What You See Is What You Get: My Autobiography (43 page)

*

Amstrad was on the move again in April 1984. We had acquired the freehold of an office block in Brentwood which was to become our headquarters. The premises had great communication links, as it was smack bang opposite the mainline station with direct trains to the heart of the City and just a few minutes' drive from the M25. To put into perspective how far we had come, my first premises (part of the factory space in Great Sutton Street) was 1,000 sq ft; our new offices in Brentwood were 40,000 sq ft!

In the early eighties, Amstrad was a unique consumer electronics company, different from the likes of Sony, Philips, Ferguson and the rest, given that our range of products comprised audio, TV and VCR, as well as portable radio cassettes and car stereos. As I've said, by spring 1984 it was clear that our amazing growth had levelled out. The market had become accustomed to our profits doubling and they obviously felt that we'd run out of steam, and this was reflected in our share price. Typical of the media, there were comments like, 'Has the boy from Clapton run out of steam? Was this all a flash in the pan?' The same people who had bigged me up in the rags-to-riches stories had now turned the other way. I guess this was my first taste of adverse publicity, so understandably I was a bit worried. One of the things that has always fascinated me is how some so-called friends and associates of mine were always quick to report to me when they'd read a negative article in the paper, but never made a point of telling me they had seen the positive stuff. I have to say, that still goes on to this day.

Amstrad obviously needed to diversify. We had done a good job on audio, but I could see we had reached saturation point. Competitors had caught up and prices were dropping and I knew that we needed to move on and find another sector or product to bring us back to profit growth. We had been observing that personal computers made by Sinclair, Commodore and Atari had gained popularity over the last few years, and Bob Watkins and I had discussed getting into this market. Having bought lots of samples, we looked inside the machines and found there were hardly any electronics in there - just a few chips on a PCB. Ignoring all the high-tech bullshit spouted by the nerds who were seen to be the pioneers of the industry, we looked at how much this lump of plastic and silicon would cost to make.

We decided to make our own personal computer. I thought - and later publicly said - that the Sinclair computer looked like a pregnant calculator; it didn't look like good value for money. Also, at the time, people would buy a Sinclair computer and then have to buy a separate cassette player to connect to it. On top of that, they'd have to wire it all to the back of their television sets, which they'd use as monitors.

My concept was simple: Mum and Dad don't want little Johnny taking over the TV set, so our computer should come with its own monitor, have a full-sized keyboard and a built-in cassette mechanism for loading software
and
hit a target price of PS199. This way, little Johnny could have it in his bedroom, freeing up the family TV. A great concept.

At the speed of light, we drew all this up and made mock-up samples. There was just one problem - we were clueless when it came to things such as software and didn't know how to make the main PCB. So Bob engaged the services of a couple of long-haired hippies who had helped us out previously by tracking down an epidemic problem with some Orion TVs. These two were obviously a pair of bright sparks and Bob asked them if they knew anything about computers. Being anoraks, they immediately responded that of course they did and that they would easily be able to design a PCB to go inside this beautiful exterior hardware we showed them. We were moving into unknown territory and what happened next, I guess we can only look back on and laugh at.

Again, I don't want to be too technical, but I'll try to explain in simple terms what went wrong. Computers work using an operating system (OS) of some kind and the software for these operating systems takes thousands of hours to write. What one of these hippies claimed he could do in a month would, in reality, have taken a team of fifty software engineers a year, but Bob and I didn't know that at the time.

This new OS software would be contained in one chip known as a ROM (read-only memory) and the rest of the circuit was made up of other memory chips, a Zilog Z80 processor and a few other bits and pieces. Basically, the whole heart of the computer lay in the software on this single chip. We designed the PCB, which effectively had a hole in the middle, ready for the chip to go in. At this point, our computer would spark into life.

After about four weeks, Paul Kelly, the lead rocket scientist of this dynamic duo, went on the missing list. Bob received a phone call from his father one day saying that he'd found Paul drunk, lying on the floor. Paul's father accused Bob of being responsible because of the undue pressure we had put him under, hassling him for the software.

What a bloody cheek! We'd not put anyone under any pressure. Naively Bob and I had simply believed what this fellow had told us - that he could provide this software within a month.

The guy
did
eventually provide us with some software, which we believed was the finished item. We sent this software to Toshiba, the chip manufacturer in Japan, only to find that the data we had sent was total garbage. The fellow had just supplied us with some rubbish to keep Bob quiet. It was an absolute disaster. We'd learned very quickly that this computer business wasn't just a case of chucking a bunch of chips into a box and putting a plastic cabinet around it. We were entering a new world and, with our heads in our hands, we wondered what we would do next.

Bob knew another guy, Bill Poel, who was an academic, rather tall, very well spoken and knowledgeable about computers. Bill is one of those people who has great ideas, but has difficulty in executing them. There was no question of a doubt that Bill could spot trends in the computer industry - he still does to this day - but there is something missing, something which I've never been able to put my finger on, which means he's been unable to turn his expertise into pounds, shillings and pence.

Bill introduced us to another brilliant chap called Roland Perry, a shortish fellow with thinning hair. When this Cambridge graduate and Bill discussed things, you could see they knew what they were talking about and, in fact, a week later, they'd come up with a masterplan. They wanted to base our OS on a design by the company Locomotive Software, who had already developed approximately 80 per cent of what we required. Bill and Roland also suggested that we engage the services of Mark Jones, a hardware specialist.

