Slashdot Mirror
Nanowires Four Times Faster Than Silicon
evileyetmc writes "Advances in nanowires have shown that they may be the future in cheap, high-performance electronics. Researchers at Harvard have shown that nanowire transistors are are least four times faster than existing silicon ones. These nanowires show promise in being able to be embedded in plastics, and could lead to devices such as flexible displays that process information in the screen itself."
The article says that we won't see this technology in computers and PDAs for a while because the relatively high cost of implementing mass production of nanowires cannot be justified by a mere 4x increase in speed. Its application will be limited to scientific research for now.
Still, there is hope for implanted computers.
Homestarrunner.net -- It's Dot Com!
Integrated with things like electronic paper, this would be brilliant - it would eliminate the need for a bulky separate processing unit. Imagine being able to hold a piece of paper that acts as a (very) basic computer...
Now the signal doesn't just get decrypted in the monitor, it doesn't even get decrypted and displayed until it reaches the display surface itself. Still doesn't close the analog hole, though...
PHEM - party like it's 1997-2003!
I do see a lot of potential for this technology for embedded systems use--particulary 'smart maps'--if we can embed display control electronics physically closer to the displays (lighter, thinner, etc). Once costs are researched down, some really neat shit is in the offing (OLED + nanowidth signal processors, anyone?).
"I've spent my whole life figuring out crazy ways to do things. It'll work." -- Montgomery Scott, "Relics"
What?!
Regardless of what Apple's marketing team tries to imply that camera is clearly build into the shell with the lens peaking through an opening above the monitor.
I know Apple likes to make their technology sound like it's more advanced than it really is, but rest assured that the display itself doesn't have a camera built in.
I'm not an EE, so I might be wrong about some of this, but this is how I understand things - please corroborate or correct as appropriate.
If the "hardware" is actually 4x faster than silicon, then that's a 4x increase for similarly scaled systems, right? The thing is that this technology can generate huge improvements in one of the primary focal points in chip design (aside from materials) over the last couple decades: smaller scale. There are several advantages to this: speed, heat, and power consumption, to name the top 3.
If you only have to send a signal 1/10th the distance to get it processed, that's a 10x increase in the throughput. If the processing also takes place in an area 1/10th the size, that's a full 10x increase in speed for the same construction material. (I pulled that 1/10th out of the air for ease of use, I realize nanowires could potentially construct circuits much smaller than this scale compared to current silicon architecture.)
Now, make that material 4x faster on top of the scaling improvements, and you have, not a 4x improvement, but a 40x improvement, right? Is there some glaring technical detail I'm missing?
I have a freind who does nanotube research.
The problem, as I understand, is sorting.
Not all nanotubes are conductive, and they can't be manufactured selectiveley.
But otherwise they behave similarly.
It's like me giving you a pile of billions of wires and saying: "Here, some of these conduct, and others don't. Now start sorting."
-... ---
People are throwing money at nano-this and nano-that because it has great PR, but nothing as yet has come remotely close to being a credible alternative to silicon CMOS for ULSI devices. Consider where silicon CMOS is at the moment - we can put a billion transistors all together on the same logic chip for tens of dollars. A bit of DRAM costs less than a billionth of a dollar. This is what we can do now - think how much further it will have gone in 15 years, when the new nano-stuff is supposed to be competing. Any new technology will have to be considerably better than what is already available for anyone to invest in it, and looking at the current state of things it's just not going to happen. They are banking on miracle breakthroughs. There is also a credibility issue with manufacture and interconnect. It's one thing to make one super-fast nanotube transistor and say "ooh, look how good it is!" But it's quite another to be able to put a trillion of them on the same chip, all wired together, for cheaper than CMOS. That is what they are going to have to do to compete with where silicon will be in 15-20 years. To be fair, the guy in the article seems well aware of this.