Slashdot Mirror
Nanotechnology To Replace Conventional CMOS
neutron_p writes "There is a lot of hype around nanotechnology these days, but some things are going to work in a near future anyway. IMEC announced a program aimed at seeking alternatives to the current CMOS manufacturing technologies using nanotechnologies. IMEC will investigate the use of semiconducting wires, carbon nanotubes and spintronics or electron spin." (IMEC describes itself as "Europe's leading independent research center" in the fields of microelectronics and nanotechnology.)
Finally an application for nanotechnology that's getting some attention that doesn't involve little machines crawling around my brain and floating through my blood stream.
The term "nanotechnology" is becoming a bit inflated. Companies know that merely mentioning the word nanotechnology in a press-release ensures a lot of extra attention. It is kind of sad that good ol' Slashdot seems to fall for this cheap trick as well.
Not too long ago, nanotechnology was about wonderful fantasies of small machines at nano-level assembling molecules or even medicines. Too bad that simple physics prevent this from becoming a reality (the resistance of air at nano-level is too large, for example). Now, nanotechnology is about punching very small holes in metal.
Maybe it is about time that everyone realizes that nanotechnology is not as exiting as it used to be.
My karma ran over your dogma
If this happens soon then it'll cause some trouble; colleges generally don't teach nanotech because it's still very much under development. I'm in a program right now that focuses on CMOS integrated circuit design ... am I going to be obsolete before I get out of grad school?
I guess this happens whenever things shift to a new technology, but still, it's sort of depressing.
How many cores will there be on a single chip a decade from now? About 64-128 if Moore's law holds (remember, it's the number of transistors, not the speed). I highly doubt we'll see more than 4-8 cores on desktop computers, even given a whole decade. There will come a point where adding more cores won't be economical or the sensible answer for more computing power, just as making chips faster isn't the answer right now. The paradigm will then shift towards either making the individual cores smaller in order to pack more on a single chip and/or making each individual core run faster and cooler. Nanotechnology is positioning itself as the answer ten years from now, you just can't see it. Putting any more than 4 cores in the same form factor as today's CPUs will absolutely require this kind of technology.
Seriously; all this talk about nanotechnology, is mostly bunk; we ALLREADY have nanotechnology, most of you are using it RIGHT NOW!
The modern computer processor has bridges that are etched in silicone, a mere 70nm across; that's right 70 nanometers. So instead of saying nanotechnology like it something strange that has never been done before the article should say 'new chips to use nanotubes and spintronix(sp?)'
-Millions of Monkeys, Millions of typewriters, 6 hours of sorting through faeces encrusted pages to find: This post