price range Williamson felt he needed to charge for his own offering. Williamson was aware not only of the development kits being offered by the likes of Intel and Motorola — respectively, the SDK and MEK series of products — but also undoubtedly of the amateur-oriented System 68 design promoted by Electronics Trade International magazine in the Spring of 1977. There’s no sign Williamson was a member of the Amateur Computer Club, so he probably hadn’t encountered the 77-68 at this point, but even if he had, he would have understood that, like the System 68, its owners would have to have shelled out several hundred pounds or more to complete the system, and certainly more than his own products’ anticipated price point to get the most basic workable parts up and running.
And, unlike Scrumpi, System 68 and the 77-68, Williamson’s product comprised more than just hardware. In the mid-1970s, Cambridge Consultants co-founder Tim Eiloart established Cambridge Learning Enterprises, a company which produced and sold multi-part, book-based self-instruction courses covering the new microprocessor and digital logic technology, including Design of Digital Systems, Digital Computer Logic and Electronics and The Algorithm Writer’s Guide. Williamson hadn’t written any of these himself — he would before the decade was out, contributing to Computer Programming in Basic and penning Microprocessors and Microelectronics: a Self-Instruction Course himself — but they provide a template for his kit’s second and then unique component: a microprocessor study guide.
Cambridge Learning Enterprises was part of the group of companies which grew out of Cambridge Consultants, Williamson’s employer. It’s worth having a brief look at Cambridge Consultants, not simply because of this connection, but because of its intimate links with Sinclair Radionics and, later, Sinclair Research.
Cambridge Consultants was founded in 1960 by two Cambridge graduates, Tim Eiloart and David Southward, to “put the brains of Cambridge University at disposal of the problems of British industry”, as the company put it at the time. Eiloart knew a third, older Cambridge University alumnus, Rodney Dale, who was then trying the establish a print and design business, and the two companies became closely connected. Dale eventually went to work for Cambridge Consultants
One of Cambridge Consultants’ first customers was Clive Sinclair, who through a meeting with Eiloart in 1962 outsourced the fulfilment side of his new business to the consultancy. Mailing out packs of electronics components wasn’t Cambridge Consultants’ core competency, but it was the least Eiloart could do for his fellow Mensa member. This arrangement continued until the mid-1960s when Radionics became large enough to warrant bringing this key function in house. Remember The Mill, Radionics’ later home? It was originally acquired by Cambridge Consultants as a home for some of its offshoot businesses, most notably AIM (originally Advanced Instrumentation Modules), the company which designed Radionics’ DM1 digital multimeter and developed some of Radionics’ hi-fi amplifiers too. Radionics took over The Mill in October 1970 when Cambridge Consultants ran into financial difficulty and was kept going with a £25,000 payment from one Clive Sinclair.
Richard Cutting, who was appointed Cambridge Consultants’ Managing Director in 1970, joined Sinclair Research in 1983. Nigel Searle, Sinclair Research’s Managing Director during its most successful period, was originally an academic mathematician in Edinburgh with an interest in artificial intelligence. He was brought down south by Tim Eiloart with the promise of a job. The role failed to appear, but he did manage to find work at Radionics — he steered the development of the company’s first programmable calculator and became its chief representative in North America for a time, a role he reprised for Sinclair Research in the ZX80 days.
Both Dale and Southward left Cambridge Consultants during the 1970s, but both eventually wound up working for Sinclair Radionics and, later, Sinclair Research. Dale collaborated with Ian Williamson on the aforementioned Basic programming course for Cambridge Learning Enterprises, on a 1980 book about the microprocessor, Myth of the Micro, and on a guide to the microprocessor kit Chris Curry was about to launch. The final thread in this web of connections: who did Clive Sinclair encourage in the early 1980s to write the official history of his life and business? Rodney Dale.
All of those publications were a very long way off when Williamson joined Cambridge Consultants in 1974. Cambridge Consultants had by then been bought by US-based global consultancy Arthur D Little — the acquisition took place in 1972 — but retained a formidable reputation for R&D consultancy built on hiring bright, capable and driven engineers and scientists straight out of Cambridge University. Or, like Williamson, those Cambridge graduates who had already left to start careers in industry. Williamson’s education was an extensive, five-year apprenticeship with GEC’s Marconi Space and Defence Systems in Stanmore, Middlesex. Williamson went in as a 17-year-old, fresh out of Manchester Grammar School. GEC-Marconi put him through university, where he studied electrical sciences at Christ’s College under influential information engineering scientist Peter Rayner. During vacations, he’d be straight back to work at Stanmore, and he returned there after graduation to develop digital control systems for satellite ground stations.
