Google's Quantum Computer Proven To Be Real Thing (Almost) | Hacker News Reader

Google's Quantum Computer Proven To Be Real Thing (Almost)

wired.com

32 pointsLibertatea13 comments

"They must have switched this thing on today - all my Google searches now return pictures of cats that are both dead and alive."

--best comment for days in that linked article, lol

The title is very misleading IMO. There have been quite a lot of questions around D-Wave's quantum computing claims. The linked story is about a paper that shows that the D-Wave machine indeed shows signs of quantum entanglement.

The question whether D-Wave's quantum computer can achieve non-linear speed-up WRT classical computers is still not answered (but a lack of quantum entanglement would have ruled out such a speed-up).

The lack of scientific characterization of what the D-Wave machine actually does is quite odd, given that they have been selling these machines as "quantum computers" for quite some time.

As other commenters have pointed out, Scott Aaronson's blog is a very good resource for scientific discussion of D-Wave's claims: http://www.scottaaronson.com/blog/?p=1400

Very misleading title. The story is entirely about the D-Wave (supposed) quantum computer, which Google has now bought. It's D-Wave's quantum computer, Google really had nothing to do with it except that it bought one.

So it's like nearly everything Google is famous for then?

It's Dwaves creation but Google bought it. Last time I checked the search engine was created in house.

Yea, the search engine is about the only thing. Which is why I said "nearly everything" and not "everything".

Here is a paper that disproves the USC paper mentioned in the story. http://arxiv.org/abs/1305.4904

I appreciate the physics are rather complex, but I'm really struggling to understand why the D-Wave machine isn't regarded as a quantum computer by many. Doesn't D-Waves machine use qubits?

Also, why hasn't D-Wave come forward and explained their process to put those critics to rest? Is it just a matter of trade secrets (ie they want to retain their monopoly)? If that's the case, I thought that was one of the few credible reasons for patents.

I really get the feeling that I'm too stupid to grasp all this properly :(

A good summary by leading D-Wave skeptic Scott Aaronson http://www.scottaaronson.com/blog/?p=1400

Scott actually had some good thoughts about D_Wave at some point. A year or so ago, D-Wave actually gave him a personal tour of the place, and convinced him that they were legitimately doing good research (He wrote about this in his blog: http://www.scottaaronson.com/blog/?p=954 ). But things change over time, and with a lot of the largely unwarranted hype recently, he's opinion of them has changed

(Edited for confusingness.) Check out https://en.wikipedia.org/wiki/Adiabatic_quantum_computation under the paragraph "In practice..."

It's possible that such a computer will solve some interesting problems quickly, but no one even knows if an adiabatic quantum system is capable of solving the same problems as a "real" quantum computer.

What is a "real" quantum computer and how does D-Wave's machine differ? (I have a basic understand of quantum mechanics, but I feel woefully out of my depth when reading comparisons on this subject)

Adiabatic just means that the computer is kept in the ground state the whole time. There are lots of different non-adiabatic systems. The advantage to an adiabatic system is that it's easy to avoid outside interference most of the time. But a non-adiabatic computer doesn't have to worry about such tiny energy differences, so it can go faster. That's good, because it's generally more vulnerable to decoherence and the faster you get it done the better.

Probably the most common is a trapped-ion computer https://en.wikipedia.org/wiki/Trapped_ion_quantum_computer But there are also photonic ones: http://spectrum.ieee.org/computing/hardware/chip-does-part-o...

Edit: ooh, a new solid-state one! That looks promising http://www.nature.com/nphys/journal/v8/n10/full/nphys2385.ht... and even a weird nuclear physics one http://arstechnica.com/science/news/2011/07/yellow-diamonds-...

Reply on news.ycombinator.com