Feeds

back to article Quantum cruncher beats today's computers by 1080

An international team of scientists has created a quantum device with “the potential to perform calculations that would require a supercomputer larger than the size of the known universe.” So says the University of Sydney's Dr Michael Biercuk of the University's School of Physics, the Australian Research Council's Centre of …

COMMENTS

This topic is closed for new posts.
Anonymous Coward

"While very exciting, Bieruck also admits that no known computer can check the accuracy of calculations performed with the new quantum simulator."

Since there are so many supercomputers in the world, why not take work they have done and asked this computer to do the same task? Given how much faster it should be, it shouldn't take long for the results to come in. if it can manage to do the knowns accurately, then the unknowns should be accurate as well.

9
0
Silver badge

The impression I get from the paper's abstract is that it's performing computations that aren't practical to verify with current conventional computing technology. It doesn't appear to be a conventional general purpose digital computer.

0
0
Silver badge

Oh. So you need a bigger one of it to check it?

Research funds, lads!

1
0
Bronze badge
Black Helicopters

*Shifty eyes*

Sorry guv, doesn't work like that, but you know that unbreakable crypto you're using?, yeah that's all fucked now.

1
0
Silver badge

Exactly right

Quantum computing is theoretically useful for some applications. Pretty pointless for others.

0
1
Paris Hilton

doesn't really make sense

surely there are lots of NP-complete and CoNP-complete problems they could throw at it and verify the solutions in polynomial time on a standard machine?

2
1
Silver badge

Re: Exactly right

Quantum computers can solve certain types of problem (factorisation being the best-known) that are inaccessible to standard computers. So it won't run Crysis 3 (sorry).

But I don't see why they couldn't use it to 'instantly' factorise a 90-digit number created by multiplying two 45-digit primes. For comparison, the largest such number factorised by conventional computers was 768-bit (231-digits) and would have taken 2,000 years on a conventional desktop.

3
0
Silver badge
Holmes

Re: doesn't really make sense

NP-complete problems?

Quantum computers are most likely as shite at solving these in polynomial time as are classical ones.

--> http://qwiki.stanford.edu/index.php/Complexity_Zoo:B#bqp

2
0
FAIL

Its Deep Thought all over again.

1
0
Silver badge
Thumb Up

Yup, all I thought was "You won't like it"

(yeah well, the best laid plans of mice and all that )

0
0
Bronze badge

Re: Exactly right

So you're saying my dreams of running half-life at 50x10exp80 FPS are doomed ?

1
0
FAIL

Re: doesn't really make sense

contrary on the popular belief: the quantum computers cannot solve NP-complete problem. They can solve NP-incomplete, though.

0
0
Trollface

Re: Exactly right

"my dreams of running half-life at 50x10exp80 FPS are doomed ?"

Dude, Half Life came out in 1998 or something. My -laptop- can run it at 50x10exp80 FPS.

0
0
Coat

Creamy rice pudding and income tax

But when it says "42", how do you know that's the right answer? Especially when you weren't very clear on the question...

Could somebody hand me that cage with my pet mice?

14
0
Anonymous Coward

"Bieruck also admits that no known computer can check the accuracy of calculations performed with the new quantum simulator."

Based on this, one would assume that this computer is useful for phenomenal accuracy rather than phenomenal speed. As far as I know, however, that's really quite useless - I don't recall significant digits to be a serious limiting factor in most computing problems.

And if that's not it, well - if you're doing fluid dynamics or weather simulation, the idea of checking accuracy is meaningless to begin with.

What gives?

0
0
Boffin

I'm not sure there are any quantum algorithms developed yet which use floating-point. Quantum computation is all about integers and set theory, at this stage. When they say "accuracy" I think they mean correctness, not floating-point precision. That would be correctness as in not broken, plus I believe quantum computers always have some probabilistic chance of giving the wrong answer; that probability is also a kind of accuracy (level of confidence).

For brokenness the problem seems easy - try a bunch of NP problems and verify them (in P) on a classical computer. But the result is a mixture of brokenness and probabilistic accuracy. You then have to figure out what proportion is brokenness.

So I believe the issue is that they really need to simulate the quantum computer on a classical computer and calculate what the probabilistic accuracy should be, so they can cross-check with the empirical figures and see what the brokenness level is. But a 300-qbit computer is so large that it can't be simulated.

1
0
Anonymous Coward

"But a 300-qbit computer is so large that it can't be simulated."

Well, so, get a good idea and then just start using it. If the hurricane goes north rather than south, and your building falls over, and the enemy renegs on the treaty and invades, its brokenness is too high!

0
0
Joke

I know this is old but.....

I have to ask it:

Can it run Crysis?

3
0
Anonymous Coward

Re: I know this is old but.....

Maybe on medium settings

0
0

This post has been deleted by a moderator

Anonymous Coward

Here I am..

Brain the size of a Universe, only to get beaten by some quantum pocket calculator

8
0
Silver badge
Coat

Re: Here I am..

The evolution of a brain the size of the universe would first require the evolution of a reproductive tract the size of the universe, and nobody wants that.

0
0

Re: Here I am..

a reproductive tract the size of the universe

George Osborne?

5
0
Silver badge

Re: Here I am..

Your brain is about 1/4 its eventual size at birth so not so much need for trouser supplements.

