[QUOTE]For the first time, a commercially available quantum computer has been pitted against an ordinary PC – and the quantum device left the regular machine in the dust.
Source:
[URL]http://www.newscientist.com/article/dn23519-commercial-quantum-computer-leaves-pc-in-the-dust.html[/URL][/QUOTE]
The future is here.
There are no graphs in the article, this is hardly scientific.
But can it run crysis [sp] im so sorry [/sp]
Well, time to sell my pc to some poor sap.
"qubits" and "niobium"
I feel like the guys who name these things do it just to fuck with us.
Cant see it saying anything about how may flops it does. If they cant even state how much faster it is by numbers, i will remain highly sceptical about the function of this one.
Now im just waiting, i hope 5-10 years from now they will succeed and ridiculously fast computers will be available for an awesome price.
[QUOTE]D-Wave, a company based in Burnaby, Canada, has been selling quantum computers since 2011, although [B]critics expressed doubt that their chips were actually harnessing the spooky action of quantum mechanics. That's because they use a non-mainstream method called [/B][B]adiabatic quantum computing.[/B][/QUOTE]
Another article dealing with the mentioned method.
[QUOTE]Aaronson also questions the quantum nature of the calculation. [B][U]"I didn't see any evidence that quantum behaviour [/U]played a role in finding the factors of 143,"[/B] he says. Rather, the experiment might have reached the conclusion in a classical way, he suggests.[/QUOTE]
[QUOTE][B]The speed tests are also not quite fair, because generic computers will always perform less well than a [U]device dedicated to solving a specific problem[/U], says McGeoch.[/B] "A next step would be to build a conventional processor optimised for this task, for a fairer comparison," says O'Brien.[/QUOTE]
But if Lockheed Martin are using their computer, I'm going to guess that they got to have something special.
Unfair comparison in many ways:
1) They pitted a very, VERY expensive "quantum" server against a desktop. They could have gotten similar results by comparing, say, a Roadrunner cabinet against the same desktop.
2) They're comparing against a very specific problem that their quantum computer was designed for. Moving even to similar problems required a software front-end to convert it, and significantly lowered their competitive edge. So unless you literally have millions to spend on a computer that will only be used to run a very specific problem, you're out of luck.
The latter is actually an inherent problem in true quantum computing - quantum computers only get the advantage of being quantum on certain algorithms (complexity class NP). In anything else, they'd be no better than a non-quantum computer of the same design. But here, they're not just failing on non-NP problems, they're failing on different NP problems.
Now, there are a lot of NP problems that commonly need to be solved in real-world usage. But not all of them. Really, because the only way for a non-quantum computer to attack an NP problem is with brute force, nearly any task currently done with computers will be non-NP.
[editline]11th May 2013[/editline]
[QUOTE=Zeneros;40605845]Cant see it saying anything about how may flops it does. If they cant even state how much faster it is by numbers, i will remain highly sceptical about the function of this one.[/QUOTE]
The problem is, quantum computers don't work using "flops". They act, in many respects, more like a dedicated ASIC.
The only way to compare them is through their performance at various problems, at various problem sizes.
[editline]11th May 2013[/editline]
[QUOTE=Madtoker;40605883]Now im just waiting, i hope 5-10 years from now they will succeed and ridiculously fast computers will be available for an awesome price.[/QUOTE]
While I won't rule out quantum computers ever being common consumer products, I can't say I expect that to happen within the next 50 years or so. They just aren't useful for many of the things you use a computer for.
What I can see happening is it becoming common for corporate use. Companies who regularly need to work with problems that are suitable for quantum computers will probably have one, the same way those who have a lot of data will buy a SAN.
[QUOTE=SpartanXC9;40605835]"qubits" and "niobium"
I feel like the guys who name these things do it just to fuck with us.[/QUOTE]
qubits are a form of quantum data processing and Niobium is an element
[url]https://en.wikipedia.org/wiki/Niobium[/url]
Still, you guys got to admit. The word "Quantum computers" sounds amazing.
