Clive Sinclair, the home electronics pioneer and entrepreneur, has just died at the age of eighty-one. Although I am not officially a fan of entrepreneurs, being rather left wing, I nevertheless have a soft spot in my heart for his products. The drama-doc ‘Micro Men’ covered his story and rivalry with Acorn in the late ’70s and early ’80s quite nicely, but there’s more to Sinclair and his two major companies than the events and products of that period. Incidentally, the two of us share a birthday.
He was born in 1940, in Richmond, and founded Sinclair Radionics five days before his twenty-first birthday, having raised funds by writing magazine electronics articles. This first company was bought out by the National Enterprise Board and he was paid off in the late ’70s, and he proceeded to start the company which actually made the famous computers and C5, Sinclair Research Ltd. This was later taken over by Amstrad, but he continued with another new company, releasing the Z88 in 1987, and a number of other products. I would say his products are characterised in three ways: they were cheaper than their rivals, they tended to get announced way earlier than they were released and they often had teething problems due to their relatively short development phases.
As far as I can remember, Sinclair’s first product was an amplifier in 1962, followed by a pocket radio the year after. This second product was self-assembly, as were several of his products up until the ZX Spectrum two decades later. This presumably made them cheaper, but this wasn’t unusual at the time. By 1966, he’d designed a pocket television, 405-line if I remember correctly, whose design was unfortunately too complex to get it beyond the prototype stage, but I’d say this is still an achievement for that time:
This was at a time when pocket trannies were quite a novelty, although I’m aware of much older DIY projects to build crystal sets to fit in wallets so as a concept they weren’t actually very new. This design looks very ‘sixties. I can imagine it turning up on ‘Star Trek’ TOS. Although this was advertised but didn’t happen, establishing a familiar pattern, the Microvision did eventually come to market in 1977, and I remember it doing so clearly. There was a display model in the window of Barrett’s toy shop in Canterbury for ages. Like its predecessor, it had a two inch screen and I remember it being advertised as being able to pick up TV transmissions from all over the world, which I found dubious at the time and imagine was either untrue or only true in the sense that if you took it to a country with the same TV system as ours, it would also pick up programmes there.
Looking at those controls, which I remember from the time, it looks like it could switch between PAL and NSTC (the American system) and possibly also 405-line transmissions, so the claim seems misleading but technically true. One of my friends hooked this up to another of Sinclair’s products, the ZX81, and was able to display what amounted to pretty high resolution graphics on it, in the sense of pixels per inch, but I’m getting ahead of myself.
In 1972, they introduced the first slimline pocket calculator, the Executive for £79.95 plus VAT. This tendency to quote prices without VAT was really irritating and seems to have stopped happening nowadays. Amstrad also did it. The use of LEDs on calculators and digital watches at the time made them quite power hungry compared to LCD displays which came in later. Sinclair was for some reason very hostile to the idea of LCDs for a very long time, not including them in any of his products until the late ’80s. You can see from this device the beginning of a tendency to have rather uncomfortable and impractical keyboards, which continued with his computers in the next decade.
Nineteen models of calculator were produced by Sinclair over the years. The once I remember best is the above, Sinclair Cambridge Programmable, which was advertised in the ‘New Scientist’ in 1976. It had a maximum of three dozen steps in its programs and included a conditional branch instruction. A later model extended this to eighty steps, but both were only accurate to four significant figures. One of the oddities of ’70s programmable calculators is that they didn’t lead smoothly into microcomputers. You might think that the design of these devices would become steadily more advanced until they actually became like home micros in a way, but instead, and this is across the board with the exception of Hewlett-Packard, microcomputer design starts again from scratch. The one exception is the HP85, but this didn’t lead to anything else in the long run. In the case of Sinclair this may have been because he lost the intellectual property rights to his calculators when the NEB took over his company, but I’m just guessing.
Another Sinclair innovation was the Black Watch, a self-assembly LED digital watch. A couple of interesting things about this design are that the LED has the same blue colour (red when illuminated of course, as red LEDs preceded the other colours) as the calculator above, and the black colour, shared with Sinclair’s first calculator and to be repeated on its computers later. The Black Watch was not successful because it didn’t keep good time due to the quartz crystal running at different speeds at different temperatures and the batteries only lasting ten days as opposed to the advertised length of a year, and they were then difficult to replace. Also the circuitry was vulnerable to damage by static electricity. Therefore a lot of the features of later products were already discernible, if you allow the word “feature” to include the ideas expressed
A lot of kits were left over after the watch was withdrawn and in order to use these up they were remarketed as a clock for car dashboards. I actually admire the ingenuity of doing this and the economy of using components which were just lying around appeals to me. In practical terms though, I wonder whether the technical problems were resolved or if they didn’t affect a dashboard clock as much as they would a wristwatch.