0
0
Anonymous Coward

Your mother

It was like throwing a hotdog down a hall^H^H^H^Hsupermassive black hole.

0
0
Silver badge

Re: Your mother

Nah they showed how it works in the animation in 'Meaning of Life'.

It's El Ron's Thetans at it again!

0
0
Silver badge

Re: Here I am..

Your brain is about 1/4 its eventual size at birth so not so much need for trouser supplements.

Quite right, a foolish oversight on my part. I was clearly dumbfounded by the concept of a universe-magnitude reproductive tract...

0
0
Mushroom

oops

Here goes our simulation/matrix theory. If it works, it means our universe is not simulated, or else the simulating machine will crash...

WAIT, DON'T RU

2
0

Wally got there first

http://www.dilbert.com/2012-04-17/

2
0
Anonymous Coward

"The quantum device in question uses a pancake-shaped crystal of beryllium ions – 300 to be precise – that Bieruck says the team built “from scratch, atom by atom” with a told ABC radio that the apparatus required to capture the crystallise the atoms uses lasers, pumps and vacuum chambers to do the job, but that the rig needs occupies only a single room."

jeeeezussss - can't anyone @ el reg form a proper sentence ?

6
0

This post has been deleted by a moderator

This post has been deleted by a moderator

Silver badge
Boffin

Headlines 2013

Dr Michael Biercuk wins the lottery five times in a row, banned from buying any more tickets.

4
0
Anonymous Coward

Yawn...

"has computational potential" ... translation: "hasn't *actually* done any computation yet"

"no known computer can check the accuracy of calculations" ... translation: "is unable to do any computation which might be useful"

Useful work would include factorisation of products of large primes. That's trivial to verify, just multiply the answers together to see if you get the question.

1
2
Silver badge

Re: Yawn...

That would have been my assumption, but doubt that it has escaped the brains in question. Therefore I suspect that I have missed something- very likely, since I haven't delved in the link provided by the article.

1
0
Silver badge

Re: Yawn...

I don't have access to the full article, but the precis says that this is a quantum simulator that can be used to model sets of quantum particles beyond the range of current computation (which maxes out at about 30 particles). So it's not a general quantum computer, more a quantum analogue computer.

1
0

This sentence should be taken out and shot!

The quantum device in question uses a pancake-shaped crystal of beryllium ions – 300 to be precise – that Bieruck says the team built “from scratch, atom by atom” with a told ABC radio that the apparatus required to capture the crystallise the atoms uses lasers, pumps and vacuum chambers to do the job, but that the rig needs occupies only a single room.

0
0

Re: This sentence should be taken out and shot!

It makes more sense if you read it in the voice of Borat.

3
1
WTF?

They haven't built a "computer" by any ordinary definition.

What they've built is an analogue simulator. It uses quantum behaviour on a small and manageable scale - the disc of beryllium - to model the quantum behaviour of much more intractable systems. It's like modelling tsunamis in the kitchen and calling your sink a "computer".

I'm not sure at what point between the Nature paper and the Reg article this became so completely befuddled. It's not rocket science.

Well ... ok ... it is slightly rocket science. But somebody with a background in physics and an ability to read even just the abstract of the Nature paper could have grasped the vague idea of what they've done.

3
0
Silver badge
Childcatcher

Re: They haven't built a "computer" by any ordinary definition.

Well, the guy in the video says:

"If you look at the most powerful supercomputers today, we have beaten their performance, we have beaten what they can do".

NOPE!

Well, what they say is not exactly a lie. It's kinda half-true. Damn young ones with their steamy marketing messages.

1
0

Re: They haven't built a "computer" by any ordinary definition.

Comparing performance is not really the same as calling it a computer.

It's like bragging about the power of a jet engine by comparing it to "the most powerful internal combustion engine".

Bill Hicks was right about marketing.

0
1

Re: They haven't built a "computer" by any ordinary definition.

I just had a bizarre mental image of a jet-engine that was not also an internal-combustion engine. SteamPunk Harriers, anyone?

0
0
Facepalm

Re: They haven't built a "computer" by any ordinary definition.

@Mike 16

Ah yes. Sorry. I dun a stoopid. What I meant to compare was a piston engine and a jet engine. I'll go and sit in the herp derp corner along with whichever Reg editor wrote the title of this article.

0
0

I guess I don't quite get what they've actually built here. Is it really an equivalent to a silicon wafer? Or...?

But anyway the most important discovery here may just be the significant but unnamed eye-candy assisting Dr. Biercuk. Woo-hoo! :)

0
0
Gold badge
FAIL

Tense

The title and sub-title both refer to the capabilities of such a beast using the present tense.

Then the very first paragraph goes an spoils it all by saying "potentially".

We all know that "potentially" means "not" in this context.

A bit of an El Reg Fail if you ask me.

0
0
Bronze badge

Re: Tense

"A bit of an El Reg Fail"

Still, at least it is a well written article with no garbled sentences at all.

1
0

What they've really built...

It's an ice-breaker at parties.

"What's that?"

"It's my new quantum computer."

"What does it do?"

Ten hours later...

"Good morning! Join me for breakfast?"

0
0
Anonymous Coward

Re: Ice breaker at parties

But you first have to give it a really strong, hot cup of tea.

0
0
This topic is closed for new posts.