Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Supercomputing Science

Quantum Computing and Optically Controlled Electrons 74

Posted by samzenpus from the greased-lightning dept.
eldavojohn writes "Researchers have released a new paper on quantum computing theorizing how to use optically controlled electrons to make an ultrafast quantum computer. From the article, "Scientists have designed a scheme to create one of the fastest quantum computers to date using light pulses to rotate electron spins, which serve as quantum bits. This technique improves the overall clock rate of the quantum computer, which could lead to the fastest potentially scalable quantum computing scheme of which the scientists are aware.""
This discussion has been archived. No new comments can be posted.

Quantum Computing and Optically Controlled Electrons More

Quantum Computing and Optically Controlled Electrons

Comments Filter:

  • Ultrfast? (Score:4, Funny)

    by dotslashdot ( 694478 ) on Thursday August 16, 2007 @01:26AM (#20246017)
    Ultrfast--so fast, there is no time for the "a". Either that, or the "a" is like Shroedinger's ct. What's the likelihood of tht?

    • Re: (Score:2, Funny)

      by Anonymous Coward

      Ultrfast--so fast, there is no time for the "a". Either that, or the "a" is like Shroedinger's ct. What's the likelihood of tht?

      Well, way to change the word by measuring it. Could have been UltrEfast, UltrIfast, UltrOfast, etc. all at once! But nooooooooooooo, you had to go and find out it's actually "Ultrafast" way to ruin it for everyone. Well, at least you've provided us with other words that we might enjoy the duality of.

      Mmmmmm, that is some mutherfuckin fine duality.

    • Re: (Score:1, Redundant)

      by l0b0 ( 803611 )
      Lost in computation...

    • Re: (Score:2, Funny)

      by Anonymous Coward
      Ultrfast: fast enough to cause spontaneous vowel movements.

  • Still too slow... (Score:2, Interesting)

    by Var1abl3 ( 1021413 )
    Sorry as I am not "up to speed" on my quantum computing but from TFA "In the article, we give the limit of about 100 GHz, which is assuming a very high magnetic field, which would require superconducting magnets to achieve. "

    and then...

    "Proper error correction may reduce the speed of the quantum computer to 1-10 MHz."
    I already have 3+ GHz machines so why would I want to have the cost of a superconduction magnet and the cooling that goes with it to get a machine that is slower than an Intel P?

    I am nu

    • Re:Still too slow... (Score:5, Informative)

      by stevelinton ( 4044 ) <sal@dcs.st-and.ac.uk> on Thursday August 16, 2007 @01:37AM (#20246099) Homepage
      Because a quantum computer does fundamentally more at each clock.
      Factoring an 1000-bigt integer takes CPU centuries on modern procesors but would be just afew million operations for a quantum computer.
    • Quantum computers are about as fast at usual computer tasks as an average computer... that's not what they're for, that's not what they're made for. Their real power lies in their ability to calculate and analyze certain things - namely, Quantum stuff, which it can do much better than your average pentium.

      At least that's the explanation I got some time ago from a physics-guy friend.

      • Re: (Score:1, Flamebait)

        by baldass_newbie ( 136609 )

        Their real power lies in their ability to calculate and analyze certain things - namely, Quantum stuff, which it can do much better than your average pentium.

        At least that's the explanation I got some time ago from a physics-guy friend.

        Your 'physics-guy friend' *cough* faggot *cough* doesn't know shit about systems architecture. Bone up [wikipedia.org].

    • Re:Still too slow... (Score:4, Informative)

      by Spikeles ( 972972 ) on Thursday August 16, 2007 @01:52AM (#20246165)
      You wouldn't want a quantum based computer unless you had some type of problem that could be broken down into appropriate search spaces that are compatible with quantum based computing. You won't be seeing massive speedups in Quake 4 with this, you won't see Windows Xp/Vista start up any faster, that's not how they work [wikipedia.org].

      • Re: (Score:2, Funny)

        by someone1234 ( 830754 )
        Take out the error correction and you'll get an ultrfast Vista compatible computer.