The secret to low-cost computers in those days was to condense lots of electronic circuitry into one chip. We had been led down the garden path by the two hippies - Mark Jones explained that for
their
version of the computer to have worked we would have needed two giant PCBs filled with hundreds of chips. However, Mark told us it was possible to condense all these chips into one superchip known as a gate array. This would leave you with a small PCB on which you stuck the gate array, together with a few memory chips and the chip containing the OS.

The development of this gate array was going to take at least ten to twelve weeks. The technology was in its infancy, but Mark Jones was one of the foremost experts in the field. Bill and Roland had put together the A-Team and we were pressing ahead.

Our computer project was given the code-name 'Arnold', an anagram of Roland. It was typical in those days to give such projects a name because in
the fast-moving, ultra-competitive market, manufacturers wanted to keep their work confidential. So rather than quote a model number, which would give the game away, we would liaise with component suppliers and ask them to quote on materials for Project Arnold.

I could write a book about the development of our first computer, the CPC464. This book, however, is about my life's journey, so I'll have to restrain myself from talking too much about one of the greatest events in my business life.

We designed the hardware based on lessons learned from the success of the tower system. It was clear in my mind that the solution to be sold to the public had to be one of
simplicity,
not only in the operation of the computer, but also in the way it was set up. We decided that the computer system would comprise just two items: the monitor and the main keyboard. We'd also make it look like a
real
computer, like the thing people saw at the airport when they were checking-in. The charcoal-grey keyboard was about two feet long with multicoloured keys and a built-in cassette data recorder. The monitor also looked great, with the same charcoal-grey, high-tech cabinet, but in reality it contained nothing more than standard black-and-white television circuitry converted to remove the TV tuner.

The industry was full of snobs who spoke in haughty, intellectual terms, trying to imply that the electronics involved in computers was something way above that used in the general consumer electronics industry. Fortunately, I recognised at a very early stage that this was a load of bollocks. The way I looked at it, chips are chips, PCBs are PCBs and plastic is plastic. Simple as that.

In the case of the monitor, we originally designed it with a standard twelve-inch black-and-white TV tube. However, when one saw the characters displayed on the screen, it didn't look very good. So I decided to get the manufacturer to change the phosphor inside the tube to green, so it looked like a computer screen - when you typed the characters, they appeared light green on a dark green background.

We also decided that the power supply to drive the main computer keyboard should be built into the monitor. This would provide the same simple solution we employed on tower systems. In other words, plug in the monitor at the wall, then plug the keyboard into the monitor. Easy. This was my brief to Bob Watkins and his team.

One of the things about Emperor Otake, when you put to one side his bullshit, arrogance and mad ways, was that he was a fantastic manufacturer. His speciality was getting products straight to market, so he could capitalise
on sales in new trends. Like everyone, he had his good points and his bad points. On this occasion I had to put up with his nonsense and massage his ego in order to motivate him to enter into this new computer venture with me. Orion was much like Amstrad and the wise old fox Otake knew that computers were an up-and-coming market.

Despite the fact that he'd screwed me on the VCR deal with Dixons, I put my personal feelings to one side simply because of his ability to act fast. Besides, there was no way he could screw me on computers because we were entering a world where intellectual property meant that we owned the product and nobody could sell it without our permission. What's more,
we
would supply him with the two special custom chips, which were the heart of the computer, on a free-issue basis and as these chips were made exclusively for Amstrad, the chip-maker could not sell them to anyone else.

I had a meeting with Otake in London and I explained to him that we would have to do business in a completely different way when it came to these computers.
We
were the designers of the product;
he
had the assembly facility in Korea, as well as very good contacts with the component manufacturers. I persuaded him that what we should do was discuss the cost of this product on an 'open bill of materials' basis. We pooled our knowledge of component prices and established the cost of materials, then we established a labour cost for assembly and, in view of the fact that he had no risk (as our orders would be irrevocable), agreed on a 5 per cent profit margin for his company.

Initially, this was difficult for Otake to accept. He didn't like the idea of being treated as a sub-contractor and often reminded me not to call him that. Nevertheless, once his ego got over it, he agreed to this arrangement. To get him to agree, I played him at his own game and implied that I had other options open to me. I told him there were plenty of Taiwanese manufacturers who would be willing to work in this way. In truth, I didn't have any other maker in mind, as my best experience to date had been with Orion. My bluff worked and he backed down and agreed to my suggestion.

This open bill of materials method of doing business was to be the way forward for Amstrad in the computer industry. We were to learn that the prices of components would tumble, week by week, and it was important that we never placed too many orders for certain components, particularly things like memory chips.

Our biggest struggle was with the British electronics company Ferranti, who allegedly were experts in the gate-array technology we were using. We had a bad experience with them and I'm ashamed that this company, which was supposed to be a glowing example of British technology, caused us such
problems. I won't go into detail, but at the eleventh hour we realised we were getting nowhere with these people. The promised production and launch dates were getting close and there we were without the heart of our computer.

One of the things I've learned is how to cut through all the smoke and mirrors to get to the heart of the matter when faced with problems and delays in production. I tended to get involved with the technicalities which, in the early days, upset a few of my people, who saw it as interfering. Under normal circumstances, the boss of a company such as mine would ask his lieutenant, in this case Bob Watkins, to sort out the problem. Unfortunately for people like Bob, despite me doing just that, I still wanted to hear things first hand. To Bob's frustration, before he knew it, I was on the phone to Bill, Roland and Mark Jones, as well as the technical people at Ferranti, wanting to understand why the thing wasn't working and why we wouldn't be receiving our order on time. Having built up a load of enthusiasm about our forthcoming computer, the reality was that we weren't going to be able to ship any because Ferranti was holding us up. Armed with all this information, I asked my people a simple question, 'Isn't there anyone else who can supply this stuff?'