If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Wired)   Quantum computer has finally been proven to be real. Maybe. The box in the picture looks awfully large for just one cat, though   (wired.com) divider line 43
    More: Followup, cats, quantum, quantum superpositions, Google, Daniel Lidar, quantum computer, classical physics, qubits  
•       •       •

3778 clicks; posted to Geek » on 28 Jun 2013 at 11:29 AM (41 weeks ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



43 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread
 
2013-06-28 09:45:07 AM
FTA:

The rub is building even a single qubit is difficult. When you look at - i.e. read information from - a quantum system, it decoheres. In other words, it turns into an ordinary bit capable of holding only a single value. It ceases to be a quantum computer.

We have yet to be able to counter the observer effect. How in the fark are we supposed to build a quantum computer when we can't even handle that?
 
2013-06-28 10:16:41 AM
Install linux.
 
2013-06-28 10:18:27 AM

xanadian: Install linux.


Not everyone is an ubergeek who are willing to drop to the terminal to do anything useful.
 
2013-06-28 11:04:22 AM
If Google has one, the NSA probably can get their hands on one (or has already had one for a couple years).

512 qbits would seem to imply any crypto system using fewer bits than that for the key is now breakable, even without resorting to rubber-hose cryptanalysis.
 
2013-06-28 11:47:42 AM
The box in the picture looks awfully large for just one cat, though

Well, duh!.  Everything is multicore these days.
 
2013-06-28 11:51:46 AM
Anyone remember the Starbridge Systems FPGA computer hoax from a few years back? This doesn't sound all that different...
 
2013-06-28 11:55:11 AM

abb3w: 512 qbits would seem to imply any crypto system using fewer bits than that for the key is now breakable, even without resorting to rubber-hose cryptanalysis.


The chip isn't programmable, can only solve one specially-designed problem, is only capable of approximations, not exact solutions, and has less processing power than a Xeon Phi.
 
2013-06-28 11:56:26 AM
Reminds me of something...

harkuchh.com
 
2013-06-28 11:58:31 AM

cman: FTA:

The rub is building even a single qubit is difficult. When you look at - i.e. read information from - a quantum system, it decoheres. In other words, it turns into an ordinary bit capable of holding only a single value. It ceases to be a quantum computer.

We have yet to be able to counter the observer effect. How in the fark are we supposed to build a quantum computer when we can't even handle that?


If the entire computation is performed when thinking of the equation as a wave instead of particles (i.e. bits) the uncertainty principle doesn't come into play. It only happens when you read the result. We won'
t be able to fully advance in the field until we can model and know what goes on as a wave function.

Consider, if a water particle hits an obstruction, where does it go: to the left or right? If you think of the problem as a wave, you don't have to know which way the particle went. The particle went whichever way it was most likely to go. This is akin to the electron double slit experiment where the exact same apparatus will result in two completely different outcomes only depending on whether or not you look at which slit the electrons went through. If we used electron waves, for instance, to do computations we would need to be able to read the diffraction pattern to know the result of the computation. Directing the flow of quantum data is where it's at.
 
2013-06-28 12:08:27 PM

Stibium: cman: FTA:

The rub is building even a single qubit is difficult. When you look at - i.e. read information from - a quantum system, it decoheres. In other words, it turns into an ordinary bit capable of holding only a single value. It ceases to be a quantum computer.

We have yet to be able to counter the observer effect. How in the fark are we supposed to build a quantum computer when we can't even handle that?

If the entire computation is performed when thinking of the equation as a wave instead of particles (i.e. bits) the uncertainty principle doesn't come into play. It only happens when you read the result. We won'
t be able to fully advance in the field until we can model and know what goes on as a wave function.

Consider, if a water particle hits an obstruction, where does it go: to the left or right? If you think of the problem as a wave, you don't have to know which way the particle went. The particle went whichever way it was most likely to go. This is akin to the electron double slit experiment where the exact same apparatus will result in two completely different outcomes only depending on whether or not you look at which slit the electrons went through. If we used electron waves, for instance, to do computations we would need to be able to read the diffraction pattern to know the result of the computation. Directing the flow of quantum data is where it's at.


static.giantbomb.com
 
2013-06-28 12:17:58 PM
near as i can tell.  quantum computers may, or may not, actually work.
 
2013-06-28 12:20:10 PM
I think that's the box they locked the Doctor in.
 
