Canadian Company Demonstrates 16 Qubit Quantum Computer
February 15th, 2007If these guys have a 16 qubit system, what does NSA have?
Via: Scientific American:
A Canadian manufacturer today unveiled what it called “the world’s first commercially viable quantum computer.” D-Wave Systems, Inc., “The Quantum Computing Company,” during a much ballyhooed rollout at the Computer History Museum in Mountain View, Calif., hailed the new device as a big step toward the age of quantum computing, decades earlier than scheduled.
But experts say the announcement may be a bit—er—premature. Even if the computer were to work as advertised, it still would be nearly 1,000 times too small to solve problems that stump ordinary computers. Moreover, researchers do not know whether it will work at bigger sizes.
“There are still a lot of ifs and maybes here,” says quantum computing researcher Seth Lloyd of the Massachusetts Institute of Technology. But he credits D-Wave for its willingness to test the idea. “From the scientific perspective,” he says, “what they’re doing is very interesting.”
A working quantum computer is the dream of every national security official and hacker on Earth. The bits inside existing computers constantly flip between 0 and 1 as they perform small steps such as “if 0, then 1.” But quantum physics allows particles like atoms, electrons and photons to be in two places at once—meaning they can represent 0 and 1 simultaneously, allowing more complex calculations.
Researchers believe that by combining many of these quantum bits, or qubits, they will be able to perform certain tasks that are currently out of reach. Chief among them: the ability to swiftly crack encrypted communications.
The NSA may have a bigger quantum computer, but for factoring purposes (Shor’s algorithm), you have to set up a quantum state with hundreds of qubits. There is an incredibly hard decoherence problem there, but even if it is solved, it is not clear that Shor’s algo can be applied on real-world problems. It assumes that the equations will be accurate to hundreds of decimal places; this is a very unrealistic assumption, the most accurate physical theories (such as QED) have not produced agreement to more than around 30 decimal places.
Of course, it may be that QCs can be usefully applied in crypto in other areas we may not have thought of, and all analysis is out the window if the gov really does have some future-tech.
And ROI-wise, it would almost certainly be easier to attack an RSA’d crypto link at the endpoints than cracking the middle.
I’m not convinced there’s anything to that company’s announcement. They haven’t demonstrated capabilities beyond that of a desktop computer. Meanwhile, who would they be allowed to sell quantum computers to? The NSA has no sense of humor when it comes to technologies that threaten national security (which a super-code-breaking-computer would be).
I think they are trolling for venture capital in a bull market…