Forgot your password?
typodupeerror
Announcements Hardware Technology

Quantum Computing Using Traditional Transistors 323

Posted by CowboyNeal
from the reusing-old-hardware dept.
Ocean Consulting writes "UCLA is reporting progress on the quantum computing front by announcing success in controlling the spin of a single electron using an ordinary transistor." It's been a long road for the researchers involved, and even the project lead, Hong Wen Jiang admits, "...our initial theoretical calculations were very favorable, and gave us confidence to persevere."
This discussion has been archived. No new comments can be posted.

Quantum Computing Using Traditional Transistors

Comments Filter:
  • Awesome! (Score:4, Interesting)

    by erick99 (743982) * <homerun@gmail.com> on Thursday July 22, 2004 @06:51PM (#9774324)
    Once they get the cost down for actually reading the the state of an electron this will be awesome. Imagine only needing 100 transistors to:

    "With 100 transistors, each containing one of these electrons, you could have the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around," Yablonovitch said. "And why stop with 100 transistors?"

    That is pretty amazing.

    Cheers!

    Erick

    • I don't know... what does "implicit information storage" refer to? How am I supposed to take its... err, implicitness? Most of the time, I want my data to be pretty darned explicit, thank you very much. How much explicit storage (think: how many gigs of, erm, valuable research data ;) could you get off of these?
      • Re:Awesome! (Score:3, Interesting)

        by Sgs-Cruz (526085)
        It has to do with all the possible quantum states of the system. I.e. if each transistor has two states, there's 2^100 quantum states of the system when the system contains 100 transistors.

        Actually making use of those squillions of quantum states is something else entirely. It's not like you can just store that much information in 100 transistors, it's that it contains all possible combinations of those 2^100 quantum states while it's running.

        • Karma: pi (Mostly due to circular reasoning in posts).

          Sir:

          Your sig is irrational.

          Just thought you'd like to know that.

    • Yeah, may I be one of the first to say "YEEEEEHAWWW!"

      Quantum computing in general is one of the most exciting technological developments ever. I love reading about progress in this field.
    • Re:Awesome! (Score:3, Insightful)

      by argent (18001)
      Don't be too impressed... it doesn't mean as much as it sounds.

      It's kind of like saying a room full of monkeys implicitly encode all the works of Shakespeare.
      • It's kind of like saying a room full of monkeys implicitly encode all the works of Shakespeare.

        Yes, and no... Classically, no, but if they were quantum monkeys...

        The point of quantum computing is that each electron can simultaneously have two spin states - used as a 0 and 1 - and can therefore be both a 0 and 1 simultaneously. 3 qubits - (|)(|)(|), being both 0 and 1, encode 8 states, 000, 001, 010, 011, 100, 101, 110, and 111... all at the same time.

        So, a roomful of quantum monkeys, each of whom presse

    • They should compress the data down to 1 bit.
  • by agm (467017) * on Thursday July 22, 2004 @06:51PM (#9774326)
    Quantum computing, which holds the promise of nearly unlimited processing power, secure communications and the ability to decode encrypted conversations by terrorists and others, is a significant step closer to becoming a reality today with new research published by a team of UCLA scientists in the journal Nature.

    So which is it, secure communications or communications that can be spied on? It can't be both.
    • I had the same thought when I read that passage. The only thing that comes to mind is some form of secure communication that does not rely on or solely rely on encryption. And if that is the case, I wish the article had more detail because that would be very interesting.

      Cheers!

      Erick

      • The quantum channel is only used for creating one-time pads (OTP). The channel is eavesdrop-sensitive so you know which bits are compromised and you don't use them in the OTP. Then when you have generated a large enough shared OTP, you use it to encrypt the message and simply send that over regular channels, and since no one else has your securely generated OTP, that message is unbreakable.

    • Most encryption algorithms rely on it being easier to multiply numbers than to factorise them. Quantum computers can easily factorise a large number into a product of primes.

      This is how quantum computers can break encryption

      I'm not sure what they mean by the encryption that is secure though; Quantum encrytion as such is completely separate from Quantum computers, it is just a clever method using detection of the polarisation of light.

