D-Wave Announces Commercially Available Quantum Computer

New submitter peetm writes "Computing company D-Wave has announced they're selling a quantum computing system commercially, which they're calling the D-Wave One. The D-Wave system comes equipped with a 128-qubit processor designed to perform discrete optimization operations. A qubit is the basic unit of quantum information – analogous to a bit in conventional computing. For a broader understanding of how qubits work, check out Ars Technica's excellent guide."

Comments

Re:So how many frames will this get in Crysis?

[Splodgey]

I would imagine that operations are instant. Unfortunately all the data gets sent to an identical 'you' in a parallel universe.......

Quantum annealing

[fph il quozientatore]

The name "quantum computer" is a bit misleading, since this thing as far as I understood is a classical computer that performs quickly an algorithm called quantum annealing (http://en.wikipedia.org/wiki/Quantum_annealing). If I understand correctly, the "128 qubits" part is snake oil, and it has nothing to do with the explanation of qubits given by Ars Technica in the other link.

Re:Quantum annealing

[zrbyte]

Exactly. The summary should of at least mentioned that there are serious controversies around the working and "quantumness" of the machine. http://en.wikipedia.org/wiki/D-Wave_Systems#Criticism http://www.nature.com/news/2011/110531/full/474018a.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+nature%2Frss%2Fcurrent+(Nature+-+Issue)&utm_content=Google+Reader

Re:Quantum annealing

[Coryoth]

As far as I'm aware the 128 "qubits" aren't entangled at all, which means it is useless for any of the quantum algorithms that one generally thinks of (Shor's algorithm for factoring, for example). It simply has 128 separate "qubits" that are queried individually, and is, essentially an augmented classical computer that gains a few minor advantages in some very specific algorithms (i.e. the quantum annealing algorithm) due to this qubit querying, but is otherwise indistinguishable from a really expensive classical computer for any other purpose.

Re:Quantum annealing

[Anonymous Coward]

Sorry, I stopped reading at "should of". Hurts my eyes too much.

Re:Quantum annealing

[paimin]

"Should of" is not a usage, it's a misspelling of "should've". "Should've" is a usage.

Re:Quantum annealing

[Anonymous Coward]

I did. It has nothing to do with this. While pronunciation of "should've" and "should of" are phonetically similar, there simply is no such thing as "should of"(which should, of course, be clear by now).

You are attempting to cover up for this erratum by distorting information. Please stop attempting to cite archaic linguistics that you don't understand.

This has the same central problem as before

[JoshuaZ]

This has the same central problem as before. D-Wave's computers haven't demonstrated that their commercial bits are entangled. There's no way to really distinguish what they are doing from essentially classical simulated annealing. And the set of problems which their machines can supposedly works on is an NP-hard problem minimization problem involving Ising spin where it isn't even clear that from a complexity standpoint that the the problem can be more quickly solved in general by a quantum system. (Essentially we don't know the relationship between BQP, the set of problems reliably solvable on a quantum computer in polynomial time http://en.wikipedia.org/wiki/BQP and NP http://en.wikipedia.org/wiki/NP_(complexity). Recommended reading that is skeptical of D-Wave's claims is much of what Scott Aaronson has wrote about them. See for example http://www.scottaaronson.com/blog/?p=639, http://www.scottaaronson.com/blog/?p=198 although interestingly after he visited D-Wave's labs in person his views changed slightly and became slightly more sympathetic to them http://www.scottaaronson.com/blog/?p=954.

Re:So how many frames will this get in Crysis?

[Anonymous Coward]

How many frames will this get in Crysis?

All of them, at the same time. Until you look at it, at which point you'll just get one, which might not be the one you want.

The Author speaks (and cringes a bit)

[TheAlexKnapp]

All - author of the piece speaking here. Yes, I'm aware of the D-Wave controversies, and talked with Scott Aaronson in a later piece at the time of the announcement. I'm cringing a little bit as I re-read this post because I know a heck of a lot more about quantum computing now than I did then. My take on D-Wave's computer now is that it's probably not a 'true' quantum computer in the sense that it involves any quantum speedup or entanglement. That said, I think that their annealing process is interesting in and of itself. I see their quantum computing tag as being akin to calling something '4G' in the wireless world. For those more interested in quantum computing, I updated the post to include some of the Q&A's I did about D-Wave at the time, as well as some of the quantum computing research I've covered since then, including some conversations with quantum computing researchers.

Re:So how many frames will this get in Crysis?

[julesh]

Yes, but it isn't the same type that D-Wave's processor solves, which is to say solving a particular class of differential equation in up to 16 8-bit variables.
Most of the time spent in ray tracing is actually in database searching (finding objects that a ray intersects with), which can be sped up by a quantum algorithm, but it isn't one that can be implemented on D-Wave's machine.

This article is a year old!

[SecurityTheatre]

Apparently, a quantum computer allows slashdot editors to see backwards in time.

Since this article was posted.....

  5/17/2011 @ 2:34PM

Right, that's almost a year ago that this "announcement" took place.

Whoops!