I have no idea what this means
Boffins make noise about D-Wave chip: it seems quantum
Researchers from University College, London, and the University of Southern California, have weighed into the ongoing “is it quantum?” D-Wave debate with an interesting approach, testing the device under a variety of noise conditions. As their paper at Arxiv explains, the thermal environment of a D-Wave chip isn't directly …
-
-
Thursday 3rd April 2014 08:14 GMT Vociferous
The D-Wave is sold as a quantum computer, but it uses a completely different approach than everyone expected quantum computers to use, and because of this different approach it isn't as universal as a "true" quantum computer would be. For this reason there is differing opinions on whether the D-Wave is a "true" quantum computer, or just a computer sold with quantum hype. The cited paper shows that it handles noise like a "true" quantum computer would.
-
Thursday 3rd April 2014 09:07 GMT Anonymous Coward
Correction: a universal quantum computer can solve pretty much any type of problems while some quantum computers are specialised on specific type of problems, that doesn't make them less quantum. That's the definition of universal and non-universal quantum computers.
The debate surrounding Dwave is simply whether it's quantum or not.
-
-
Thursday 3rd April 2014 08:08 GMT Vociferous
I don't get it.
I don't get the "we don't know if there's a quantum speedup" comment. The D-Wave is commercially available, the researchers clearly had access to a D-Wave machine, how could they conceivably NOT know if the device is significantly faster at solving problems suited for a quantum computer than a normal computer is?
-
Thursday 3rd April 2014 08:22 GMT Werner McGoole
Re: I don't get it.
That makes me scratch my head a bit too. Either it's a lot faster than a classical computer or it doesn't really matter whether it's a quantum computer or not.
I could say I'd sold my soul to the devil in return for the skills to make blisteringly fast computers, but if the computers I sold weren't actually fast, what would be the point?
-
Thursday 3rd April 2014 10:45 GMT ramsey66
Re: I don't get it.
Just because it's not yet faster than conventional processors doesn't mean it's not going to be. There's still a lot of architectural and engineering challenges to be solved to expose the quantum advantage. The technology is in the very early stages. Think about this, when the Wright brothers built their historic airplaine, did it go as fast as the fighter jets of today? The Wright brothers provided the basic principle for long-distance flight and pointed everybody in the right direction.
-
Thursday 3rd April 2014 11:08 GMT Anonymous Coward
Re: I don't get it. - @ramsey66
I think that is a very stretched analogy.
Cayley and Lillienthal (separately) established the principle of flight. The Wrights worked out how to add an engine to a glider. But at no point did anybody, as seems to be the equivalent case here, watch a glider and try to decide whether it was "flying" or not. It was sufficiently obvious.
-
-
-
Thursday 3rd April 2014 09:07 GMT ramsey66
Re: I don't get it.
That is easier said than done. In order to determine a speed up they need to make a fair comparison and that means finding the appropriate problem to solve to show the difference, otherwise they will end up comparing apples and oranges. FInding that problem is not easy. The last comparison they did found no evidence of speed up on Dwave but even that was not conclusive because of the type of problem being solved. Plus the Dwave archiitecture is not yet mature.
-
Thursday 3rd April 2014 09:48 GMT Phil Endecott
Re: I don't get it.
On average, it clearly solves problems faster than a "brute force" search would.
But there are better non-quantum methods than brute force, e.g. simulated annealing. It is not clear whether this machine is faster than non-quantum annealing. This is hard to measure because both classical annealing and the quantum mechanism have randomness.
Another complicating factor is that the device is small, so everything is fast. The differences would be much more obvious in a machine that could handle larger problems.
-
Thursday 3rd April 2014 10:13 GMT James Micallef
Re: I don't get it.
As per D-Wave's Wikipedia article (usual Wiki caveats), the D-wave shows remarkabel speedup in some applications and none (or even much slower than desktop) in others. I guess it's related to what a poster above referred as difference between 'quantum computer' and 'universal quantum computer', ie it probably does act as a quantum computer on the problems that show high speedup, but the performance gain is still not niversally applicable.
This is not surprising... even a quantum computer that can only solve a limited set of problems in a quantum way is a HUGE achievement.
-
-
Thursday 3rd April 2014 09:05 GMT Anonymous Coward
More Dwave evidence
This is another experiment that supports the Quantum behavior of the Dwave processor, however, it's not the final proof and it doesn't proove that it will be faster than classical computers.
In this experiment, they compared Dwave to 3 classical model and 1 quantum, and found that it behaves more like the quantum model than the classical ones. One of the classical models is the one from IBM that was debated recently and discussed here in the Reg. This is Dwave rebuttal to that experiment. Here's the previous Reg article about the IBM experiment:
http://www.theregister.co.uk/2014/02/04/boffins_say_dwave_machine_could_be_a_classic/
-
-
-
Thursday 3rd April 2014 10:50 GMT Anonymous Coward
@FartingHippo
Well, hostesses have on average got wider since DNA was present in earthly manifestation. One foot to the left could cause serious internal injuries, so we wouldn't want that to happen.
On the other hand, possession of a working hoverboard would probably mean that you could strew your path with discarded womens' undergarments, if that's what turns you on.
Posting this tripe because I can't comment intelligently on the article, except to note that this really does seem to be a solution in search of a suitable problem. This makes DNA references rather appropriate.
-
-
-
Thursday 3rd April 2014 10:44 GMT ramsey66
It seems everybody here has a hard time understanding about the "noise" in the experiment. "noise" (heat, magnetism) in a quantum machine generally destroys quantum processing (not good), the more noise the less quantum is the machine. The less noise the more quantum it is, and this is how they observed the behavior of the Dwave machine. This means it is consistent in behaving quantumly.
-
-
Friday 4th April 2014 08:16 GMT ramsey66
I used the term "quantumly" because based on the experiment, the Dwave machine may not fully utilize the quantum effects of the qubits depending on noise. Therefore it can be more quantum or less quantum. For example, it will entangle more or less qubits depending on the noise. The Dwave machine can still give you answers even when it's not fully entangled, you will just get better or worse answers. There's almost no right or wrong answers here.
You are probably thinking of the gate model quantum computing where if the entanglement is not perfect you will get a wrong answer. Dwave is based on Adiabatic Quantum Computing (AQC).
-
-