Cambridge gave Williamson the technical qualifications he needed to further his career, but it also exposed him — as university does for young, intelligent minds — to new ideas and new ways of looking at the world. In short, he found himself, and he discovered that he was uncomfortable with the defence focus of the work he had to do at GEC-Marconi. It was time to move on. He applied for other jobs, one of them with Cambridge Consultants. He turned down an offer from Sinclair Radionics. The R&D powerhouse was really where he wanted to be.
“I jumped to go to CCL in the Digital Systems Group… and worked on various projects involving digital system design and was therefore was involved in some of the first microprocessor applications. So I was in a fairly privileged position, and by 1975-76 I was fairly experienced in the using the very early chips,” he told me a few years ago.
Cambridge Consultants was no mere think tank. It was at the heart of what would become known in the 1980s as Silicon Fen, Britain’s answer to the United States’ Silicon Valley and home to many of the nation’s most dynamic technology start-ups. Thanks to the influence of Arthur D Little, it was by now a thoroughly commercial operation and it encouraged its engineers to think entrepreneurially. Workers at other companies could rely on being told what to do by their managers, but Cambridge Consultants’ smart young things were expected to come up with ideas and to go out and sell them to potential clients. An easy matter, perhaps, for a sales rep with a charming manner and a glib tongue, but to be able to pitch successfully, an engineer or scientist needs not only a high degree of technical competence but also a deep belief in the concept he or she is selling. That confidence inevitably encouraged some of Cambridge Consultants’ people to decide that the best way to take a concept forward was to take it to market themselves.
The interest in education and training, which would see fruition in his later Cambridge Learning Services guides, now shaped Williamson’s conception of what his low-cost kit could achieve and how it would tie in with the written materials he would produce. Williamson wasn’t setting out to make a cheap general-purpose computer but a package of book and hardware that would teach engineers and ordinary folk about the new microprocessor technology.
A microprocessor kit… but not the MK14
The first step: build a prototype. The National Semiconductor SC/MP was the obvious choice for a low-cost microprocessor kit since it was readily available and the cheapest viable 8-bit microprocessor on the market. It was unsophisticated and thus relatively easy to learn to program. Memory was easy to obtain too; 256 bytes would be sufficient for the product Williamson had in mind. But what about the display and the keyboard? He didn’t want to inflict data entry through toggle switches, as per the Scrumpi and 77-68, on his students, or leave them to decode binary values flashed up on a set of LEDs. Williamson reasoned that an ordinary calculator might prove a cheap source of both a keypad and an eight-digit LED display, so he bought himself a Sinclair Cambridge calculator for around a fiver, took it apart and examined the components. Hooking up the input and output pins on the Cambridge’s main chip, a Mostek MK50321, to an oscilloscope, he quickly worked out how the device scanned the keypad and how it presented numbers on the display. With that information, he wrote a 512-byte monitor program to allow users to enter their own program instructions using the calculator keypad, run them on the SC/MP one at a time — a technique called ‘single stepping’ — and see the results on the calculator display. For the tutorial he wrote a set of simple programs that users could type in, run and see immediately the effects the code made on the state of the machine.
There was a broad template for this: National Semiconductor itself offered a low-cost SC/MP evaluation board called the IntroKit for around the same £50-60 price as Miller-Kirkpatrick’s Scrumpi and the kit Williamson wasplanning. It was designed to be hooked up to a considerably more expensive teletype machine, a kind of electric typewriter able to transmit key strokes to a computer and to print on paper the output coming back from its host. However, as an alternative, and cheaper, method of input and display, the IntroKit could connect to the National Semiconductor Kit KB, which was a pocket calculator repurposed as both an input device and a screen. The Kit KB cost around £70. Unlike Ian Williamson’s design, which simply used the calculator as a source of parts, the Kit KB, also referred to as the KBDKit, used the entire calculator casing too.