      The sending computer begins by sending photons in one of four con
      • Disclaimer: IANAPhysicist, and I know that because I read The Code Book by Simon Singh.

        What?

        Let me see...

        Oh, here it is, The Code Book by Simon Singh, page 244: "RidiculousPie is not a physicist."

        Must have been kind of weird to find that out in such a way.

    • by lenhap (717304) on Thursday July 22, 2004 @07:09PM (#9774482)
      It is secure and it allows encrypted communications to be spied on. What they don't tell you is that the encrypted comunications are encrypted using standard encryption methods around today. Things that can be cracked by exhaustive search.

      Using a quantum computer it can search every possible key simultaneously, cracking the encryption almost instantly. An example to understand this, you are in a building searching for your briefcase. Normal computers would go through every room one by one until they find it. A quantum computer would find the briefcase by existing in every room at the same time, finally settling (existance wise) in the room with the briefcase.

      They also mention quantum cryptography being uncrackable, this is true. If someone eaves drops on communication that is encrypted, it inherently destroys the data. The users will recognize intrusion and the eavesdropper cannot decrypt the message because the data has been destroyed.

      So yes, quantum computers can decrypt normal encryption that can be broken by exhaustive search and they can be used to provide quantum cryptograph which is a theoretically unbreakable form of communication.
      • by cephyn (461066) on Thursday July 22, 2004 @07:39PM (#9774674) Homepage
        Here's what i never understood, maybe you or someone can help me out...

        if eavesdropping on the encrypted transmission destroys it, couldnt the eavesdropper do so on purpose everytime, effectively jamming all transmission? Little point in having a secure way to communicate if no message can ever get through.
        • Yes, quantum communication is not magically DOS proof.

        • by bigberk (547360) <bigberk@users.pc9.org> on Friday July 23, 2004 @01:21AM (#9776771)
          if eavesdropping on the encrypted transmission destroys it, couldnt the eavesdropper do so on purpose everytime, effectively jamming all transmission?
          Definitely. The main problem with practical quantum crypto communications is this issue of information loss due to noise or tampering. If you could send photons over a lossless link (impossibility) then you guarantee entirely protected communications, or easy detection of tampering/eavesdropping.

          But since real transmission lines (even the best optic fibers) will always lose photons, you have to start adding on complicated processing to deal with the losses. Were the photons lost due to natural causes, or is someone eavesdropping? And if data is duplicated to account for losses, the system can possibly be tricked by an attacker into revealing information. This is a delicate subject and a great cause of complication in the field!

          The communications can also be jammed of course but the focus of the technology is delivering a secure link.
      • Eavesdropping can't possibly destroy the information in every case. The recieving unit has to be able to recieve the information so their must be some way to decode it (would be a pretty useless system if their wasn't, lol). If there is some way to decode it then I have a hard time believing that it is unbreakable.
        • Evesdropping will change the packets as they are enroute. This will cause a measured error rate to go up significantly, which will reveal the evesdropping.

          Google for sites that explain the whole process. It does in fact work.

          Basically, you transmit a one-time-pad over the encrypted link. If it was unintercepted, then you use the pad to encrypt your data, send that over normal communication routes, and decrypt that at the other end. A one-time-pad is of course unbreakable.

          If the pad was intercepted ov
      • It is secure and it allows encrypted communications to be spied on. What they don't tell you is that the encrypted comunications are encrypted using standard encryption methods around today. Things that can be cracked by exhaustive search.

        Not strictly true in all senses: some implementations use quantum effects to share a string of truely random -- not pseudo-random [wikipedia.org] -- bits which are then used in a one-time pad [wikipedia.org]. If someone eavesdrops on the exchange, it can be detected, and the message aborted (or change

        • Duh -- ignore the last bit about fake messages; if it's eavesdropped, there is no way to continue the message enciphering, unless the eavesdropper isn't aware that this is a Q.C. exchange, which is impossible. Shit, this stuff has me confused too and I read about it all the time. :/

          - Oisin
      • Using a quantum computer it can search every possible key simultaneously, cracking the encryption almost instantly.