2013-06-28 12:27:20 PM
The Quantum computer is designed to automatically troll 3-d printing and space threads with uncanny efficiency.
 
2013-06-28 12:29:13 PM

Copper Spork: abb3w: 512 qbits would seem to imply any crypto system using fewer bits than that for the key is now breakable, even without resorting to rubber-hose cryptanalysis.

The chip isn't programmable, can only solve one specially-designed problem, is only capable of approximations, not exact solutions, and has less processing power than a Xeon Phi.


Only integers and ratios give you exact solutions. Floating point arithmetic is nothing but an approximation, and it's by far the most used type of number in data processing.

But as far as programability, we went from punch cards and vacuum tubes to transistors and EPROMs in only a few decades. This is only a test case of a quantum circuit that does one thing. Once we get a better grasp on how to build them, feed them data, and interpret the result we can know how to link new circuits together. Eventually you can get enough circuits together to form a processor not unlike modern chips, which are nothing but a bunch of different specialized binary circuits linked together.
 
2013-06-28 12:37:28 PM
Anyone know and/or can conceive of the possible future implications/uses regarding the day there are full-blown quantum supercomputers?

For example, is it conceivable that so much data could be crunched that the future/"probable outcomes" of yet to occur events may be predicted to a close proximity?
 
2013-06-28 12:46:24 PM

Stibium: Only integers and ratios give you exact solutions. Floating point arithmetic is nothing but an approximation, and it's by far the most used type of number in data processing.


No, not that kind of "approximation" (and incidentally, floating point is exact, but not precise). The "it can't guarantee it's given you an optimal solution" kind of approximation. It can't, for example, give you the best solution to a travelling salesman problem. It can only give you a probably-quite-good solution.
 
2013-06-28 12:55:32 PM

Copper Spork: Floating point ... it's by far the most used type of number in data processing.


The word you're looking for there is decimal as in BigDecimal, if you're using floats or doubles, you've already killed us all.
 
2013-06-28 12:58:07 PM

Any Pie Left: The Quantum computer is designed to automatically troll 3-d printing and space threads with uncanny efficiency.


Nice.
 
2013-06-28 01:02:38 PM

Copper Spork: Stibium: Only integers and ratios give you exact solutions. Floating point arithmetic is nothing but an approximation, and it's by far the most used type of number in data processing.

No, not that kind of "approximation" (and incidentally, floating point is exact, but not precise). The "it can't guarantee it's given you an optimal solution" kind of approximation. It can't, for example, give you the best solution to a travelling salesman problem. It can only give you a probably-quite-good solution.


So maybe you make 400 of them work the travelling salesman job in parallel and set one to thinking about which of the 400 is going to provide the best result.
 
2013-06-28 01:03:15 PM

Any Pie Left: The Quantum computer is designed to automatically troll 3-d printing and space threads with uncanny efficiency.


I may or may not get that.

Stibium: But as far as programability, we went from punch cards and vacuum tubes to transistors and EPROMs in only a few decades


But once we're at the atom by atom level, you'll never see that kind of progress again.

http://www.youtube.com/watch?feature=player_embedded&v=NGFhc8R_uO4
 
2013-06-28 01:03:41 PM
fark quantum cats bullshiat.

And fark Mike deGrassi High School Tyson.
 
2013-06-28 01:08:51 PM
www.angelfire.lycos.com

"Ziggy says there's a 68% chance quantum computing will lead to Skynet."
 
2013-06-28 01:11:28 PM
freesoupwithpurchase.files.wordpress.com
"Ziggy also says there's a 100% chance you'll be in Star Trek: Enterprise."
 
2013-06-28 01:23:51 PM

neongoats: So maybe you make 400 of them work the travelling salesman job in parallel and set one to thinking about which of the 400 is going to provide the best result.


So long as you don't mind the solution potentially being arbitrarily wrong, sure. But if all you're looking for is a usually-decent approximation, classical methods work just fine.
 
2013-06-28 01:27:05 PM
The only reason why the Internet fawns all over Neil deGrassi High Tyson is because he's black. The Internet thinks this is some sort of karmic perfection: black astronomer.

I SAY fark ALL THE farkING farkERS
 
2013-06-28 01:38:21 PM

heinrich66: The only reason why the Internet fawns all over Neil deGrassi High Tyson is because he's black. The Internet thinks this is some sort of karmic perfection: black astronomer.