Using the Cambridge calculator components put some limits on what Williamson’s machine could do. To be friendly to novice programmers, a computer would ideally provide what was called a ‘high-level language’, such as Basic, Algol, Cobol or Fortran. But that required a compiler or an interpreter to convert the language’s English-like syntax into numbers, the ‘machine code’ which the microprocessor was built to understand. Compilers and interpreters are complex software applications, and there wouldn’t be enough memory for those. In any case, the whole point of the product was to teach people about microprocessors so it was actually advantageous to work in machine code. Kit like Scrumpi and the 77-68 entered those numbers using binary notation: each of their toggle switches represented a single binary digital, up for 1, down for 0. A common alternative to both binary and decimal that was — and still is — favoured by programmers is ‘base 16’ hexadecimal, or ‘hex’, notation. All microprocessor documentation of the time, and for many, many years later, listed the chip’s set of instructions as the two-digit hex codes used to represent each of those instructions.
Unfortunately, the Cambridge calculator chip, which Williamson decided could be put to work monitoring the keypad and controlling the display electronics, could only output numbers. It could easily present any single digit from 0 through 9 but not the letters A to F. And of course the keypad lacked sufficient keys for those characters. That immediately ruled out being able to display hexadecimal numbers. So Williamson employed another notation: octal, or ‘base 8’, which can represent any value using only the digits 0 to 7. The keypad had those, and it left Williamson with two extra keys, ‘8’ and ‘9’, to use for processor control functions. Ordinary folk worked in decimal and some techies might know hex, but few people worked regularly with octal numbers. Entering and reading octal numbers might be a hassle, but Williamson decided that anyone who could only afford a £50 kit would be willing to put up with it.
Curiously, Amateur Computer Club member M A Baker proposed using octal with a calculator keyboard as a method of data entry more than a year earlier, in 1976. And Jonathan Titus’ early American microcomputer kit, the Mark-8, had also been designed to be programmed using octal notation. So Williamson was perhaps bending the mould rather than breaking it.
“I figured beggars can’t be choosers. You could buy a £300 kit from the US, or a £50 kit in the UK and you would still learn the fundamentals of microprocessors even if you were using octal. The calculator chip was driving the digits. I interrupted the IO of the calculator chip so when the Scamp wanted to display something it just sent eight digits to the calculator — I just drove an eight digit value into the register. It was a classic bit of down and dirty electronics and a very, very cheap way of giving a reliable IO, but it meant that your machine code had to be expressed with numbers – I didn’t have the option of A, B, C, D, E and F as a display output.”
Williamson built his prototype on a general-purpose circuit board. It had the kit’s key components: the SC/MP processor, a RAM chip containing 256 bytes of memory for temporary data and program storage, and a 512-byte ROM to hold Williamson’s monitor software. They were all connected together by wiring rather than the fine, etched metal lines of a finished product’s printed circuit board (PCB), but it was just a prototype. Hanging off the board was the red Sinclair Cambridge Memory calculator with the back missing, wires running from the Scamp to the Cambridge’s MK-50321 chip, now used solely to handle display output and keypad input. So, £10 for the processor, a fiver for the calculator and perhaps three quid for the other components: a fully functioning, if rudimentary, computer for under £20. And Williamson planned to sell it for more than twice that. Nice profit.
And then Williamson’s career took a sudden turn. Between the time Williamson started working on his microprocessor kit, in the Spring of 1977, and July that year, he applied for and was offered a job at Leyland Vehicles, nationalised car firm British Leyland’s buses, trucks and tractors division. The position was based in Preston, Lancashire and he would become the firm’s Principal Electronics Engineer, a promotion. Three years previously he had quit GEC-Marconi to escape what was for him the morally questionable world of defence work, and even in Cambridge he found himself frequently being dragged back into work with a military angle. Accepting the job offer from Leyland Vehicles would allow him to do turn his back on such work, but it was nonetheless a considerable career change involving a move far away from his friends and colleagues in Cambridge and from the buzz of the being at heart of Britain’s technology sector. He weighed up the role’s opportunity and cost, and decided to do it — even though he would now no longer be able to pursue his microprocessor kit. Yet he was unwilling to shelve the opportunity. He could still make it pay, he reasoned, if he could sell the device and its documentation to a third-party with the resources to offer the kit as a commercial product.
Early on in the development process, Williamson had approached his bosses at Cambridge Consultants to see if they would like to commercialise the kit. They turned him down. Now, while he worked out his notice period, he approached two other companies in the Cambridge area who he felt might have the imagination and expertise to bring his product to market. One was a maker of electronic instrumentation and simple kits for enthusiasts, but it proved to be no more receptive to Williamson’s ideas than Cambridge Consultants. The other company was Sinclair Radionics.
It was an obvious choice. Sinclair Radionics had, after all, made the very calculator which Williamson had used for parts. Then there were those close ties between Cambridge Consultants and Radionics. And Williamson had an entrée: when he had been looking for a way out for GEC-Marconi, he had been offered a job at Radionics, though he decided not to take it. The chance to become an engineer at the company even then considered to be at the forefront of the British consumer electronics industry had been made by none other than company founder. Would he remember the rejection? Not one to be intimidated by the irascible entrepreneur, Williamson steeled himself and called Clive Sinclair.
Williamson’s microprocessor kit might hark back Radionics’ hobbyist roots, of course, but it wasn’t the kind of advanced consumer electronics product Sinclair believed the company should be offering in the late 1970s. But Sinclair immediately thought it sounded like the kind of thing that Science of Cambridge might choose to take up, so he offered to put Williamson in touch with Chris Curry.
Three men in a pub
The three men met at The Fox in Bar Hill, the small township outside of Cambridge where Cambridge Consultants was now based. If Sinclair was only interested in products that might be used to provide funding for his favoured projects, or had his mind on other, more pressing matters, at least Curry was keen to hear what the young engineer was proposing. Not that Williamson’s discussion of the potential market and a demonstration of his prototype prompted Curry to sign up there and then. “It was a very non-committal meeting, and then Chris turned up a week to ten days later at Cambridge Consultants with a cardboard box full of broken calculators and asked if I could use any of them.” Curry wanted him to see if his design could be simplified — for which read ‘made cheaper’ — with different components.
Curry was taken with Williamson’s concept. He too had noted the growing interest in microprocessors and microcomputing, and the more he thought about the kit, the more he felt it was the right product for Science of Cambridge. “I had been very interested in the computer market, buying the US magazines and seeing what was happening,” Curry said in 1981. “I actually tried to negotiate an import agreement with an American company which had what they called a ‘computer in a book’.” Curry never struck a deal with the unnamed American firm, but he soon had Williamson’s kit instead. He appreciated not only Williamson’s design but also the young engineer’s talk of the potential market among electronics buffs for a kit priced well below the competition — a classic Radionics and, later, Science of Cambridge tactic. It’s impossible to imagine someone in Curry’s situation in the middle of 1977 not being aware, through the electronics press, of home-grown kits like Bywood Electronics’ Scrumpi, imports from the States of MOS Technologies’ KIM-1 and RCA’s Cosmac Microtutor, not to mention the recent hints being dropped by component sellers Sintel and Lynx about the microcomputers they were developing.
And so the deal was on. Williamson was offered £5,000 for the kit up front, with royalties to follow when the product went on sale. Williamson readily agreed to the arrangement. A contract was drawn up and sent to him. He signed on the dotted line and posted it back to Curry at Kings Parade. A copy of the contract, now with Clive Sinclair’s signature, would surely be sent back to him in due course. It was the start of September, and Williamson was preparing to travel up to Preston and begin his new job with Leyland Vehicles. His mind was on more important matters than electronics kits and contracts, so it didn’t register immediately that the signed contract was taking longer to arrive than he had been given to expect.
NatSemi muscles in
After some weeks, Williamson finally received a call from Curry. He listened politely while the Science of Cambridge chief asked if he would be willing to redesign his kit to use only components made by National Semiconductor, the American chip maker which had designed the low-cost SC/MP processor Williamson had selected for his project. As an engineer, he knew that Curry’s requested changes were entirely feasible — they would just take time to implement, especially given the use of an alternative chip for controlling the display. But it was disheartening to be asked to dispense with the design he had spent so much time devising. Curry also said he wanted to put the kit on the market for much less than £50.
“I was reluctant to do it because you’d have to start from scratch in terms of the design, and it wasn’t clear how they were going to get a functional keyboard at that price,” says Williamson. He put that very question to Curry, who answered that the kit would would use a low-cost membrane keyboard. This comprised two flat wire circuits placed on separate thin plastic sheets kept apart by a grid-like spacer. Press a button and the wires briefly made contact and allowed a current to flow that could be read by the processor. Cover this electronic sandwich with a layer of printed buttons and you had your membrane keyboard. Anyone who has used any of Sinclair’s later home micros will know what it was like to type on. The design’s sole virtue was that it was cheap. Williamson promised Curry he would think about it.
Before he could answer, Curry called again. This time clearly very embarrassed, Curry now informed the engineer that the deal was off. Science of Cambridge would not be using Williamson’s kit, either in its original form or the revision they had recently discussed. Williamson says Curry told him that Sinclair had vetoed payment for the kit because it was no longer needed: National Semiconductor had offered to do all the work for free.
When Curry approached National Semiconductor to negotiate a purchase order for the SC/MP microprocessor that Williamson had placed at the heart of his kit design, the American chip company quickly spotted an opportunity: sell him all the other components he needed too. Curry would have to buy all these extra parts anyway, but working with a single supplier would simplify the purchase process and potentially make the overall deal less expensive. Hence his call to Williamson inquiring about the feasibility of redesigning the kit. The conversations between Curry and National Semiconductor would ultimately have come around to the nature of the product he intended to offer, and on learning that it was a microprocessor learning kit, National Semiconductor’s sales team offered him one of its own, as a sweetener. It might also have been miffed at the similarity, if only externally, between Williamson’s ‘Scamp plus calculator’ design and its own IntroKit and Kit KB combo. In addition to the IntroKit, National Semiconductor also offered the LCDS (Low-Cost Development System). It was a large black box costing £335 onto which could be slotted a number of circuit boards, one for the SC/MP chip itself and others for memory and input/output connections, but it was also fitted with a cheap membrane keyboard almost identical to the one Curry had told Williamson he wanted to use. From all this hardware, Curry could take a circuit layout, a list of required chips and a copy of the monitor software — the all-important application that allowed users to enter and run their own programs — all for free. The chip maker just wanted Curry’s commitment to buy its latest SC/MP microprocessor and associated chips.
The National Semiconductor kit also had built-in expansion possibilities, an opportunity Curry would later make use of to sell buyers add-on products, including extra memory and chip ‘blowers’ — equipment used to fix software into read-only memory ROM chips — after the board itself had gone on sale. So did Williamson’s prototype. The young engineer had already lightly sketched out an interface system that would allow his kit to be connected to a TV screen as a display. He also planned to add both a Basic language interpreter to allow user to program the kit. It would be loaded in from cassette. He even had a notion that this expanded system might be operated through a kind of on-screen keyboard with a lightpen — a stylus with the ability to measure the intensity of light at any point on the screen.
“Before you know it, you’ve got a full computer. There was in my mind a roadmap that lead down that route, but I effectively stopped working on it in the middle of 1977.” The Leyland job offer had put paid to any plans he might have had to expand his kit into a complete home computer system, and his ideas for expansion existed more as a plan for the future than as actual circuit designs.
Curry would dismiss Williamson’s original work a few years later, saying: “We nearly went ahead with an inexpensive home computer which would have been based on the use of a calculator chip for keyboard and display. Eventually, we went away from the use of a calculator and used more conventional interfaces to provide a display and keyboard — and produced the MK14.”
Enter Sinclair’s first micro: the MK14
Yes, the kit now had a name: MK14 — ‘MK’ for ‘Microprocessor Kit’ and so pronounced ‘em-kay’ not ‘mark’. It wasn’t a straightforward reproduction of either of the existing National Semiconductor SC/MP evaluation systems and it would use the SC/MP’s cheaper successor, the SC/MP II. Always mindful of how much the consumer might have to pay, Curry had the National Semiconductor board ‘cost engineered’ down: the design was tweaked to make use of fewer, cheaper parts. This process was a well-established one from the early days of Sinclair Radionics. National Semiconductor Field Applications Engineer Tony Amendt liaised with Curry on the conversion of the chip maker’s designs into a physical product. The pair later contributed a description of the machine to the journal Microelectronics and Reliability.
