back to article Simon's says quantum computing will work

One of the hard parts of quantum computing is turning laboratory qubits into a calculation of anything. Now, South African scientists claim they've tested a handful of qubits against a 20-year-old algorithm to demonstrate that yes, a quantum computer can run it “faster” than a classical machine. Regular readers of quantum- …

  1. Destroy All Monsters Silver badge
    Headmaster

    WTF am I reading?

    To put that another way: if we know that a quantum computer can be made to outperform a classical machine, we've got a bit more confidence that the research money isn't being wasted.

    We DO know the former. If it turns out that for some as-yet-unknown reasons, the QC actually does NOT manage to follow the mathematical description of QM, that would be monster spectacular and open the road for re-evaluation. Shurely not a waste of money one way or the other.

    1. jzlondon

      Re: WTF am I reading?

      Win / win, right?

      If quantum computers work, that's great because they'll be incredibly useful.

      If quantum computers don't work, that's great because it opens up new science.

      1. Destroy All Monsters Silver badge
        Thumb Up

        Re: WTF am I reading?

        Yes

  2. dajames
    Holmes

    It's harder than it looks

    Quantum computing is certainly an interesting field, but it's a long way from being able to solve the sorts of real-world problems that it is supposed to be able to make trivial -- such as the factorization of RSA moduli.

    There are three big problems: construction of a system with enough qbits to be useful, working out how to present the problem, and working out what the hell the output means.

    This, for example:

    http://phys.org/news/2012-04-largest-factored-quantum-algorithm.html

    shows that progress is being made on at least the second of those problems, but systems with (say) 2048 qbits are still comfortingly (for those of us who would like RSA to continue to be worth using) far away!

    1. Pascal Monett Silver badge

      Sounds like the skiddies won't be hacking on qbit computers.

      Good.

    2. Michael Wojcik Silver badge

      Re: It's harder than it looks

      it's a long way from being able to solve the sorts of real-world problems that it is supposed to be able to make trivial -- such as the factorization of RSA moduli

      No one in the know has ever seriously claimed that QC would make RSA factorization "trivial".

      Shor's algorithm offers exponential reduction in number of trials, but the exponent is just 0.5 - that is, it reduces the work factor to its square root.

      That's equivalent to cutting the key length in half. There's a known remediation: double the key length.

      QC may someday be used to factor old, short RSA products. There's no reason to believe it will ever be used to break RSA in general, in practice, because users can just move to larger keys. That's an arms race QC cannot win.

  3. phil dude
    Black Helicopters

    research...

    is never "wasted". Noone ever said D-Wave was doing something serious, if nothing else it was a really cold fridge project.

    If true quantum computing is discovered, will we be allowed to use it?

    P.

  4. Brian Miller

    Why bother?

    With ASICs, FPGAs, GPUs, and eye-popping networks on a chip, will quantum computers really amount to anything? Either it's going to be too late, or else it's going to be too expensive for anything actually useful, except by Big Government and Big Corporation. Really, does anyone expect a quantum computer priced like a PC?

    By the time a quantum computer will be breaking encryption written today, the art of encryption will have moved beyond what would make that quantum computer practical. And no, it wouldn't be cheap enough for crims to purchase to crack our online transactions.

    I predict the obvious: bad implementation and practices are our biggest problem. Always have been, always will be. "Passwords? We don't need no stinkin' passwords!" "Encryption? Uh, I saw a movie with that in it."

    1. asdf

      Re: Why bother?

      >will computers really amount to anything? Either it's going to be too late, or else it's going to be too expensive for anything actually useful, except by Big Government and Big Corporation. Really, does anyone expect a computer priced like a typewriter?

      Is this what your father/grandfather said in the 1940s?

    2. Anonymous Coward
      Anonymous Coward

      Re: Why bother?

      Another few Moore cycles and "normal" ICs will be well and truely into the quantum realm. Already brushing up against it.

    3. Michael Wojcik Silver badge

      Re: Why bother?

      By the time a quantum computer will be breaking encryption written today, the art of encryption will have moved beyond what would make that quantum computer practical.

      Sigh.

      The "art of encryption" is already well beyond anything threatened by known QC algorithms. It's the practice of encryption that's potentially at risk, but that's trivial to fix.

      QC does not threaten general-purpose cryptography. It just doesn't. This is one of the most prevalent myths about QC, and there's no reason for it to go uncontested on a technical site.

      If we ever do have practical QC, its utility will be in making a larger pool of real instances of NP problems (and possible some problems that aren't in NP but simply have really bad polynomial complexity) tractable.

  5. Rich 2 Silver badge

    Question

    Can snyone out there explain how a quanrun computer works? Ie- how do you tell it what to do. How do you get data into it (and what does it look like)? What does a 'program' look like? That sort of thing

  6. Anonymous Coward
    Anonymous Coward

    Presumably a quantum computer will work and yet not work simultaneously. Much like Windows 8.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like