        My understanding was that this is not true. At best you get the square root of the number of steps that would be required for a non-quantum brute force search. This means that key sizes are effectively halved, but that isn't an insurmountable problem.

        A bigger problem is that some algorithms are intrinsically vulnerable to quantum computing (or to rephrase, take far, far fewer steps to revers
        • by randombit (87792) on Friday July 23, 2004 @11:08AM (#9779610) Homepage
          My understanding was that this is not true. At best you get the square root of the number of steps that would be required for a non-quantum brute force search. This means that key sizes are effectively halved, but that isn't an insurmountable problem.

          Bingo. Which is why the AES competition required support for 256 bit keys, when even 128 bits is out of reach by any conceivable technology.

          Factoring is one such case, which is why quantum computing spells the death of RSA.

          Not true, necessarily. Shor's algorithm is algorithmically faster than the generalized number field sieve, but there is a constant in there. We don't know how big that constant is, and we won't until we have a quantum computer big enough to run Shor's algorithm (30 qubits or so, IIRC). It's entirely possible that Shor's algorithm is only faster then the GNFS once you hit keysizes of 10,000 bits, in which case it doesn't matter. OTOH, if Shor's algorithm is faster than the GNFS on 256 bit keys, we are, indeed, in some trouble. Of course running Shor's algorithm on a 1024 bit RSA key would take quite a large quantum computer, too.

          And, as you mention, there is no algorithm for compute discrete logarithms much faster than usual on quantum computers. I haven't heard about such an algorithm, anyway. Doesn't mean it doesn't exist, of course.

          Summary: algorithmic complexity is not the sole determinant of algorithm running time.

          In summary, quantum computing is powerful, but not a magic wand that makes all classical encryption schemes invalid.

          Thank you. Every time a quantum crypto or quantum computing store pops up here, the hype level seems to increase by several orders of magnitude. It's really annoying.
  • by zeux (129034) * on Thursday July 22, 2004 @06:52PM (#9774338)
    From the article:
    "With 100 transistors, each containing one of these electrons, you could have the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around," Yablonovitch said. "And why stop with 100 transistors?"

    Of course, because with 101 transistors you could store as many Library of Congress as there are electrons in the visible universe on a disk the size of 2 square hogs for a duration of up to 3.4256 parsecs.

    Unfortunately, it will take up to as many (1/98742) of year as it took in seconds for Apollo 11 to reach the moon from the launch pad to design such a hard-drive.

    Why is it scientists always use weird units? I have absolutely no clue of what "the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around" actually represents in bytes.
    • i'm not a quantum physicist (sorry for not using a lame IANA* acronym) but i believe that since quantum bits can represent both 1 and 0 at the same time and thus through all the permutations of those 100 bits (2^100 states) could represent that much data -- since all possible states are represented simultaneously? again, i'm fuzzy on the details, but this could get you started wikipedia entry for qubit [wikipedia.org] and there's some info on quantum entanglement that i havent chewed through yet that seems to be the basis
    • I have absolutely no clue of what "the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around" actually represents in bytes.

      Well, let's estimate that 1 million hard drives are made each year, at an average capacity of 200gig (both figures pulled out of the air; I have no idea how many hard drives will be made this year). That's a total capacity of 0.2 billion GB/year.

      The age of the universe is esti
      • I'd guess slighty more than 1 million hard drives are made in a year. According to this [yahoo.com] article, Apple shipped 876,000 Macs and 860,000 iPods last quarter. That's 1.6 million hard drives shipped by one (small) vendor in a quarter, so you'll need to up your numbers :-)
      • 600,000 1.8 inch drives per month according to a very quick google search. Total monthly production of all drives from all manufactureres will certainly exceed 1,000,000.
    • >Why is it scientists always use weird units? I have absolutely no clue of what "the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around" actually represents in bytes.

      6.0 X 10^8 drives x 100 GB/drive x 15 x 10^9 years x 1.1 x 10^9 bytes/GB = 9.9 x 10^29 bytes. More or less. Definitely a BFN. This should be enough for most mp3 and pr0n collections. For reference, the number of electrons in

    • From what I understand of quantum computing, the "implicit storage" would be storage used during the process of the calculation -- discrete data cannot be pumped in (beyond a classical 100 bit start state).