I had never imagined that we'd see progress in this area this soon, and I hope more money is put into it.
[QUOTE=SpartanXC9;40605835]"qubits" and "niobium"
I feel like the guys who name these things do it just to fuck with us.[/QUOTE]
Discover the brand new superquantic computer
Now runs on Elitistium
i cant understand quantum. how is it possible?
[QUOTE=cucumber;40606357]i cant understand quantum. how is it possible?[/QUOTE]
One of the fundamental elements of quantum physics is something called "superposition". At a quantum scale (meaning at or below the level of atoms) when an event happens that could have multiple outcomes, that outcome isn't "fixed" until it is observed.
Say you flip a coin, except a coin so small quantum physics applies to it. Until you look, the coin is in a superposition of states - it is both heads-up and tails-up. And you can do things to that coin while it is in a superposition, as long as you don't actually look at the face.
Quantum entanglement is an extension of that. Let's use an actual example - a decay happens that produces two photons with opposite phase. Until you observe the phase of either one, both are in a superposition of phases. But when you observe one, the other one immediately drops out of superposition. The two were entangled.
The way you use that for computing is to have "bits" that can be in a superposition of their 1 and 0 state. You can use this to run certain problems in parallel, on the same "circuit". So instead of needing 10,000 cores to test various possible solutions to the problem, you run it once using qubits, and the only correct answer will be what you see when you finally observe it.
As a bit of a side note, other quantum effects already happen in traditional processors, as we're getting them to a scale where quantum effects can be significant. That's the main reason Intel had to switch to 3D transistors - quantum tunneling (where, because the position of particles is "fuzzy") was allowing too many electrons to "tunnel" through transistors they shouldn't have been able to go through. Some are actually trying to exploit these effects, most notably to make 1-transistor SRAM. But these are not quantum computers, as they don't use the superposition of states as a computing tool. They merely use quantum effects to help make more efficient circuits.
There is one other "quantum" computer that I wish more research was done on - rapid single-flux quantum computing. It essentially exploits certain quantum effects in superconductors to make faster and more efficient transistors. And by faster, I mean "early computers would be in the 50GHz range, scaling into the 100GHz range, and single transistors have been tested to terahertz frequencies". Sadly, between "superconductors need to be cooled with liquid nitrogen or helium" and "we can't design circuits that run at 10GHz, much less 50GHz" everyone seems to have given up on it.
Glorious quantum computer gaming master race.
The first buyable quantum computer should be called Schroedinger's box.
[QUOTE=gman003-main;40606457]One of the fundamental elements of quantum physics is something called "superposition". At a quantum scale (meaning at or below the level of atoms) when an event happens that could have multiple outcomes, that outcome isn't "fixed" until it is observed.
Say you flip a coin, except a coin so small quantum physics applies to it. Until you look, the coin is in a superposition of states - it is both heads-up and tails-up. And you can do things to that coin while it is in a superposition, as long as you don't actually look at the face.
Quantum entanglement is an extension of that. Let's use an actual example - a decay happens that produces two photons with opposite phase. Until you observe the phase of either one, both are in a superposition of phases. But when you observe one, the other one immediately drops out of superposition. The two were entangled.
The way you use that for computing is to have "bits" that can be in a superposition of their 1 and 0 state. You can use this to run certain problems in parallel, on the same "circuit". So instead of needing 10,000 cores to test various possible solutions to the problem, you run it once using qubits, and the only correct answer will be what you see when you finally observe it.
As a bit of a side note, other quantum effects already happen in traditional processors, as we're getting them to a scale where quantum effects can be significant. That's the main reason Intel had to switch to 3D transistors - quantum tunneling (where, because the position of particles is "fuzzy") was allowing too many electrons to "tunnel" through transistors they shouldn't have been able to go through. Some are actually trying to exploit these effects, most notably to make 1-transistor SRAM. But these are not quantum computers, as they don't use the superposition of states as a computing tool. They merely use quantum effects to help make more efficient circuits.
