SCI/TECH: Optical Computer made from Frozen Light

Scientists demonstrate control over light to form a core, or CPU, of an optical computer. Using ultra cold atoms called BEC's, experiments have shown that information can be retained or stored in the light. Previous experiments have failed to do this.




While we don’t have an optical computer spinning out computations, this breakthrough is as significant as the first silicon circuit. The ability to retain information in light that has been slowed and controlled is the quantum step to take us out of the silicon world of computers.


Related News Links:
www.deas.harvard.edu

[edit on 4/12/05 by NoPhobos]

[edit on 4/12/05 by NoPhobos]

Incredible! I woner why they need to slow it down though. Wouldn't circuts that moved at the speed of light shareing information be better then well... below the speed of light?

Nonethe less this is really cool news.

Certain data needs to be held in memory for reuse or later use in computations.

Thats what Cache ram on current CPU chips is for.

When working with light based computing, storing this data in silicon based memory would slow the entire light computing down to the speed of the cache chips, loosing any edge you have.

Cache on a CPU is propably the most important part when it comes to the actual speed of a CPU.

Thats why slow cheap CPU's that run at same frequency's as their performance counterparts, only have 1 real difference, and thats cache size.

Good examples of this are the First Celeron, wich didn't have L2 cache but had the exact same core as a P2 wich in turn had 512KB cache.
Then the 2nd Celeron, wich had 128KB L2 and the same core as a P3.

Still same with current gen CPU's, with Celeron vs P4 and Duron & Sempron vs Athlon. Where the cheap counterparts have primarely less L2 cache and in some cases also less L1 cache.

Creating memory for Light computing that runs with light too removes this speed limit imposed by the cache ram.

This really sounds like a great achivment
It will really be neat to see what beacomes of it, how it will be used, and to what degree of preformance.
Unforchantly I just dont see my self liveing long enuff to find out.

Originally posted by Red Golem
This really sounds like a great achivment
It will really be neat to see what beacomes of it, how it will be used, and to what degree of preformance.
Unforchantly I just dont see my self liveing long enuff to find out.

You don't see youself living another 5-10 years? How old are you?

i wonder if BEC stands for Bose-Einstein condensates...interesting. BEC is supposedly a phase of matter. This phase is in the opposite direction of plasma. BEC is a super cooled solid where plasma is a superheated gas. The speed of light is retarded severly in a BEC due to the density of the medium in which it is trying to travel through.

sardion2000
Actually I am in my early 40s. I just very much doubt I will see this get to the pratical stage because of what has become of the other next generation computer technologys. I rember how excited some people were about holographic memerys 18 years ago and nothing has happened with those that I know about. Granted from what I have read quantum computing is getting closer, there have been several patents for some of the componets ect... It is just if the time frame holds true for this form of light technology, odds are I just might be pushing up daiseys long before this beacomes avaible to the public.

Well it depends on what you mean by public. It will probably be very expensive at first like hundreds of thousands of dollars and being produced for research purposes, it will come down eventually and quickly. I remember when a top of the line PC cost 5 grand, now you can get one many many times better and still considered good for 1200 bucks monitor included. I keep kicking myself for not waiting longer when upgraded as it feels like a waste of money. But anywho, we NEED a replacement tech within 10 years anyhow if we expect to keep Moores law alive and well The tech companies LOVE thier little planned obsolecenance skam bah... Chances are they already have a working prototype as I remember reading a study around a year back that we would have that capacity RIGHT NOW. So prolly being used by military and research institutions atm....

The Future Of The Past

Originally posted by Red Golem
I just very much doubt I will see this get to the pratical stage because of what has become of the other next generation computer technologys.

Ah yes. I remember back in the '70s we were supposed to be driving hovercars by the Year 2000.

Well here it is the Year 2005 and no hovercars.

Where's my damn hovercar?

Originally posted by Majic
The Future Of The Past

Originally posted by Red Golem
I just very much doubt I will see this get to the pratical stage because of what has become of the other next generation computer technologys.

Ah yes. I remember back in the '70s we were supposed to be driving hovercars by the Year 2000.

Well here it is the Year 2005 and no hovercars.

Where's my damn hovercar?

Talk to this guy.

He seems to be the closest to that dream everyone seems to have I'm young still so I'll be patient. When I'm in my 50s then it will be different if it's not around.

yeah, but can I overclock it?

Imagine a beowulf cluster of these!

*sigh* too much slashdot.

now if they could make a room-temperature, solid state version of this, then we're talking!

sardion2000,
it is nice to think that the Gov already has a working model of this Tech. It would not be the first time it would happen. I just dont see it as to likely. I really do hope to see this get to the public domain, and I guess if it does, you can come back and say I told you so.

Originally posted by Red Golem
sardion2000,
it is nice to think that the Gov already has a working model of this Tech. It would not be the first time it would happen. I just dont see it as to likely. I really do hope to see this get to the public domain, and I guess if it does, you can come back and say I told you so.

Why don't you see it as likely? The IT industry wants to keep the status-quo, of faster, smaller and cheaper. Only way to continue the trend past 2015 is to go all optical or nano-analog, either way is fine with me lol

this is really interesting.... but will it be usefull and cost effective if the quantum computer comes into play?

How about a quantum computer, operating at close to the speed of light?

The two technologies could be combined to create the next great computer, or the next great world disaster.

That's the thing with this technology, it natually leads to QC which is still 50-60 years away from the public. Too many security issues with Quantum Computing at the moment...

Another rout to QC is something called Spintronics, this article talks about Optronics. There are two obvoius ways to get there personally I think since Optics are very well understood Optical based Quantum Computing is closer then "solid state" QC. Just my guesstimate though, so take with a grain of salt.

[edit on 14-4-2005 by sardion2000]

Is this actually slowing down/capturing light or simply delaying photon reemission by freezing an atom ? In the latter case, the prospect of information treatment by this technology may be overestimated, as with quantum computing...

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.

It would seem they can do both now.