27 September 2015

Limitless

No, this is not a missive about a soon to premiere TeeVee show. Hardly. No, more musings on the evident reaching of the asymptote of human understanding of the physical world. More specifically, the glass is 99.44% full, not 10% as it was in the 19th century. Don't tell the Rightists, their heads will explode.

More confirming reporting today.
Technologists now believe that new generations of chips will come more slowly, perhaps every two and a half to three years. And by the middle of the next decade, they fear, there could be a reckoning, when the laws of physics dictate that transistors, by then composed of just a handful of molecules, will not function reliably. Then Moore's Law will come to an end, unless a new technological breakthrough occurs.

Some, with whom I've roundly disagreed in various forums, have suggested (not uniquely or first by any means) that some sort of bio based artificial neuron will displace the silicon transistor. Think about this for a second. At 14nm a transistor is composed of 36, more or less, atoms
Not quite, Si is FCC so the lattice constant is greater than the close packed atomic spacing by a factor of √2; you could have up to 36 atoms spanning 14nm depending on crystallographic orientation.
here
How are organic molecules going to top (bottom?) 36 atoms? Not hardly. In any case, as we see with driverless cars and computers that beat the best human chess players, the human brain has its limitations. Even Chinese industry is replacing cheap hands with robots. Oh, where will it end?

How many atoms does a transistor need to be this side of probabilistic? Haven't found a definitive statement, but 36 doesn't leave much room for production error, etc.
In July, Intel said it would push back the introduction of 10-nanometer technology (a human hair, by comparison, is about 75,000 nanometers wide) to 2017. The delay is a break with the company's tradition of introducing a generation of chips with smaller wires and transistors one year, followed by adding new design features the next.

Some say, quoted in the Times' piece, that slowing of Moore is good for competition. On the other hand, it could also mean that the first-to-Xnm will have a solid monopoly, since being second or third will be profitless: we've still not seen the legendary 450mm wafer, since there isn't sufficient demand. Xnm will increase output for that first mover, but without a Windows/Office cycle sink, where will the demand come from? IoT, say some. Yet the Jeep remote hijacking and VW perfidy make IoT a bit more of Big Brother (or, Borg) than might have been previously obvious.

A man's got to know his limitations. And the limits of the physical world. Einstein allowed that the universe is finite but unbounded.

No comments: