New records set for silicon quantum computing

Oct 16, 2014

Credit: Stephanie Simmons, UNSW

By Science Daily

Two research teams working in the same laboratories at UNSW Australia have found distinct solutions to a critical challenge that has held back the realisation of super powerful quantum computers.

The teams created two types of quantum bits, or “qubits” — the building blocks for quantum computers — that each process quantum data with an accuracy above 99%. The two findings have been published simultaneously today in the journal Nature Nanotechnology.

“For quantum computing to become a reality we need to operate the bits with very low error rates,” says Scientia Professor Andrew Dzurak, who is Director of the Australian National Fabrication Facility at UNSW, where the devices were made.

“We’ve now come up with two parallel pathways for building a quantum computer in silicon, each of which shows this super accuracy,” adds Associate Professor Andrea Morello from UNSW’s School of Electrical Engineering and Telecommunications.


 

Read the full article by clicking the name of the source located below.

7 comments on “New records set for silicon quantum computing

  • I’m glad I read the article before commenting. Because my first reaction was 99% accurate is not exactly what I would consider a “very low error rate”. In a traditional chip 1% failure would be a terrible error rate but if I’m understanding it the quantum computing paradigm has ways built in to detect failure and correct for it.

    Although I still think that quantum computing is over hyped. Even if it ever works, at least from what I understand, it only provides orders of magnitude improvement for very specific kinds of problems that are amenable to distributed solutions that can benefit from qbits such as code breaking. Also, it literally is a whole new paradigm so none of the software that runs on any existing chip could run on a quantum computer. And it’s not just a matter of recompiling the code the way it would be for other kinds of new chips. Because it’s a different paradigm you have to write new code from scratch that can take advantage of qbits rather than bytes.

    It’s still very cool but I think any usable technology is a very long way away.



    Report abuse

  • I remember learning a theorem in math that said if you could create components with accuracy anything other than random, it would be possible to build a computer with arbitrary accuracy. This requires massive redundancy, but it at least shows the problem is in theory soluble.

    The thing that makes me nervous is the CIA may have already funded various quantum code-cracking and already has the ability to routinely snoop on all encrypted communications in real time. If they had succeeded, they would certainly not tell everyone.

    In any case, if there is such a breakthrough in the public domain, suddenly encryption everywhere fails. We must fall back on one-time pads and true random keys.

    The other possibility for getting increased accuracy is you may be able to verify or refine rough quantum solutions with classical ones.



    Report abuse

  • The thing that makes me nervous is the CIA may have already funded various quantum code-cracking and already has the ability to routinely snoop on all encrypted communications in real time.

    It works both ways. If you can use quantum computer to crack regular binary computer files, it also means you can use quantum computer to encrypt the files, at the quantum level. You will then need a Higg’s Boson computer to crack the quantum encryption. Reminds me of those every receding turtles.



    Report abuse

  • it’s still slower progress that I was hoping but encouraging. Has no one thought to get Deepak Chopra involved in this project? I believe his knowledge of quantum mechanics is all too often overlooked



    Report abuse

  • a theorem in math that said if you could create components with accuracy anything other than random, it would be possible to build a computer with arbitrary accuracy

    What theorem was that?

    the CIA may have already funded various quantum code-cracking and already has the ability to routinely snoop on all encrypted communications in real time.

    Virtually zero chance that is true.The CIA doesn’t fund that kind of advanced research at all. That all goes through DARPA or if it would be anyone it would be NSA but even they would be very unlikely to do something that far out. Just for practical reasons, it would require a very big long term investment that might not pay off at all. Anyway, the NSA can ALREADY pretty much break into anything they want. As far as people like you or I are concerned they don’t need quantum computing to spy on us. The NSA already has access to Internet traffic via their relation with people who provide the infrastructure for the Internet backbone. The people that need to be worried about quantum computing are people like other governments who encrypt things with technology that the NSA can’t break or have back doors to. Your average citizen doesn’t bother with that.

    Back to the research though; another reason the NSA would be unlikely to do this in secret is the best physicist and computer science researchers don’t like to work that way. You can’t win the Nobel prize if you make a ground breaking discovery in physics but you can’t tell anyone about it.



    Report abuse

  • I think it is fair to say that any advancement like this has as many advantages as disadvantages, and it basically boils down to the purposes it is used for…

    NSA building a ‘quantum computer’ capable of breaking all forms of encryption

    It may also be used for many valuable applications other than cracking encryption:

    Extremely complex predictive analysis
    Machine learning where you need to assess many variables and many patterns and test models against it.
    Investment analysis
    drug discovery, for example, when you have trillions of combinations of amino acids to cycle through to find that single protein.
    solving complex optimization problems



    Report abuse

  • It may also be used for many valuable applications other than cracking encryption

    That’s something I’m not so sure about. Everything I’ve read talks about very specific code breaking problems as the only kinds of examples. Those examples are ideal for taking advantage of the qbit architecture. I have the same intuition as you though, it’s hard to believe there wouldn’t be more ways to exploit the technology. My guess is no one has bothered doing the hard engineering work to figure out how to design new algorithms or heuristics that take advantage of qbits because they are still so far from being practical and that code breaking is just the one people use because it’s an obvious good fit and because it’s one of the best ways to get funding given the emphasis of US R&D these days. It has to be military but even better if it can be spun as “homeland security” as code breaking can.

    But it is worth pointing out that at a minimum this would require some serious additional effort because the Qbit architecture would be the most radical departure from traditional Von Neumann chip architectures since digital computers replaced analog ones after WWII.



    Report abuse

Leave a Reply

View our comment policy.