The Speculist

Yeah Frank, it says here that it's the same kind of work, but closer to the ground.

KurzweilAI points this morning to an interesting article on 3D computer chip designs (a subject Phil and I spoke about in our latest edition of Fast Forward Radio).

The reason that manufacturers would be interested in 3D computer chips is that Moore's Law (which predicted the exponential improvement of the 2D integrated circuit) will soon fail. Gordon Moore himself said earlier this year that the law will soon fail as transistors reach the limits of miniaturization at atomic levels.

Just as civil engineers of the 1880s began building skyscrapers in crowded cities, [James] Lu is pioneering chip real estate by developing high-rise, 3-D chips to alleviate congestion in integrated circuits.

This may seem too obvious. You run out of room at the bottom of your beige box for a single layer of integrated chips, just install a second board above it, right?

Well, getting more 2D integrated circuits into your beige box by installing a second board above the first is not an answer to the problem. The key to greater and greater performance of integrated circuits has been a shrinking of the distance between transistors on the chip. The limits of the integrated circuit are not overcome by the sort of 3D computing that is really just 2D computing "folded over" to fit in a box.

What Lu is attempting to do is get the transistors on the second floor working directly with the transistors directly below them on the first floor - as well as the transistors on all sides.

In addition to keeping the computation level of Mr. Spock's chess set on its exponential track after Moore's Law has failed, 3D computing could allow other innovations, many that haven't been imagined yet.

Wafer-level stacking also allows for short connections between different types of chips. “Particularly today the industry is trying to combine memory with the processor, and more than half of the chip is taken up by memory,” Lu explains. “When we stack layers, we have a processor on the bottom and layer the memory on top, with a short access time between them.” Lu says the reduction of memory access time would be a huge advancement for large-scale computer clusters calculating nuclear reactions and weather broadcasting, for example.

And what would be a huge improvement for the big iron guys becomes a huge improvement for we PC users a cycle or two later.

“You are also creating new functionality,” says Nalamasu. “Such technology has vast implications, for example, integrating biochips with silicon chips.

If this would allow different types of computers to work well together, I wonder if this technology could also allow quantum computers to work closely with traditional computers. Quantum computers theoretically offer unimaginable power for a certain class of problems, but are, apparently, useless for other things. This might allow the best of both types of computing.