If you were to ask most people to name Sinclair computers, the first one to come to mind would probably be the ZX Spectrum, followed by the ZX81, probably with the QL and ZX80 sharing a fairly distant third place. However, the ZX80 wasn’t Sinclair’s first computer. That honour goes to the 1977 product, the Mk 14. Now I have to say that I find most mid-’70s hobbyist microcomputers rather confounding, and if you wanted an Acorn equivalent to a Mk 14 it would probably be the System 1. Other similar computers include Commodore’s KIM-1 and the MPF-I, which I imagine ceased to exist when Apple decided to sue the heck out of the company which made it when it moved on to the rather Apple ][ -like MPF-II. Anyway, this is a Mk 14:
These were not user-friendly machines by any stretch of the imagination. Moreover, unlike the System-1 and KIM-1, which both used the 6502 CPU, for some reason Sinclair opted to use the “Scamp”, also known as the 8060. I have never understood why the 8060 is designed as weirdly as it is. It sounds like it’s supposed to be an Intel CPU like the x86 series or 8080 and 8085, and maybe that was a marketing ploy, but it was made by National Semiconductor. Although it can access a 64K address space, it does so by changing the function of several of its pins which are quite important in what they’re doing already, and it increments the program counter before fetching instructions, meaning that address 0 cannot contain an op-code unless it branches back to it. I also seem to remember it didn’t have a stack pointer, so subroutines would be very difficult to implement. It was used as a lift (elevator) controller and some of them are probably still in use, and that’s fine, but it doesn’t seem to lend itself very comfortably to writing general purpose programs, which is what Sinclair was using it for in the Mk 14. It was also, probably due to its unpopularity, much more expensive than the mass-market Z80, 6502 and 6809. It seems perverse to have such a fiddly piece of hardware in the first place then be made even less user-friendly by employing the 8060.
The Mk 14 cost only £39.95, had half a kilobyte of ROM and 256 bytes of RAM. It would presumably have had a machine code monitor as firmware and be programmable using hex opcodes using the LED seven-segment display for output. Remarkably, in the light of future developments, Clive Sinclair wasn’t keen on the idea of bothering with computers at all, and didn’t actually even use them himself well into the 1980s, and this seems to have been a factor in the genesis of Acorn. Chris Curry managed the Mk 14 project and soon went on to found Acorn with Herman Hauser in 1978 to build the System 1 and eventually design the ARM chips which now power tablets and mobile ‘phones, and I’m wondering if this was due to Sinclair’s failure to appreciate the potential of computers. It’s quite strange to think of this now.
One thing the Mk 14 did manage to do was persuade Sinclair that there was a market for home computers, and he went on to design and release the ZX80, in 1980:
I like the design of the ZX80 case, as in 1980 terms it looks very futuristic. It has the whiteness of the Cambridge calculators and of course a flat panel keyboard, which was very en vogue at the time in the form of hi-fi and music centre controls. It was also a nightmare to type on. It could be bought either complete or as a kit, and in the former condition it was the first computer ever to be sold for under £100. People tend to think of it as very primitive. My mother considered a friend of mine to be uppity because his parents bought him one when it came out. It had 1K of RAM and 4K of ROM, which included a BASIC interpreter which could only work in integers. Also, it was a bit like President Ford who couldn’t walk and chew gum at the same time, because it could either do computing or show things on the TV but not both, so when it was actually running a program the picture would disappear. However, it was also very fast because of the integer BASIC and it was able to use full keywords when the later ZX81 had to abbreviate, such as “CONTINUE” and “GO SUB” instead of “CONT” and “GOSUB”. This leaves one with the impression that the ROM is quite spacious when in fact it only had about as much information in it as two sides of handwritten foolscap. It was made entirely from readily-available parts rather than commissioned or in-house chips, as was usual at the time, and apart from the CPU, which thank goodness was a Z80 rather than the silly 8060 used in its predecessor, RAM and ROM, was composed mainly of integrated circuits in the form of discrete logic, which was again pretty standard for micros of that time. It sold about 50 000 units. It is possible to get it to produce a steady display through using interrupts carefully, but it wouldn’t do it out of the box. My perception of it at the time was that it was still very much a niche product about which I knew practically nothing. In the publicity, SInclair claimed it was powerful enough to run a nuclear power station but I’m unaware of any supporting evidence for that.