      Say you have an algorithm that needs to store all numbers from 0 to 2^100, then the algorithm excludes sets of them until arriving at an answer. The algorithm might truly need 2^100 words of 100 bits each to proceed on a classical computer, but on a quantum computer the calculation chugs along on our i

  • "Quantum Computing Using Traditional Transistors"

    That should like a LOT of soldering.

  • by Anonymous Coward on Thursday July 22, 2004 @06:53PM (#9774343)
    The article said something totally different brfore I clicked it.
    • It was in a quantum super-state. The act of observation (By clicking Read more...) made it evaluate to a particular single state.

      You might also notice that, now that you know what the whole article says, you don't know how long it took to load. If instead you had timed the page load, you wouldn't have been able to read the article.
  • Hrmm... (Score:5, Funny)

    by Arcanix (140337) on Thursday July 22, 2004 @06:53PM (#9774345)
    I thought about reading the article but will it change if I look at it?
  • wow! (Score:4, Informative)

    by quelrods (521005) * <quel@nOspAm.quelrod.net> on Thursday July 22, 2004 @06:53PM (#9774346) Homepage
    This would be something to help drive down the cost. Quantum computing on the desktop would finally be a evolutionary step in computing. (Up'ing clockspeed constantly and decreasing chip size is not evolutionary.) Though, quantum computing on the desktop probably means time to stop using passwords due to sheer power to brute force them.
  • by 7Ghent (115876) on Thursday July 22, 2004 @06:55PM (#9774366) Homepage
    They're actually using pulsed microwave bursts to manipulate the electron's spin, not the transistor itself, really.
    • Well seeing as you can lay down a junction on a silicon die that can produce that microwave burst just as easily as you can lay down a transistor the basic principle that you can do quantum computing with silicon is still being demonstrated.
  • by CommanderData (782739) * <.moc.oohay. .ta. .ihnivek.> on Thursday July 22, 2004 @06:58PM (#9774388)
    From the article:
    Quantum computing, which holds the promise of nearly unlimited processing power, secure communications and the ability to decode encrypted conversations by terrorists and others (emphasis mine)

    Take special note of the word others, which should be read as everyone. The government will be falling all over themselves to support this research and inherit a technology that makes encryption virtually useless.

    I'm all for advancing technology, and no doubt quantum computing will be a great leap forward. It's just a shame that our privacy will be sacrificed in the process.
    • Of course, the government also uses encryption, and they don't want theirs to be useless.
      • ...mostly the government uses "one time pads". They're immune to any decryption because for an encrypted message of length N, any plaintext of length N is identically plausible. The way to break govenment encryption is to copy the pads, or subvert the guy who gets to read the plaintext.
        • I don't doubt they use them with their most secret operatives. But do you really think every piece of data on government computers uses a 1 time pad? That all of their military commands do? The problem with 1 time pads is they aren't very practicle- its very hard to get pads to people, and losing a single pad totally fucks up communication
          • It's hard to get pads to people? Step one put pad on some sort of media, step two send the pad to someone before they need it. If you are willing to spend the money you can get anything you can fit in, say, a truck between any two points on the globe in 72 hours or less. Given that the assorted militaries of the world are generally all too happy to spend great quantities of the populace's money, I suspect there's a lot more 1TP use than you think there is. This is not to say that all military communications
    • Yes, but while Quantum computing allows current encryption to be cracked, it also allows use of stronger encryption which is equally difficult to crack.

      So supposing there's some impossible miracle breakthrough and by this time next year we've all got quantum computers on our desktop. Any communications you made previous would be crackable, but everything you were doing at the time would still be secure. I can't speak for everyone, but for most places you use encryption in your daily life, the information b
    • No worries. In the past, we use something not much better than ROT13. The best one that we got was probably ENIGMA before the computer age. They are pretty robust against the arithmatic ability of human brain.

      The currently using ones (public key encryption, DES3 etc) are all developed with silicon computers in mind. While they will become obsolete with quantum computing, I will be surprised if we cannot develop something better by then...

