There was a loose end in my piece on the astonishing proliferation of ARM microprocessors. How did they come to dominate the low-power niche? The first processor designed in 1985 by ARM’s ancestor Acorn was for a desktop computer the size and shape of a large typewriter, run off the mains. Surely power efficiency wasn’t a priority?
The answer is: good old British penny-pinching, with an assist from good old Brummagem standards of manufacturing.
From a story by Chris Bidmead for The Register:
According to [Acorn’s co-founder Stephen] Furber. â€œWe designed the ARM for an Acorn desktop product, where power isn’t of primary importance. But it had to be cheap. Cheap meant it had to go in a plastic package, plastic packages have a fairly high thermal resistance, so we had to bring it in under 1W.â€
When the first test chips came back from the lab on the 26 April 1985, Furber plugged one into a development board, and was happy to see it working perfectly first time. Deeply puzzling, though, was the reading on the multimeter connected in series with the power supply. The needle was at zero: the processor seemed to be consuming no power whatsoever.
As [the designer Sophie] Wilson tells it: â€œThe development board plugged the chip into had a fault: there was no current being sent down the power supply lines at all. The processor was actually running on leakage from the logic circuits. So the low-power big thing that the ARM is most valued for today, the reason that it’s on all your mobile phones, was a complete accident.”
Wilson had, it turned out, designed a powerful 32-bit processor that consumed no more than a tenth of a Watt.
Of course, somebody would have developed a low-power line of processors: the large potential market in mobile devices was obvious, and the RISC idea had been around for a while. Indeed, ARM has smaller competitors today, such as MIPS. However, it’s a matter of some interest in Britain how, against the odds, and the powerful economies of networking of Silicon Valley, a sleepy university town in the Fens managed to grab and hold on to a gratifyingly large slice of one of the world’s key technologies.
Pure luck? Not quite.
The designer of the chip, Sophie Wilson, was a young man of very superior talent. That’s not a typo: she is transgendered, and at the time was Roger Wilson. He was part of a very small team – including Chris Curry and Andy Hopper as well as Furber and Wilson – brought together at Acorn by an young expatriate Austrian venture capitalist called Hermann Hauser.
Why didn’t Wilson go and sell his high talents for lots of $$$ in Silicon Valley? Cambridge (England) is a nice environment apart from the weather, but so’s the Bay Area. My guess is that his/her transsexuality enormously raised the bar. Cambridge has long been tolerant of sexual diversity; think of Keynes, G.H. Hardy, the Apostles, and Alan Turing, not to mention Burgess and Blunt. Transsexuality is the ultimate test of such understanding, and if you are transgender and have found a safe and accepting environment, you will be very reluctant to leave it.
San Francisco is tolerant too, but Silicon Valley was created by strait-laced mid-Westerners (according to Tom Wolfe’s fine essay on Robert Noyce). Isn’t the hard-driving culture macho in other senses? The risks of moving in 1980 would have been very high to Roger Wilson. Sophie Wilson, presumably a wealthy woman, still lives a life of retiring domesticity in a satellite village near Cambridge. She is clearly strongly attached to the city.
So I suspect that the success of ARM is partly down to Cambridge’s exceptional tolerance. (It’s exceptional by British standards, themselves more relaxed today than middle America’s.) Discrimination is inefficient as well as wrong; and doing the right thing can pay off in the Kingdom of Serendip where we all live on alternate days.
A great engineer: