The first micros: Intel 4004's forgotten peers

The first micros: Intel 4004's forgotten peers

Summary: The 4004 chip is 40 today, so it's about time it had a mid-life identity crisis. Its creation story is well known — here's ZDNet UK's coverage from its 30th birthday — and it is indeed the first commercially available microprocessor.

SHARE:
TOPICS: Emerging Tech
1

The 4004 chip is 40 today, so it's about time it had a mid-life identity crisis. Its creation story is well known — here's ZDNet UK's coverage from its 30th birthday — and it is indeed the first commercially available microprocessor. Yet it didn't appear on a tablet of stone in one blinding flash of light (the genesis myth of so much new technology, especially if you talk to patent lawyers), and there are some intriguing twists in its story.

The canonical version of events is well-known. Busicom, a Japanese electronics manufacturer, designed a custom chip for its 141—PF calculator and asked Intel to produce it. Intel said that it would be more sensible to create a more general-purpose design in standard packaging; Busicom agreed and that company's Masatoshi Shima and Intel's Frederico Faggin, set to work, with Faggin doing the design and Shima the software. Ted Hoff, later to become Intel's first Fellow, is widely credited with coming up with the initial architectural concept.

But were they first with the idea? Almost certainly not: by the late 60s the minicomputer had democratised digital design, with lots of people realising they could create computers without needing to be an IBM. At the same time, digital integrated circuit design was progressing fast, with increasingly sophisticated dedicated logic circuits being created as single-chip designs. While plenty of people — including Intel — had solid doubts whether a general-purpose microprocessor would sell, it took a knowledge only of two of the most high-profile areas of electronic development to realise that such a thing was technically possible.

Evidence of this comes from a number of places, including Glenrothes in Scotland. Here, a company called Pico Electronics had designed and fabricated a single-chip RISC processor, including ROM and RAM on the same silicon, by 1970 — again for the calculator market. Because it couldn't work with external ROM it couldn't run general-purpose code: this was primarily a marketing and economic decision, and one that made sense — the chip and its successors sold well, ending up in Clive Sinclair's calculators among others.

Another close-but-not-quite design was the CADC, the Central Air Data Computer in the F—14 Tomcat jet fighter, which was a set of six chips making up a parallel processor running at around 375 kHz. That was fast enough to control the jet and run cockpit displays: the design was complete in 1970, but was classified by the US Navy and couldn't be discussed in any detail until 1998. This was a true reprogrammable high-density logic computer with off-chip program storage in ROMs but, of course, was never commercially available.

There are other contenders for attention. One of the highest profile examples is Texas Instruments, which introduced its TMS1802 in September 1971, ahead of the 4004. Again, it was definitely a microprocessor at heart, but permanently configured as a calculator chip. As you might expect with so many similar ideas appearing so close together, there were competing patents and lawsuits, cross-licensing agreements and out-of-court deals, and the real sequence of who invented what is hard to make out through the clouds of dust such activity throws up.

One final claim deserves relating, although it is by far the least well documented, relying on a few scraps of paper and the recollections of a man who left school in 1968 and was hired by Intel in 1970. Wayne D Pickette worked on the 4004 project but claims that prior to that, during his job interview with Intel founder Bob Noyce, he showed the company a block diagram of a microprocessor he'd started to work on three years previously when he was 17.

It had come about through his experience of the PDP 8/S minicomputer and his interest in discrete digital electronics; when he'd got a 1967 datasheet on the 74181 single-chip arithmetic logic chip, he realised that silicon design was good enough to create a complete computer on one circuit. That design, he posits, could be the true basis for the 4004, as he claims it was passed on to Ted Hoff after Pickette's interview.

That's certainly possible. It's also possible that Ted Hoff saw the potential in the original Busicom design, as the official story relates. And it doesn't matter: the invention of the microprocessor, like so many major and minor inventions, came about because many different factors aligned. If Intel had not even existed, another chip introduced at roughly the same time would have done the job just as well.

With computers becoming so important and silicon design getting so good, the microprocessor's time had come.

Topic: Emerging Tech

Rupert Goodwins

About Rupert Goodwins

Rupert started off as a nerdy lad expecting to be an electronics engineer, but having tried it for a while discovered that journalism was more fun. He ended up on PC Magazine in the early '90s, before that evolved into ZDNet UK - and Rupert evolved with them into an online journalist.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

1 comment
Log in or register to join the discussion
  • Good, interesting read Rupert, thanks for posting this. It brings back some nice memories - my first microprocessor course was assembler on a 4040.

    By the way, great picture...

    jw
    j.a.watson@...