What an ironic choice of journal that was. The MK14’s keyboard was certainly anything but reliable, and individual units often failed. “The first [MK14] weakness, the keyboard, is more or less acknowledged by the manufacturers in as much as they provide an edge connector on the printed circuit board for an external keyboard,” wrote Personal Computer World contributor Clifford Clark when reviewing the product. “The second shortcoming is the manual, which jumps from simple memory switching to sample programs laid out in assembler format. There are statements about outputting to the display with no instructions on how to do it.”
Don’t blame Ian Williamson for that: the MK14 shipped with someone else’s documentation. While Williamson was later paid £2,000 to write a full programming manual for the MK14, perhaps to salve Curry’s conscience, Science of Cambridge ultimately never used it, despite frequently telling reviewers and reporters frustrated with the booklet supplied with the kit that an expanded manual was in the pipeline. Curry mentioned this first in the Autumn of 1978, but even by the following Spring, reviewers were still stating that “apparently a larger programming manual is in preparation”. The book, Understanding Microprocessors with the MK14, was eventually co-written with Rodney Dale, Williamson’s fellow Cambridge Learning Enterprises contributor. The book eventually came out in 1980, published by Macmillan.
There is no byline printed on the more simple guide that shipped with the MK14 itself, the manual Clifford Clark and many other reviewers criticised, but it was written by Tony Amendt, Chris Curry’s liaison at National Semiconductor. David Johnson-Davies, one of the growing number of electronics and computing buffs, Cambridge University students all, who had by now become regular visitors to Science of Cambridge’s Kings Parade office, wrote the manual’s sample programs. “Chris Curry got my name through the Cambridge University Processor Group, a club aimed at people interested in building their own computers. I demonstrated a 6800-based Motorola evaluation kit, the D2, that I had assembled and programmed to play Bulls and Cows, and on the strength of that he asked me to develop some programs for the MK14.” Many of Curry’s student circle, Johnson-Davies among them, would go on to form the core of a new microcomputer company, Acorn, as we’ll shortly see. Another was Steve Furber, a maths graduate by then trying to pursue a doctorate in aeronautical engineering but increasingly find himself distracted by a new-found love of microcomputer technology.
Furber build the very first MK14 prototype using the National Semiconductor schematics and chips: “I got the circuit diagram and built one using Verowire and solder in my front room. The MK14 was basically a copy of a Nat Semi Scamp development kit, and they’d taken a mass-programmed ROM and copied into two linked PROMs on the MK14. Only they managed to copy it wrong, and I debugged this thing in my front room. That was the first piece of work I did for them.”
With bugs such as these squashed and the circuit diagram reproduced on a printed circuit board ready to accept and connect the kit’s 30 resistors, capacitors, chips and a 4.43MHz clock crystal to tap out the tiny micro’s heartbeat, the MK14 finally went on sale in February 1978, though it took three months more for Science of Cambridge to begin pushing the kit out in appreciable numbers. It sold for £39.95 plus £3.20 sales tax, and 40 pence postage and packing. Once assembled, it could “handle dozens of user-written programs through the hexadecimal keyboard” and was “the only low-cost keyboard-addressable microprocessor!” Assembly could be carried out by “anyone with a fine-tip soldering iron and a few hours’ spare time, using the illustrated step-by-step instructions provided”, the Chris Curry-penned adverts claimed.
The user’s programs were stored in 256 bytes of RAM, coincidentally the same amount to be found in the Scrumpi and in Ian Williamson’s kit. The MK14’s monitor software, providing a user interface for entering programs in machine code, was placed in the two Programmable read-only memory (PROM) chips Steve Furber referred to above: 512 bytes in total, 256 bytes per chip. An eight-digit segment LED display mounted just above the keypad presented the results. There was space on the board to add a further 256 bytes of RAM, which Science of Cambridge soon offered as an optional extra. The chips weren’t soldered on but pushed into holders which were themselves soldered to the board. Finally, the board had two edge connectors to clip on peripherals or to tap the MK14’s data.
“The MK14, produced by Science of Cambridge, makes up the cheapest complete home computer I have yet seen. The cost is less than that of the [integrated circuits] when bought individually,” wrote an enthusiastic Nick Toop in Personal Computer World. “The basic kit enables one to enter programs written in machine code through a hexadecimal keyboard and to show the results on a calculator type display. There are also several channels for control applications. Toys, robots and laboratory applications would find the small size and power consumption an attractive possibility. An interesting domestic application would be the control of heating and lighting since the large number of inputs make it possible to act on the outside lighting, the room temperatures, the number of people in each room, etc. One could make sophisticated fire alarms or security systems which operate various emergency procedures including phone calls.”