Sinclair would dedicate the next few years to mainly producing exclusively home computers and peripherals, and in 1981 they started to produce the computer which I think is still the cheapest new computer on first introduction ever: the ZX81at £69.95. In hardware terms, the unexpanded ZX81 is functionally equivalent to a ZX80, but internally it made the major innovation of putting all the logic on the integrate circuits into a single chip, resulting in the board only using a total of five chips compared to the ZX80’s seventy-eight. There was now 8K of ROM, including almost a full floating point BASIC lacking READ, DATA and RESTORE, and the RAM was expandable to 56K. The compromises in the BASIC were due to the inclusion of a number of instructions for interfacing with the new ZX Printer, a thermal printer which required special metal coated paper. I actually consider this an unfortunate decision which was probably connected to marketing the printer. The display was steady but this was achieved by getting the computer to multitask between running programs and displaying the screen, which made it four times slower running the same code than the ZX80, and it was further slowed if a RAMPack was used because this meant dynamic RAM rather than static, which is slower. I’m guessing that the initial decision to use a Z80 CPU was made with an eye to such a later expansion as it has its own built-in RAM refresh facility which can double as a kind of quick random number generator. This machine was probably responsible for the microcomputer boom of the ’80s. My perception of it is rather dominated by the fact that it was our first home computer. It was also frustratingly limited even at the time, but this spurred third party developers to come up with their own expansions for the likes of colour, high resolution graphics and sound, some of which went on to produce their own computers.
1982 brought the legendary ZX Spectrum:
This was once again a huge leap forward. It was at the time the cheapest computer with sound, high-res graphics and colour. Although it once again used the Z80A CPU, it shares many of the features of the Apple ][ while improving on them, and for this reason I’ve written about it as part of an alternate history here. It is entirely feasible that a functional equivalent to the ZX Spectrum could’ve been put together from April 1976 onward, because in the following year a rather similar, though 6502-based, computer arrived on the scene. However, the chances are the world wasn’t ready for it in the mid-’70s and at that time it would probably have cost about £1500. It’s all discussed on the link. Sinclair wanted his machine to be considered as the BBC micro, but the BBC wanted a “real” computer so they chose the Acorn Proton project instead. Clive Sinclair objected to the idea of the BBC endorsing a specific model of computer because they were a publicly-funded body and he saw it as similar to advertising. I wonder if in fact he was influenced by his experiences with the NEB, which seems to have taken advantage of all his hard work and given him an insufficient golden handshake while apparently denying him the opportunity to capitalise on his ideas.
The ZX Spectrum was not intended to be a games machine but that was certainly its main use. The same applies to a lesser extent to the BBC Micro. With the Commodore 64 it was the most popular computer before IBM PC clones came to dominate everything. The original keyboard was not much of an improvement on the previous two products, and single-keyword entry, intended to circumvent the problem of having a poor quality keyboard, led to such absurdities as it taking four keystrokes to type the word “INK”. Sinclair promised a stringy floppy called the Microdrive, and Interfaces 1 and 2, all of which were both delayed and led to a long waiting list. The Spectrum persisted for a very long time, undergoing several upgrades and continued to be manufactured for some time after Sinclair became part of Amstrad, by which time it had a proper multichannel sound chip, RGB monitor interface, built-in disc drive and something approaching a typewriter keyboard, and it was possible to opt out of single keyword entry when powering on. There were many Spectrum clones, notably behind the Iron Curtain and in Latin America, some of which extended the capabilities beyond recognition and were more like early ’90s PCs in their specs, and there were also computers such as the Sam Coupe which was far more capable than the Speccy but was also compatible with it. It was an incredibly persistent computer. Sinclair also had three rather nebulous projects connected to the Spectrum called the Loki, Janus and Pandora, which however did not materialise.
Through the ’80s, Sinclair was aiming to produce a laptop like the Grid Compass. They planned to do it with the ZX80, ZX81 and Spectrum. It almost came to fruition with their next computer. No new computers were announced in 1983 in spite of the huge glut of new micros being released by loads of different manufacturers. Then, in 1984, the QL was pronounced. Standing for “Quantum Leap”, this was alleged to be a 32-bit computer and Sinclair saw itself as having leapt over the 16-bit era and just going straight for 32-bits. However, it was based on the 68008, a version of the 68000, which was internally thirty-two bit but had an eight-bit data bus. As had often occurred before, the hardware was buggy and the first QLs were released with a lumpy thing hanging off the back called a “dongle” which fixed them. The QL was the first Sinclair computer to have something like a proper keyboard, although it still wasn’t up to the standards of many other more expensive home micros. Storage was in the form of two built-in Microdrives. This formed the basis of ICL’s One Per Desk, which was a hybrid computer and communications terminal and was used by BT as the Merlin Tonto. The QL didn’t seem to catch on, but it’s hard for me to tell because it coincided with the point when I decided to go cold turkey on IT, not liking the feeling of being addicted to computers and wanting to become a more balanced person.
