Phonon

[PB666]

http://www.redorbit.com/news/science/1113411769/for-the-first-time-scientists-observe-a-two-phonon-quantum-interference-123015/

This is kinda of a science dead period for the forum and just about anywhere. Phonons are simply waveform quanta. If we remeber everything has a wavelength, the more massive the object the smaller the wavelength. 

[Shamus6200]

So, the phonons (phoni?) being easier to setup and trap are going to help quantum information processing research. I understand that to mean that this helps bring quantum computer applications a little closer? I see that as being a really good thing, especially if I can run KSP on one of those rigs at some point :-)

[*Aqua*]

Unlikely. Quantum computers are only good at solving very specific sorts of problems (for example calculating prime numbers). They suck at calculating 'normal' stuff (addition, substraction, etc.).

Afaik the real problem is shielding the ions from all kinds of interference long enough to make them useful in a quantum computer.

[K^2]

2 hours ago, *Aqua* said:

Unlikely. Quantum computers are only good at solving very specific sorts of problems (for example calculating prime numbers). They suck at calculating 'normal' stuff (addition, substraction, etc.).

Sure, a QC won't help you to add two numbers together, but if you have a thousand pairs of numbers, QC can compute all of the sums in one operation. Since that's where all of our computational technology is going, especially in graphics, this is actually very valuable.

There are very serious physical limitations in building practical QCs. If we were to overcome them, they would make absolutely fantastic general purpose computers, allowing us to do things we couldn't dream of with ICs. But I wouldn't hold one's breath. So QCs don't suck at general purpose computation. We just haven't learned how to build ones that don't. There is some fantastic work on Quantum Error Correction out there, though.

2 hours ago, *Aqua* said:

Afaik the real problem is shielding the ions from all kinds of interference long enough to make them useful in a quantum computer.

That's severely oversimplified, but not wrong. The more complex a quantum system is, the more sensitive it becomes to errors due to random interactions. Currently, it means that for general purpose QC, we are limited to something between 10 and 15 qubits. And that's just not very useful. Although, specialized QCs exist with hundreds of qubits, that is achieved by extremely narrow specialization. D-Wave QCs are a good example. While it can't solve general problems, it's catching up fast with computational clusters on ability to crunch out artificial neural nets, and is expected to surpass them in near future.

[shynung]

Looking from a computer gamer's perspective, what would a rig with quantum-computing hardware inside can do that non-quantum equivalents can't?

[Steel]

13 minutes ago, shynung said:

Looking from a computer gamer's perspective, what would a rig with quantum-computing hardware inside can do that non-quantum equivalents can't?

Very little,  other than be many orders of magnitude more expensive to own and run.

 

Essentially the way everything is designed now is using simple calculations, adding, subtracting etc. Quantum computers have no benefit for these types of calculations, so unless your software is designed to utilise the special properties, a quantum computer will be much worse than a conventional one, at least until quantum computing has had as much development as conventional systems have already had

Edited January 3, 2016 by Steel

[K^2]

19 minutes ago, Steel said:

Essentially the way everything is designed now is using simple calculations, adding, subtracting etc. Quantum computers have no benefit for these types of calculations, so unless your software is designed to utilise the special properties, a quantum computer will be much worse than a conventional one, at least until quantum computing has had as much development as conventional systems have already had

Do you happen to know how graphics hardware works? I'll give you a hint, exactly like a Quantum Computer.

[shynung]

52 minutes ago, K^2 said:

Do you happen to know how graphics hardware works? I'll give you a hint, exactly like a Quantum Computer.

I really hope that means maximum graphics settings on every game ever created in the next 10 years or so.

^{Damn, I wish Fallout 4 wasn't choppy on my rig.}

Edited January 3, 2016 by shynung

[Steel]

52 minutes ago, K^2 said:

Do you happen to know how graphics hardware works? I'll give you a hint, exactly like a Quantum Computer.

What I was trying to say was that (as far as I understand it) if you fed the code currently used by the graphics cards into a quantum computer, it wouldn't be any faster because it's not designed to suit the unique advantages a QC has. Before any differences are seen, the way software is designed would need to change.

[K^2]

19 minutes ago, Steel said:

What I was trying to say was that (as far as I understand it) if you fed the code currently used by the graphics cards into a quantum computer, it wouldn't be any faster because it's not designed to suit the unique advantages a QC has. Before any differences are seen, the way software is designed would need to change.

The unique advantage of a QC is parallel computation. Which is precisely the feature the shader code is designed for. Except, in graphics card, we achieve parallel computation by having thousands of identical cores running this code in lockstep. QC can do all of this with a single core.

[Steel]

9 minutes ago, K^2 said:

The unique advantage of a QC is parallel computation. Which is precisely the feature the shader code is designed for. Except, in graphics card, we achieve parallel computation by having thousands of identical cores running this code in lockstep. QC can do all of this with a single core.

I stand corrected!