Slashdot Mirror
Optical Computer Made From Frozen Light
neutron_p writes "Scientists at Harvard University have shown how ultra-cold atoms can be used to freeze and control light to form the "core" - or central processing unit - of an optical computer. Optical computers would transport information ten times faster than traditional electronic devices, smashing the intrinsic speed limit of silicon technology. This new research could be a major breakthrough in the quest to create super-fast computers that use light instead of electrons to process information. Professor Lene Hau is one of the world's foremost authorities on "slow light". Her research group became famous for slowing down light, which normally travels at 186,000 miles per second, to less than the speed of a bicycle."
Imagine trying to harness today's 3GHz CPUs with 1930s lab bench equipment. Digital electronics could have seemed another universe, out of reach in a universe of alternate physics "beyond radio". If photonic computation is within reach at artifically lowered speeds, we might be just about to cross the watershed, like going from transistor to ENIAC.
--
make install -not war
Obviously it's not simply a temperature thing, since most of space is absolute zero, and I can see stars and suns and stuff. So it's not freezing light as in freezing water.
So how exactly do you stop photons from moving? How does this affect relativity (e=mc^2)? How does this affect our perception of the universe - ie; if the light from the star that we think is 10,000 light years away is only moving 20mph or so, it could really be millions of light years away?
Does like, time slow down? My heads spinning. Freeze sounds like the wrong word.
I don't need no instructions to know how to rock!!!!
1) Wavelengths are too big: 1 micron is now a large number, and optics doesn't work much smaller than this.
This poster is correct. Since I have a Ph.D. in the field and the parent obviously knows something about optics, I might as well respond to the parent's critics.
IR photons are BIG. Forcing light to bend around corners is difficult. A waveguide must have a very high index of refraction if it is to be used to bend light within a reasonable radius. To the extent a Bose-Einstein Condensate helps this problem is encouraging if you don't mind cooling your computer to 2 millikelvin.
The speed of these optical computers always seems to come down to limitations of the silicon processors that work in conjunction with the light.
It's just a Bose-Einstein Condensate. These projects take time. While we are enamored with this BEC project, some poor grad student is working on carbon doping. Higher doping might improve the world of electronics far more than another optical computer claim.
I visited Hau's website and did, though, enjoy her papers. I just don't think the press release accurately portrays the low engineering potential of this work.
While those statements are true, I'm not sure if it's really legitimate to say that those wavelengths will work well inside a computational device.
Calling 13nm 'extreme ultraviolet' is marketing--those are really soft x-rays at that point. You're getting into photons that are inconveniently energetic. That's fine if you're doing lithographic etching of chips, but murderous on your hardware in daily operation.
We also don't have light sources capable of anywhere near the appropriate level of miniaturization for those very short wavelengths. Constructing one large EUV source for a chip fab plant is a very different engineering problem from constructing hundreds, thousands, or millions of such sources on each chip. The optics also get much more complex, expensive, and exotic as you move to shorter wavelengths. Once again, things that can be done in a billion-dollar chip fab are quite different from things that can be done on a hundred-dollar microchip.
~Idarubicin