A much more public failure was the notorious C5 electric vehicle. Sinclair had big plans for this, which would’ve climaxed with the C15, which was what we would call a Smart Car today and looked very similar. None of this happened of course, as the C5 itself was not a success. The only time I’ve seen a C5 in use was at one of the halls of residence at my university, where it was being pedalled around by one of the students. It’s shown at the top of this post. The C5 is an electrical tricycle with a polypropylene body designed partly by Lotus. One of the problems with it is that it’s too low to be visible from larger vehicles. The battery range was short, the maximum speed was only 24 kph and it suffered from the usual Sinclair problem of not being delivered on time. Sinclair’s vehicles division went into receivership after less than a year. However, even today it has an enthusiastic hobbyist community which has managed to soup it up to travel at over 200 kph, although I can’t say I fancy the idea of riding in or driving one at that speed. Research into developing the C5 had been going on for five years before it was released. There was, however, no effort to develop a more advanced battery than the lead-acid ones used in milk floats, with the rationale that better batteries would come along eventually from third-party manufacturers. Of course that did eventually happen, but not for decades after the C5 had bitten the dust. Reviews from the motoring press were decidedly negative. It’s considered to be one of the worst marketing failures since World War II.
Going back a couple of years, Sinclair made one more attempt at a pocket TV, this time a flat screen. Oddly, Sinclair had, as I’ve said, a fixation on the idea that CRTs would always be superior to LCDs. The idea behind the TV80, illustrated above, is to bend the electron beams round a corner to enable the electron gun to be placed beside the screen rather than behind it. In his ongoing dream of producing a laptop computer, Sir Clive planned to incorporate such displays in portable versions of his computers, and was dead set against LCDs. This may have been due to their inadequacy at the time. The TV80’s screen was also magnified by means of a Fresnel Lens, which are those magnifying things you used to see on the backs of buses and in lighthouses – flat, thin lenses which can magnify like ordinary contact lenses. However, it was noted at the time that LCDs would soon overtake this technology.
In 1986, Amstrad took over Sinclair. Amstrad continued to use the Sinclair trademark for some of its products, but from that point onward, Sinclair had no part in developing new products with his badge. This leads to the rather anomalous phenomenon of a Sinclair PC, the PC-200, which in 1988 was still using CGA and MDA. It had just two ISA slots for expansion, but the case wasn’t high enough to accommodate the cards. However, it was not really a SInclair product anyway.
In 1988, no longer able to release products under his own name, Sinclair finally achieved his dream of a portable, battery-powered computer with its own display, which was however LCD monochrome. This was the Z88, an interesting Z80-based device which included BBC BASIC and adapted versions of Acorn’s own productivity apps. When the Z88 first came out, I found it very confusing because it certainly seemed and looked like a proper Sinclair machine but wasn’t called that. It’s black, A4-sized and actually seems to have a nice keyboard for once. It feels like Sir Clive either couldn’t legally attach his name to it or didn’t think it was good publicity to do so. Compared to what he actually wanted to do, which was to have a large, possibly colour, flat screen display at an angle to the main unit, this is not it, and in fact from this point on most of his products feel like him making the best of a bad job. And I don’t feel that his stuff was actually shoddy as such, but that he was setting his sights lower henceforth. It must have felt like a bit of a comedown to have to use an LCD on this device.
I want to mention three more products which I think illustrate this sense of compromise. The first of these is an electric motor for a pedal bike called the ZETA – Zero Emission Transport Accessory. This appeared in 1994, was upgraded to the ZETA II in 1997, then the ZETA III, and was finally retired in I think 2002. It’s an electric motor powering a wheel which is fixed to a bicycle frame to boost its speed up to about 20 kph. Incidentally, the maximum speed here is the same as that of a C5 and this is no coincidence, because above that speed these devices would be officially classed as motor vehicles with the concomitant legal connotations. In fact, both the C5 and the ZETA could easily be designed to go faster, and hobbyist communities circumvent their limiters, but it changes their legal status. It kind of feels like Sir Clive was limiting himself in more ways than one with these. He also produced the Zike (he seems to like the letter Z), which was an electric bike, once again limited to 15 mph for the same reasons as the others, and weighing only eleven kilogrammes. This unfortunately failed probably because it was associated in the public mind with the C5 even though it was a complete rethink and if it had been produced by a different company it would priobably have done fine.
The absolute final bit of kit associated with the guy was the SeaScooter. This still exists and can still be bought! It came out in 2001 and is an underwater motorised vehicle scuba divers can hang onto to transport them through the sea. It goes at 3 kph and operates up to twenty metres depth, and can be recharged overnight. It’s a bit of a departure for Sinclair although once again there’s a sense of him adapting something unsuccessful to a new environment where he hoped it would achieve greater success.
Who, then, was Sir Clive Sinclair? Someone who was very much part of British private sector industry in the 1960s into the twenty-first century, whose ideas were ahead of both his time and his company and manufacture. Many of his products did have an air of cheapness about them, but they were also very impressive and high-concept, and he seems to have had a tenacity and resourcefulness you don’t see very often. It seems unlikely that we will see his like again.
A Box Of Chocolates 