      • Re: (Score:1, Interesting)

        by Anonymous Coward
        Actually, quantum computation could cause speedups for "regular" software if an exhaustive search is done for most optimized compiler output as performance changing parameters vary.

        This is the main reason I really want a quantum Lisp (or preferably Scheme) machine.
        • That's not execution. Thats more of using a quantum computer to determine when compiling the best sequence of operations to put out. That COULD indeed be represented as an appropriate quantum algorithm. And in the case of a Hotspot compiler (Java/.NET) it *could* make your programs execute faster at the expense of having to actually recompile every so often to update the performance changing parameters. Of course, since there are only 3 problems(factoring, discrete logarithm, and quantum physics simulations

      • you won't see Windows Xp/Vista start up any faster
        So not even quantum mechanics (the fundamentals behind quantum computing) can solve the problems with Windows? Guess Microsoft will have to go live in another dimension to get a good product out the door.

      • Re: (Score:1, Interesting)

        by Anonymous Coward
        You won't be seeing massive speedups in Quake 4 with this

        Be careful, once a real quantum computer is built, coming up with applications just becomes a normal engineering problem.

        Quake N could use a quantum graphics card to do quantum raytracing. Imagine having the pixel's color of each surface being represented as a superposition of colors and then evaluating the reflections of all light points at the same time. Your output qubits will settle into a final color for each pixel.

        (Granted, I have no idea how su
    • I already have 3+ GHz machines so why would I want to have the cost of a superconduction magnet and the cooling that goes with it to get a machine that is slower than an Intel P?

      Because that is called progress! Back in the old days, the multiplication tables and a proficient abacus operator could beat any proto-computer out there. Does that mean that we shouldn't even have built them? You seem to forget the direct linage connecting your 3+ GHz computer with Bell's invention of the transistor that could n

  • Did a similar thing... (Score:5, Insightful)

    by Anonymous Coward on Thursday August 16, 2007 @01:55AM (#20246175)
    this summer...
    But different. I was working with electron spin resonance in solids. The set up used a superconducting magnet and a microwave source. We could actually measure absorption changes when the microwave energy matched the Zeeman split.
    There was even some talk about using the set up as a component for quantum computers.

    However, the people at the lab have started the discover that the primary relaxation method is fast phonon interactions. This must in fact be the case, otherwise the entire upper band would be overpopulated quicker than detectable. Anyway, as things stood, the materials we worked with proved to be ineffective as quantum switches. The spin property was far too transient.

    • We might make this work now (Score:5, Insightful)

      by Steeltoe ( 98226 ) on Thursday August 16, 2007 @08:14AM (#20248169) Homepage
      this summer...
      But different. I was working with electron spin resonance in solids. The set up used a superconducting magnet and a microwave source. We could actually measure absorption changes when the microwave energy matched the Zeeman split.
      There was even some talk about using the set up as a component for quantum computers.

      However, the people at the lab have started the discover that the primary relaxation method is fast phonon interactions. This must in fact be the case, otherwise the entire upper band would be overpopulated quicker than detectable. Anyway, as things stood, the materials we worked with proved to be ineffective as quantum switches. The spin property was far too transient.

      In our tests, we have been working with another component. Preliminary tests found the electronic reflection change to be adequately measured within the interaction timeframe. Although none of our instruments were powerful enough to keep up with the fastest cycles, the information bits could be stored in cubic fashion, and then looked-up in strange cubit pairs after the fact. We theorize that a switch can be made, if only the energy is high enough to be reliably detected and stored.

      So if we can somehow correlate the high energy of the absorption and readability of the reflection, we can combine the power of the two methods to enhance eachother and cancel out the negative aspects, I think we can have something that will finally work! When properly set up, it should only be a matter of phase-adjusting the two polarities of the photon switches to be in exact oposition to eachother, while making sure no interference can be made across the photon shields. You may have to distort the angle by a tiny fraction due to stellar polarity in our locality, but that should be easy as pie once you have the two photon switches ready.

