Quantum Computer To Launch Next Week

judgecorp writes "D-Wave Systems of British Columbia is all set to demonstrate a 16-qubit quantum computer. Simple devices have been built in the lab before, and this is still a prototype, but it is a commercial project that aims to get quantum devices into computer rooms, solving tricky problems such as financial optimization. Most quantum computers have to be isolated from the outside world (look at them and they stop working). This one is an 'adiabatic' quantum computer — which means (in theory, says D-Wave) that it can live with thermal noise and give results without having to be isolated. There's a description of it here — and pretty pictures too."

I'll have to see it to believe it

[ZonkerWilliam]

There still so many issues with quantumn computing that haven't been resolved yet. Like for instance how do you get the information out without affecting it...

Re:Computer is snake oil

[MindStalker]

Umm. No duh. Its been speculated for a long time that general purpose quantum computing would be near impossible. Quantum computers will for a long time be co-processors that do special task that regular computers can't do. This one is built for quadratic equations. Which is exactly what Babbage's "computer" was initially built for. Sorry but we are still a long ways away but snake oil this is not..

Not true

[rbarreira]

That's not true, as far as I've read regarding attacks on symmetric ciphers with quantum computing. See these links:

http://www.google.com/search?q=quantum+computer+sy mmetric+ciphers+double+size&rls=com.microsoft:en-u s&ie=UTF-8&oe=UTF-8&startIndex=&startPage=1 [google.com]

Specifically, it is said that a quantum computer gives a quadratic speedup in the kind of searches involved in breaking symmetric ciphers. That means it's enough to double the size of the key in order to account for quantum computing...

Re:Quantum mystery

[Waffle Iron]

how often in modern society are you likey to end up confused.

However, avoiding any confusion can be vitally important. The next time you're straining to hold your front door closed against a mob of attacking zombies, and you yell out to your friends "Get Me A Torch, NOW!", you sure don't want them come back with some wimpy little flashlight.

Re:Single purpose... but solves NP-C, silly!

[oldmacdonald]

Gospodin is correct: Solving any NP-Complete problem is sufficient, as all the reductions are polynomial and classical. There are other complexity classes that can involve quantum reductions, but NP isn't one of them.

That said, there are good reasons to think that quantum computers can't solve NP-Complete problems anyway. Factoring, which breaks RSA, is in NP, but is not NP-Complete. So, even if D-Wave has built a true 16 qubit quantum computer (which I doubt) their claims aren't credible.

Re:Computer is snake oil

[Garse Janacek]

Quadratic Integer Programming != "quadratic equations" (though strictly speaking it does involve them). It's not about plugging in the quadratic formula or something, it's about optimizing over a set of quadratic inequalities. This is an NP-Complete problem, and I'm almost certain Babbage's computer was not built to solve an NP-Complete problem...

This could very well be snake oil, not in the sense that they don't have a device that solves what they say it does, but in their claims about the more general implications: (1) On such small inputs as we can assume they'll be using, of course it's trivial for any computer to solve that problem, so they aren't doing anything special. (2) As another poster points out, it's not even clear the extent to which this is really a "quantum computer." (3) Right now it's not even clear that it's theoretically possible for a quantum device to efficiently solve an NP-Complete problem (e.g. a quantum computer could, in theory, break your RSA key, though there are currently intractable engineering obstacles -- but it would be major news, regardless of engineering issues, if it was even theoretically possible to solve QIP efficiently). It seems odd that someone would announce a device that solves the problem (on very small inputs), without also announcing that e.g. this technique could be extended to larger inputs without exponential blowup (which after all is the only obstacle to solving the same problem classically).

Personally, I suspect that the device is partial snake oil in the sense that they are being misleading about how it really works, and that the algorithm is total snake oil in the sense that they don't really have an efficient algorithm for QIP in a more general quantum computing setting. But I guess we'll see...

IIATheoretical Computer Scientist