Nano-sized Microchips? HP Says So.

ImaLamer writes: "A C|Net News story is reporting that HP has announced they have made breakthroughs that 'help turn out powerful computers that fit on the head of a pin with room to spare.' Also in the article, that the patent announced Wednesday, will produce no two chips that are the same. 'Each one will be customized for a particular function,' says Stanley Williams, the chemist on the team. The work was done by himself, Phil Kuekes, a computer architect, and James Heath, a UCLA professor. The chips use nanowires and the chips are said to be even less than the size of bacterium. Sounds cool enough. The biggest part of the breakthrough isn't the chips themselves, but that HP plans to be able to 'fix' chips which come out with imperfections, thus saving money on an already cheap process."

Comments

One thing I'd like to see

[RC514]

All the advances in chip design and manufacturing make computing hardware a big player game. One thing I'd really like to see is a technology which enables hobbyists to create microchips on a small scale. There are a few open source hardware projects around, but when it comes to manufacturing, only high numbers can be produced at reasonable costs. This advancement, promising no two chips will be the same, sounds a lot like what I want.

Maybe this is why they are cutting loose their PCs

[drenehtsral]

This could be part of why they are cutting loose their PC division. It seems they've been doing a lot of pure research lately. I hope some of this comes to market soon (within the next 5-10 years) and they aren't just filing speculative patents.

On the humorous side, maybe they can use this tech to start making the HP48gx again and overclock it to 1ghz =:-)

So, how do we interface to these nanocomputers?

[mfarah]

Ok, let's say that they can make a REALLY small computer. How do we interface to it? Via a REALLY small keyboard? With a REALLY small monitor?

I think this makes more sense if they have some kind of networking capability, and that they'll be able to form some sort of "sensor cluster", much like in the way Vernor Vinge's A Deepness in the Sky's Qeng-Ho's network of dust computers worked. Of course, there's still too much to work on for that.

I hope HP begins work on some sort of nanoTCP/IP.

Re:Of course...

[caesar-auf-nihil]

Not so. As you get down to smaller scales, sometimes heat dissapation becomes easier as there are different methods of heat release than just fans and heat sink.
Basicailly it depends on the structure of the chip. If its inorganic semiconductors, which have to push heat through a rigid crystalline structure, then they tend to hold onto their heat longer due to poor heat conductivity. Therefore, they tend to heat up and stay heated up, and it takes more effort to cool them.

However, while no details were given, the tech probably won't be inorganic semiconductor based, and therefore could just release heat by the release of energy through the chemical bonds in the structure. You would get some heat, but some of that energy would get converted into moving electrons back and forth in each of the molecular bonds. In fact, its possible that they're relying up on the heat to get certain atoms to jump to higher energy state, thus turning a switch on or off, and when they rapidly cool back down, they activate or shut off the switch as appropriate.

Then again, its very likely they haven't considered this, and the first time they hook it up and starting running computations there is a puff of smoke and the chip is now CO2 and ash.

Re:Interesting story...

[mac.newbold]

For computer companies to make a profit, they currently have to make chips at almost 99%+ effeciency, so that they throw away very few chips per silicon wafer.

I hate to nitpick, but you really should check your stats before spouting them off like that. A chip manufacturer making something like a P3/P4 or Athlon chip with those kind of yields would own the world by now. Nobody making big (by area) chips these days has yields anywhere near that high. Even 90% is doing good. They often range much lower than that, 80% and worse even, but because of the high price of the good ones that they sell, they still barely make a profit. The way the Intels and AMDs keep on going is by selling a lot of them, not by making much of a profit on each one. For DRAM memory, yields are often around 50%. So don't assume that they need to get 99+% yield to make a profit. There are a lot more variables in that equation that they can and do work with.

Mac