  • One of the fastest? (Score:3, Interesting)

    by Jartan ( 219704 ) on Thursday August 16, 2007 @02:14AM (#20246247)
    Did I miss something? To be one of the fastest that would imply a quantum computer already exists? If that's true why isn't everyone going nuts that the man will be able to read our encrypted email soon?

    • Re: (Score:2)

      by OzRoy ( 602691 )
      It's my understanding they do. Just extremely basic.

      There was one not long ago that was able to do simple arithmetic using 7 atoms. Not exactly code cracking capable, but still interesting.

    • Re: (Score:3, Informative)

      by JustinRLynn ( 831164 )
      Well, in short they are going nuts because they simply can't do that. While quantum cryptographic analysis can break existing encryption methods that rely on factoring, new methods are being developed [wikipedia.org] that are safe and secure. No system, however sophisticated, can crack the one time pad provided complete shannon security is maintained.
      • or rather, aren't going nuts .. heh, preview is your friend... :)

      • You are misunderstanding it (Score:5, Informative)

        by aepervius ( 535155 ) on Thursday August 16, 2007 @03:18AM (#20246547)
        Quantum Cryptography in that case is using the normal cryptographic method , just use a secure line to exchange the private key. But once the data are stored and the public key is known, it does not matter whether the private key has been transmitted on secure line or NOT. It is only a matter of cracking the crypto-algorithm itself which is then as weak against quantum computer factorisation. The only way to avoid that would be to get ride of public/private key system altogether, and use an OTP as key transmitted on a secure QC line.

        But in the very end, QC is only a way of transmission without eavesdropping. It is NOT an encryption algorithm.
        • Yes, the use of quantum key distribution to enable the use of the OTP cypher is exactly what I was referring to. I'm sorry if I didn't make myself clear.
        • Which is why quantum cryptography uses a one-time pad as an encryption algorithm which, coupled with quantum key distribution, becomes secure. It's not like OTPs are rocket science. It's the QKD part that's hard.
          • Non-quantum ellipsis math is not prime factorization, and theoretically "really tough" to do on a Qomputer.

            Can a regular computer be fast enough so that it can beat a q-computer at its own game?

    • Re: (Score:1)

      by schotty ( 519567 )
      Because although the potential is phemomenonally great, it just isnt there yet to kill 1 terabit encryption yet.

      http://www.spectrum.ieee.org/feb07/comments/1710 [ieee.org]

      The computer demo aforementioned in the link (quick google search, sorry if its not the best article) is a mere 16qbits. From what I gathered, 256qbits is the minimum for a useful qpc. But I am no professional or even a hobbyist in the field. Talk to one if you want more reliable data.

    • Re: (Score:1)

      by m50d ( 797211 )
      They do exist. If you're using 4-bit keys, better worry, the man's gonna find out that 15=3x5 soon.

  • ... and the case of Missing Flux Capacitor (Score:3, Funny)

    by unity100 ( 970058 ) on Thursday August 16, 2007 @02:42AM (#20246377) Homepage Journal
    - should have read this article's title. It looks more from a movie thats a crossover between Lex Luthor finally does the superman and Back to the College Dorm 2.7

  • Just keep the cats out of it (Score:3, Funny)

    by Anonymous Coward on Thursday August 16, 2007 @03:24AM (#20246561)
    Or at least don't look until they are safely back out..

    • Re: (Score:2, Interesting)

      by Anonymous Coward
      OK.
      It's complicated, but I think I get it:

          Any story that contains the word "quantum", you post anything containing the word "cat", and the Mods rate it +3 funny.

      Right.
  • But will it run the new 3d engine of Carmack?

    Seriously, if we ever get as far as a well working quantum computer it would have a huge impact. Imagine IBM super computers that are a hundred times faster then the ones they build now. I wonder what kind of impact it would have on research that needs to calculate lots of huge formulas.