  • by Anonymous Coward
    ...to a quantum radio? I want to pull in stations from alternate universes since there is no good local music.
  • "secure communications and the ability to decode encrypted conversations by terrorists and others"

    Why does every article about anything have to have the word "terrorist" in it, these days?
    • Why does every article about anything have to have the word "terrorist" in it, these days?


      Because if you don't immediately say that it (it, as in indefinite pronoun) is a weapon against the terrorists, someone moranic will cry that it will be used by the terrorists. And morans tend to stick together when it comes to irrational fear, so you don't want that.

    • if you are for catching terrorists then you get government funding and avoid investigation.

      I you don't explicitly make the statement, you are instantly cast into suspicion..

      Welcome to this 'brave new world'... Exactly what Binny boy wanted..
      • if you are for catching terrorists then you get government funding and avoid investigation.

        On that note..
        • The NRA announces that "gun ownership prevents terrorism - we may never even know just how many lives were spared"
        • McDonalds announces "will the obese person sitting next to you shield you from the blast? Why the new McBiggerMac mitigates the threat of terrorism"
        • Michael Jackson claims "while sleeping with children I tell them stories about the nasty terrorists, and we have a good old Christian time.
    • Re:Terrorism. (Score:3, Insightful)

      by emeitner (513842)
      Fear sells.

    • Why does every article about anything have to have the word "terrorist" in it, these days?

      The guy is "Eli Yablonovitch, UCLA professor of electrical engineering, director of UCLA's Center for Nanoscience Innovation for Defense", so it'll help to get more funding (almost wrote "FUDding" there, I wonder why?).

  • by mdvlspwn99 (172473) on Thursday July 22, 2004 @07:01PM (#9774416) Journal
    is great. Until the technology becomes ubiquitous enough that even terrorists have access to it. Then what? It's secure...even from us.
    • then we'll just be where we are right now. big deal. we can't argue that all human beings have the right to free speech and the right to privacy* (*except for terrorists)

      no, we'll just have to figure out how to protect ourselves another way. maybe terrorism should just be viewed as a crime of opportunity. we can do lots of things to prevent the opportunity for terrorism from occurring without even infringing on people's rights. consider that proper pilot training and some secure cockpit doors could have pr
  • Quantum terms (Score:4, Interesting)

    by Decaff (42676) on Thursday July 22, 2004 @07:05PM (#9774456)
    I wish physicists would be more cautious in their use of language.

    In the article it states: "The UCLA team succeeded in flipping a single electron spin upside down."

    Considering that the term 'spin' is just a metaphor for a quantum-mechanical property that has no equivalent in our everyday experience, it makes no sense to talk about 'flipping' it, or the spin being 'upside down'.

    Neat achievement though....
    • Re:Quantum terms (Score:3, Interesting)

      by Ruie (30480)
      Actually this was pretty precise.

      You see electron spin space is a two dimensional complex space. It is tied to usual 3 dimensional real space via representation of SU(2) - so it is hard to say which direction corresponds to which.

      However, if one fixes a basis in usual space one can use it to fix a particular representation of SU(2).

      Furthermore, one of the basis vectors will have a particularly simple Pauli matrix corresponding to it - the direction of this vector is usually called "quantization axis".

      Of

  • by gphinch (722686) on Thursday July 22, 2004 @07:09PM (#9774485) Homepage
    "With 100 transistors, each containing one of these electrons, you could have the implicit information storage that corresponds to all of the hard disks made in the world this year, multiplied by the number of years the universe has been around," Yablonovitch said. "And why stop with 100 transistors?"

    I hope this drive lasts longer than the Quantumm Fireball I had.
  • Okay, I only skimmed the article (what's that STFA?) but this sounds like BIG news. If this holds true, existing hardware could be used for quantum computing - a very interesting possibility. Well, it's a long way from the science lab to everyday use, but I hope those guys can create something acutally usable throughout commercial computing.
    • Assuming your existing hardware is chilled to -400 farenheit, and equiped with a Maser.

      That was in the non-skimmy part of the article. It's pretty misleading, actually. A more accurate headline would have been, researchers flip the spin of an electron in a hunk of silicon that, coincidentaly, was located in an off-the-self transitor, using lots of expensive and complicated equipment.

      It's a good first step, though.
  • Ready for Doom 4! (Score:2, Interesting)

    by Evil Butters (772669)
    Cool... With one of these new quantum computers, I should be able to meet the minimum requirements for Doom 4! Now if only I could get my quantum video card to work...
  • DONT LOOK AT IT !!!

    It might go away...

    stupid lameness filter stupid lameness filter
    too many caps jackasses too many caps jackasses
    too bad you had to see this too bad you had to read this
  • Can anyone with more perspective on this, clue me in on what this really means? It seems like it would be a big breakthrough in quantum computing, making full on Quantum systems a possibility within a decade or so. Is this accurate, or are there still many more problems to solve that are more important than this one?
  • Terrorists? (Score:2, Insightful)

    by mikeg22 (601691)
    Oh, you mean the ones that use human couriers to relay messages? The ones that live in caves with no access to computers?

    No, this technology is not going to be used on terrorists. It is going to be used on a combination of normal people suspected of criminal activity (ie anyone who bothers to encrypt their communications) and actual hightech criminals.

    This technology will be effectively useless at stopping the terrorists we are worried about.
  • The actual article to appear in Nature can be found here [nature.com], which I found at the CNSI web page [ucla.edu].

    I only wish that CNSI will complete construction before I graduate with my Master's in CS... Seems like it will be a great facility to do research on this sort of thing. Oh well, there's always CENS :)

    - shadowmatter
  • Quantum computing, which holds the promise of nearly unlimited processing power, secure communications and the ability to decode encrypted conversations by terrorists and others

    I was willing to forgive a little hype until the idiocy about terrorists. Decided maybe I was just cranky, then read:

    While flipping a single electron was difficult, detecting that they had actually done so proved even harder. "We couldn't tell whether it was flipping," Jiang said. "It was like looking for a needle in a haysta

  • by reversible physicist (799350) on Thursday July 22, 2004 @09:12PM (#9775285)
    Quantum communication is already practical, and provides a secure way to communicate to replace factoring-based encryption, which quantum computation may one day make insecure. The hype in this article, though, is way over the top. 100 electron spins can only encode 100 classical bits. Not one bit extra. Yablonovitch is using a very sloppy way of talking about how hard it is to simulate 100 spins, and making it sound like he's talking about a way to store a lot of classical bits! His "implicit information storage" is nonsense. It's also worth mentioning that quantum computation is unlikely to speed up any computation you care about, unless you like to simulate quantum systems. Fast factoring is the "killer app" that got people excited about this field, but "terrorists" (and the rest of us) can just stop using factoring-based encryption.
  • What they did appears to be just flipping the spin (and detecting the flip) of a single electron.

    Creating and measuring the quantum state of a single particle is *not* new. Or that hard. The neat thing is that they did it with a commercial transistor.

    Oh, and some sort of fridge that goes to -400 F.

    The really hard part in quantum computing (as far as I can tell) is (a) creating and (b) maintaining a coherence between many particles.

    The problem is that useful coherences between particles are *very* easi
  • by bigsteve@dstc (140392) on Thursday July 22, 2004 @10:02PM (#9775633)
    Do these researchers call themselves "quantum spin doctors" or just plain "quantum mechanics"?
  • by teutonic_leech (596265) on Thursday July 22, 2004 @10:11PM (#9775684)
    Welcome to Windows Quantum 2006! We crash several ways at the same time!
  • by GogglesPisano (199483) on Thursday July 22, 2004 @10:13PM (#9775696)
    The following link may be helpful for those of us who are a little fuzzy on quantum computing: http://www.cs.caltech.edu/~westside/quantum-intro. html [caltech.edu]
  • The problem is that with each extra degree that you raise the temperature above absolute zero, you increase the random kinetic energy level that is exhibited by the particles which exist in the environment. These extraneous vibrations will make it exponentially more difficult to detect individual electron spins. Detecting individual electron spin amidst the incredibly chaotic motion that is induced by heat at room temperature would be about as difficult as hearing the chirping of a single distant grasshop

"Wish not to seem, but to be, the best." -- Aeschylus

Working...