There is one other "quantum" computer that I wish more research was done on - rapid single-flux quantum computing. It essentially exploits certain quantum effects in superconductors to make faster and more efficient transistors. And by faster, I mean "early computers would be in the 50GHz range, scaling into the 100GHz range, and single transistors have been tested to terahertz frequencies". Sadly, between "superconductors need to be cooled with liquid nitrogen or helium" and "we can't design circuits that run at 10GHz, much less 50GHz" everyone seems to have given up on it.[/QUOTE]
And to clarify, this isn't a 'true' general purpose quantum computer. It would mainly be used in optimisation problems, true quantum computers control qubits using superconducting circuits in order to process information where it has two states, the sophisticated (and useful) quantum computers will make the qubits communicate with other qubits which is the jist of a quantum computer. However for the device mentioned in the article, the 'qubits' returns to a ground state and from there the optimisation can be worked out.
I want one for reasons.
And that reasons is VIDEOGAMES.
[QUOTE=Netsc;40606698]The first buyable quantum computer should be called Schroedinger's box.[/QUOTE]
How can I tell if I got a headshot without observing the screen and changing the outcome?
I thought that quantum computers were not real. Aren't they incredible at password cracking?
[QUOTE=Desuh;40606920]I thought that quantum computers were not real. Aren't they incredible at password cracking?[/QUOTE]
Quantum computers are incredible at factoring prime numbers very quickly. Encryption deals with this, so yes they are great at password cracking. Incidentally, as far as I know quantum computers are good at virtually nothing else.
[url]http://en.wikipedia.org/wiki/Shor%27s_algorithm[/url]
[QUOTE=gman003-main;40605929]
While I won't rule out quantum computers ever being common consumer products, I can't say I expect that to happen within the next 50 years or so. They just aren't useful for many of the things you use a computer for.[/QUOTE]
Thing is, you're only looking on current known uses for a quantum computer.
People used to say private computers would probably be unnecessary at all, because (at the time) they had no practical commercial use for private individuals.
It's over, buttcoins are finished.
[QUOTE=Glorbo;40606952]Thing is, you're only looking on current known uses for a quantum computer.
People used to say private computers would probably be unnecessary at all, because (at the time) they had no practical commercial use for private individuals.[/QUOTE]
he just said that
They make custom chips for specific tasks. Why not bitcoin mining? You would be rich in week.
[QUOTE=SGTNAPALM;40606949]Quantum computers are incredible at [B]factoring prime numbers[/B] very quickly. Encryption deals with this, so yes they are great at password cracking. Incidentally, as far as I know quantum computers are good at virtually nothing else.
[URL]http://en.wikipedia.org/wiki/Shor's_algorithm[/URL][/QUOTE]
sorry this is upsetting me
you mean finding prime factors right?
[editline]11th May 2013[/editline]
[QUOTE=Reader;40606974]They make custom chips for specific tasks. Why not bitcoin mining? You would be rich in week.[/QUOTE]
because you don't need a quantum ASIC for bitcoin mining?
you can use a normal custom chip for that job and they already exist
also you'd just crash the market if anyone knew your quantum bitcoin ASIC existed (which would be fucking hilarious and totally worth it)
[QUOTE=BrainDeath;40606987]
because you don't need a quantum ASIC for bitcoin mining?
you can use a normal custom chip for that job and they already exist[/QUOTE]
Current ASICs work like normal chips do, they don't take advantage of quantum entanglement or whatever it is.
Wouldn't quantum be much faster?
[QUOTE=BrainDeath;40606987]sorry this is upsetting me
you mean finding prime factors right?
[/QUOTE]
I don't know.
[QUOTE=BrainDeath;40606987]sorry this is upsetting me
you mean finding prime factors right?
[/QUOTE]
yea finding prime numbers is easy peezy
2...there, i found one!
prime factorization is a bit tougher(hence its usage in encryption).
Sorry, you need to Log In to post a reply to this thread.