Toop concluded: “As an independent consumer, I was very impressed with this kit. Its low cost and versatile but economic design are very pleasing. I am sure that it will have great success.”
Toop, then an astrophysics post-grad student at Cambridge, would later become one of Chris Curry’s employees at Acorn, and work on the company’s first home micro, the Atom. Later still, and by then a freelance, he created the custom graphics chip used by the Enterprise micro in 1983.
Other reviewers were keen on the MK14 too — despite universally loathed keyboard and the much-criticised manual. “The MK14 is not a toy and, with the low-cost add-ons planned by Science of Cambridge, it should prove a powerful tool for those wanting a versatile MPU development system at under £80… It is the cheapest development kit that we know of,” wrote one.
“The expansions described will produce not only a reasonably sophisticated personal computer, but also a control development system of some power,” wrote another. “In conclusion, the MK14 is a well though out product nicely packaged, and easy to use with the revised monitor [software]. Indeed my firm opinion is that no electronics enthusiast or engineer should be without one in today’s technology.”
The MK14 “became a great favourite”, Curry said in 1981. “I think we sold about 1,500 of them.” Other estimates put the total closer to 10,000. Whatever the figure, the MK14 “was quite a successful computer”. That success could be attributed to positive write-ups in the press, to the very low sale price and to Curry’s professional-looking ads which stood out a mile from the more prosaic advertisements offering either rival microprocessor kits or the new, full microcomputers that had just begun to appear, such as the Nascom 1. To the potential buyer, Curry’s ads made the MK14 look like a solid product. And if the keypad had been cost-reduced to the point of at which it would break down with anything but the most tender of usage, the rest of the system held up well.
Not so Science of Cambridge’s ability to fulfil orders. As would be the case with many later Sinclair products, the MK14 suffered from long shipment delays. Curry requested an initial 2,000 sets of chips from National Semiconductor for the first batch of MK14 kits. But so many eager buyers sent in orders that Science of Cambridge was unable to fulfil them all until more components came through. That took several more months. Curry never quite got the numbers right and shipment delays plagued the product throughout the remainder of 1978 and 1979.
As late as February 1979, a year after the MK14’s introduction, Personal Computer World still felt the need to warn potential buyers about the long wait they might experience before their computer arrived: “We advise readers who are interested in obtaining MK14s to first write and obtain a firm undertaking on a delivery date for this outstandingly good value-for-money kit.”
“I received one in January  which was ordered the previous summer,” complained one reviewer. Practical Computing’s May 1979 number catalogued a number of grumbles from “about a dozen readers who are less than enthusiastic about MK14 or, more accurately, about Science of Cambridge”. Their beef? Long waits for delivery, kits with faulty components, excessive delays for return of replacement kits, and difficulties getting hold of Science of Cambridge staff to complain to. Naturally the magazine asked the company to respond.
“We do not deny that we have been beset by component supply problems and we apologise to those people who have been kept waiting,” a spokesman — Chris Curry, one assumes — admitted. “We have had very few component failures and any repairs which exhibit symptoms of original component failure are repaired free of charge. A recent batch of a thousand [MK14]s which we had to supply ready-built contained only two which did not work first time, and they were due to IC pins being bent over on insertion.
“It is extremely difficult to provide a fault-finding guide, as almost any component failure or solder bridge gives rise to the same fault indication on the display. We provide a considerable amount of fault-finding guidance over the telephone which can be very time-consuming hence the difficulty in reaching us.”
Problems aside, the MK14 was evolving. By early 1979, its board layout had been revised at least three times, and the monitor software given a significant update. At this point, Science of Cambridge began supplying the kits with the correct, 4MHz crystal, the clock whose beats mark time for a microprocessor, in place of the original 4.434MHz version which, being slightly faster than the specification called for, had led to unexpected errors when add-ons were connected. And by now the computer kit had been joined on Science of Cambridge’s full-page ads by a £33.75 VDU module capable of displaying 32 lines of 16 characters on a domestic TV set; a £7.25 cassette interface which let the MK14 user load and save programs on audio tape; and an £11.85 PROM programming board for saving programs on chips rather than cassettes. In addition to the extra 256 bytes of RAM that could be slotted onto the MK14 (provided you sent £4.14 to Science of Cambridge), an £8.97 RAM IO chip provided a further 128 bytes and allowed memory boards of your own devising to be wired up to the kit. As Ian Williamson had once envisaged his proposed kit might eventually become, the MK14 was gaining the parts a user would need to turn it into a machine more akin to what we today think of as a computer.
