D-Wave - first commercial quantum computer

[pascal12]

D-Wave has claimed that they offer a commercial quantum computer, using a 128-qubit chipset. What exactly is going on here?
I'm sure that this is not the traditional quantum computer, since the record of http://www.physorg.com/news/2011-04-quantum-bits-physicists-limits.html" was reached only a couple months ago.

http://www.dwavesys.com/en/dw_homepage.html

 

[danR]

I live in Vancouver, and am naturally interested in this occasional 'story', which has been running on and off for a few years. One recent commenter after the article D-Wave cited in Nature said they used a product of D-waves, left a question about it, and got no reply from the company.

What they appear to be selling to Lockheed-Martin is a 'black-box' computing item that apparently gives the right outputs to L-M's inputs to L-M's satisfaction. If that's what L-M wants, that proves nothing about any purported Quantum mechanism to their contraption. They have published an interesting paper in Nature, but what real scientists would want is a reproducible description, or a device they can disassemble.

I called this company's early claims BS 3 years ago, and would like to be proven wrong. But I don't think I will.
.

 

[pascal12]

So is there anything "quantum" about their chips? Or did they take a regular chip and completely lie about it? Even though that it is likely they are misleading the public in implying the qubits are entangled and all, surely they can't be outright lying?

 

[Ivan92]

I doubt that they took a regular chip and lied about it. This article states that they took 16 8-qubit cells to form 128 qubit chipset.

http://www.physorg.com/news/2011-06-d-wave-commercial-quantum.html

They claimed they have built a 128 qubit chip set back in 2008, but it was yet to be verified. It is 2011 and they have sold it to Lockheed Martin. LM paid some serious money so they should get what they paid for.

http://en.wikipedia.org/wiki/Timeline_of_quantum_computing

 

[waht]

Google Workshop on the D-Wave quantum computer:

https://www.youtube.com/watch?v=56qR0iX5A4o

 

[danR]

I doubt that they took a regular chip and lied about it. This article states that they took 16 8-qubit cells to form 128 qubit chipset.

http://www.physorg.com/news/2011-06-d-wave-commercial-quantum.html

They claimed they have built a 128 qubit chip set back in 2008, but it was yet to be verified. It is 2011 and they have sold it to Lockheed Martin. LM paid some serious money so they should get what they paid for.

http://en.wikipedia.org/wiki/Timeline_of_quantum_computing

You understand how quantum-computing works, but you don't understand business. I remember reading about a Turkish rug-dealer who sold a carpet (a real carpet here, mind you) for a promissory note, to that famous author who wrote 'Airport', 'Wheels', etc. (can't remember the name, James something).

The dealer didn't care about the money. He put the signed promissory note on his wall and showed it off to future customers.

D-wave is showing off their sale to Lockheed-Martin, and that will leverage further sales, eventually at a profit. At least to the founders. I don't intend to join the share-holders however.

Smart.

 

[f95toli]

So is there anything "quantum" about their chips? Or did they take a regular chip and completely lie about it? Even though that it is likely they are misleading the public in implying the qubits are entangled and all, surely they can't be outright lying?

Yes, there is definitely something "quantum "about it. In their recent article in Nature they demonstrated tunnelling between states. This has been done in many other systems, but they were of course doing it using a subset of they QC chip which from an experimental point of view is very impressive (I'ver done similar experiments on smaller systems, doing it using a system as large as theirs is VERY tricky).
This in itself does not mean that it is a quantum computer, but it does mean that it uses some "quantum mechanical" properties of the system when it is operated.

 

[AlephZero]

It is 2011 and they have sold it to Lockheed Martin. LM paid some serious money so they should get what they paid for.

The price of $10m in your link might sound like a lot of money (even though it is spread over a multi-year contract), but in the context of the total computing or R&D budget of a company like Lockheed Martin it's not a huge amount.

Compare it with Seymour Cray's definition of a supercomputer, back in 1980: whatever you can build for $20m. That would be about $55m in today's money.

 

[pascal12]

Thanks for the replies... the Youtube video where he was explaining what the chip was supposed to do and all those mathematical functions really went over my head.
I think they mentioned that they built several of these in labs, but people would access it remotely - did I understand this right?
And I'm not sure how each of the "qubits" is supposed to actually be a qubit, since I think he said that it can only exist in two states.

Can someone clarify a little more what makes these computers "quantum"?

 

[pascal12]

Bump-ish?
Can someone clarify a little more about these "quantum" computers?
I'm not an expert in this area, and I only know the basic concepts of what quantum computers are. But I'm very curious, and it would be awesome if someone could explain in easier terms.

 

[Ivan92]

I found this helpful when I wanted to know what Quantum Computers are. Check it out:

http://www.physicspost.com/articles.php?articleId=241

 

[pascal12]

Thanks, I already know at a basic level how quantum computers work. However, the D-Wave computers are clearly not quantum computers, although they are labeled as such, so I'm wondering exactly how the D-Wave computers work and what makes them "quantum".