    • Re: (Score:2)

      by dido ( 9125 )

      The main use of such a quantum computer, besides the use of Shor's algorithm or the other quantum algorithms known, would be to simulate quantum mechanical phenomena. As you might imagine, simulating complex quantum mechanical systems on a normal computer takes prodigious amounts of memory and processing power (e.g. the gauge quantum chromodynamics simulations used to calculate particle masses and properties in the standard model of particle physics). A workable quantum computer would make such computatio

  • When I saw the title, at first, I misread it as "Quantum Computing and Optically Controlled Elections".
  • ... there are like two worthwhile comments to this article *sigh*. I'd hope to see an insightful post about how quantum computing will helps as overcome the obstacles posed by the density/heat/energy issues the current techonolgy has. The move to multicore cpus didn't just happen because multithreading is suddenly cool ...

  • Don't get too excited... (Score:5, Informative)

    by Ancient_Hacker ( 751168 ) on Thursday August 16, 2007 @06:04AM (#20247207)
    Don't get too excited-- most quantum computing ideas are rather far-fetched-- there are really hard roadblocks that are theoretically and practically very hard to solve. The basic one is you have to keep all the electrons from interacting in ANY WAY with the rest of the universes for a considerable length of time (on the quantum scale). The slightest interaction with anything else and the quantum magic goes *poof*.
    • Nevrtheless this wonder me some phylosophical questions.

      How the Heisenberg uncertainty affects to a quantum computer?
      What is the minimun energy neccesary to store one bit?
      What it is the maximun speed a bit can travel?
      Do the information can be stored with 100% of efficiency?
      Can the information be processed and transmited with 100% of reliability?

  • Imagine (Score:1, Redundant)

    by crhylove ( 205956 )
    a Beowulf clu..... Hey, wait, did we just get lapped? What's today?!? Damn Quantum Physics!!!

    Now I just need this unit to have the AI to know how to control and build my molecular assembler. K thnx!

    rhY
  • its a good R&D will be expect good stuff

  • Quantum Computer (Score:2, Interesting)

    by javalizard ( 781952 )
    The fully quantum computer probably won't ever exist. An operating system shouldn't be in a fuzzy state... ever. Imagine it, maybe i might be running Photoshop. Anywy, my point is that quntum computers will be auxiliary chips or, when the time comes, it will be in the form of a Quantum Processing Unit (QPU). It would be like the Floating Point Unit (FPU) or the Integer Unit (IU) in a processor. The various units DO run at different speeds. Some operations take 1 cycle and others upwrds of 5 or even 10
    • I'd also like to point out that the US Military will probably classify such a processor so most people don't have access to it or even know about it. I'd bet my banana that they are doing most of the funding for the QPU.
  • If the Internets have taught me anything it is that two is always better than one when it comes to load balancing, bandwidth allocation and processing large computational tasks.

    The Internet itself is a decentralized, node based system. BitTorrent uses this approach to solve many problems. Why are we thinking about one quantum computer? It would be big, ugly and vulnerable, even if it wasn't running Windoze. Spreading the tasks over many computers in a decentralized manner would be a better use of the ex
  • a really HOT cup of tea. Meanwhile, at http://arxiv.org/ftp/arxiv/papers/0708/0708.0681.p df [arxiv.org] we learn that quantum tunneling really is FTL. I'm feeling very depressed.
  • Is one of the two electron spin states in a higher energy state than the other? In other words, does an electron require more (or less) energy to switch from "UP" to "DOWN" than the reverse flip? Or are both spin energy states the same?

    If there is an energy difference, how big is it (minimum theoretical)? And how much is the minimum (theoretical) energy required to flip the state? I'm not talking about today's first generation flippers, which probably consume much more energy than is theoretically required.

    • Re: (Score:2)

      by fritsd ( 924429 )
      IIRC, not normally, but if you apply an external magnetic field then yes. See the wikipedia article about NMR for a simple explanation (yes I know it's NMR instead of EPR [wikipedia.org] there, but the general principle's the same). Also IIRC the energy difference is tiny (radio wave type frequency as opposed to, say, light). There was a simple explanation what effect this had on the relative populations of the both spins but I've forgotten it. I thought that it didn't cost a lot of energy under normal operation but I'm su

Slashdot Top Deals

Disclaimer: "These opinions are my own, though for a small fee they be yours too." -- Dave Haynie

Close