Science of Cambridge was changing too. Not long after the MK14 went on sale, Curry began to receive inquiries from companies asking for his help applying microprocessor technology to the products they made. This interest — and the commercial possibilities it suggest, encouraged him to establish a consultancy with Hermann Hauser, an Austrian physicist doing research work at Cambridge University, but who had a dilettante’s interest in computers and electronics, and a yearning to go into business for himself. Hauser had identified the Cambridge University Processor Group (CUPG), the local society for computing enthusiasts, as a source of the expertise he knew he would need if he established some kind of microelectronics business. Hauser was soon a regular visitor at the Science of Cambridge office in Kings Parade, as were half a dozen of the CUPG’s other enthusiasts, among them Steve Furber and David Johnson-Davies.
Curry himself was soon keen to move beyond the MK14 to offer a more versatile product. Calling on the brains of the King’s Parade hangers-on, Curry and Hauser eventually launched their own microcomputer system, though this one was not aimed exclusively at hobbyists but also at industrial applications. It was introduced in February 1979, a year after the MK14. A conflict of interest? Clive Sinclair couldn’t see that the Acorn Microcomputer was anything other than a direct challenge. Perhaps he might have felt less threatened if he himself hadn’t become so angry with the deteriorating situation at Sinclair Radionics.
The company had limped on from the days when Curry dropped out to set up Sinclair Instrument. The details of what took place at Radionics between 1977 and 1979 will have to wait for a future post, but by April or May 1979, Sinclair was attempting to devise one more strategy to ensure the company’s survival. He failed to persuade his NEB bosses that it could be made to work. In June 1979, the end-game began to be played out as the NEB began the dismemberment of the company. The first of 300 staff were made redundant. The following month, Sinclair was out too. He accepted a £10,000 pay-off, sold his Rolls Royce and went straight round to the King’s Parade office of Science of Cambridge, which he’d been running since Curry left a month or so before. He brought with him the rights to the flat TV tube technology and three engineers who had been working on it at Radionics: his old stalwart, Jim Westwood, who was accompanied by Brian Flint and Peter Maydew. Sinclair’s secretary came too, and they all joined the various administrative assistants hired by Curry.
“Chris Curry had moved out by the time we got to King’s Parade,” says Flint. “The MK14 microcomputer kit was still going, so one of my jobs was servicing and fixing MK14s that had been sent back.”
And what of Ian Williamson? Chris Curry kept in touch while he was still at Science of Cambridge, to sound the engineer out about developing tutorials for the upcoming Acorn Microcomputer. Apart from co-writing those programming guides and other books with Rodney Dale, Williamson’s participation in the emerging UK microcomputing scene ended in September 1977 with Clive Sinclair’s decision to reject his kit in favour of the free one National Semiconductor was offering. Williamson’s day job put him in charge of Leyland Truck and Bus’ electronics R&D efforts. “After a few years running the technology activities, I got caught up in Chief Executive Michael Edwardes’ psychological assessment programme and they decided I would be a better manager than an engineer. So at the age of 31 or so I was moved to run a loss-making gearbox company in Coventry.”
And the rest is history…
Established in 1961, a going concern in 1962 and hugely profitable in 1975, by 1980 Sinclair Radionics was no more. Science of Cambridge’s MK14 was a relic too. “The alarming rate at which MK14 microcomputers are appearing on the classified ads pages of the various computer magazines is a useful indicator of the shortcomings of this very popular device,” noted an article on expansion opportunities for the kit, published in March that year. Though Science of Cambridge continued to offer and advertise MK14 add-ons, the product’s days were numbered. In February 1980, Clive Sinclair announced its successor, and by November it had vanished from computer magazines’ listings of board computers. Soon the Science of Cambridge name would be gone too, replaced first by Sinclair Computers and soon after by the more impressive sounding and soon much better known Sinclair Research.