I SAY fark ALL THE farkING farkERS


... No? From the mild exposure I've had, I like him because he's entertaining, and because he's good at something I aspire to and would love to be able to do for a living: Distilling his scientific knowledge and often explaining it in ways the layman can understand.

/And without also bandying about theories like string theory and many-universe theory as though they are accepted scientific fact like Brian Greene seems to.
 
2013-06-28 02:01:28 PM
I really hope we aren't in one of the half of the alternate quantum universes where the answer to "do quantum computers work" is "no"
 
2013-06-28 02:03:23 PM
The Observer Effect is a feature not a bug. Its not something a cryptographer would really want to solve. I hear that there is a quantum network based on star topology (anyone remember token ring?) that allows each node to see the results once.

The real use for these machines will be for open ended dynamic computations with really large numbers. IE: How many sodium atoms are in the sea?
How much water is there in the Universe?
What is the fusion ratio at the core of the Sun?

You will not be able to send email or play video games with a quantum computer.
 
2013-06-28 02:16:50 PM
I came for cat pictures

Left disappointed
 
2013-06-28 02:56:36 PM

Heraclitus: The Observer Effect is a feature not a bug. Its not something a cryptographer would really want to solve. I hear that there is a quantum network based on star topology (anyone remember token ring?) that allows each node to see the results once.

The real use for these machines will be for open ended dynamic computations with really large numbers. IE: How many sodium atoms are in the sea?
How much water is there in the Universe?
What is the fusion ratio at the core of the Sun?

You will not be able to send email or play video games with a quantum computer.


Unless someone develops a video game in which you need to catch all the sodium atoms in the sea.

/Which I am.
/Which I am not.
 
2013-06-28 03:05:47 PM
512 qubits?  Wow.  That's 2512 normal bits worth of information.  That's about 1.34 * 10154.  That's many, many times larger than the estimated number of all the atoms in the universe (1080), which means if you wanted to build a hard drive that could store all the potential information in 512 qubits as bits, you'd need all the matter in the universe about 1074 times over if you wanted to store the individual bits as individual atoms of two different types.
 
2013-06-28 03:07:53 PM
Felgraf:.

/And without also bandying about theories like string theory and many-universe theory as though they are accepted scientific fact like Brian Greene seems to.


You're nothing but a black apologist.
 
2013-06-28 03:15:14 PM

Any Pie Left: The Quantum computer is designed to automatically troll 3-d printing and space threads with uncanny efficiency.




LOL

/this is why I love fark
 
2013-06-28 03:28:44 PM

heinrich66: Felgraf:.

/And without also bandying about theories like string theory and many-universe theory as though they are accepted scientific fact like Brian Greene seems to.

You're nothing but a black apologist.


... Uh.

I'm not really sure how that specific part you quoted makes me a black apologist, as opposed to an experimental physicist who eyes String Theory with skepticism, but I'm going to back away slowly now, because you are clearly a crazy person.
 
2013-06-28 03:32:36 PM

heinrich66: Felgraf:.

/And without also bandying about theories like string theory and many-universe theory as though they are accepted scientific fact like Brian Greene seems to.

You're nothing but a black apologist.


And you're a really poor troll. This is a bookmark for when I'm off mobile- I need to do some housekeeping.
 
2013-06-28 03:38:16 PM

Copper Spork: Stibium: Only integers and ratios give you exact solutions. Floating point arithmetic is nothing but an approximation, and it's by far the most used type of number in data processing.

No, not that kind of "approximation" (and incidentally, floating point is exact, but not precise). The "it can't guarantee it's given you an optimal solution" kind of approximation. It can't, for example, give you the best solution to a travelling salesman problem. It can only give you a probably-quite-good solution.


Neither can regular computers...
 
2013-06-28 03:52:12 PM

Telos: No, not that kind of "approximation" (and incidentally, floating point is exact, but not precise). The "it can't guarantee it's given you an optimal solution" kind of approximation. It can't, for example, give you the best solution to a travelling salesman problem. It can only give you a probably-quite-good solution.

Neither can regular computers...


Sure they can. NP-hard problems are computable.
 
2013-06-28 04:02:18 PM

wingnut396: The box in the picture looks awfully large for just one cat, though

Well, duh!.  Everything is multicore these days.


I was told it was all ball bearings.
 
2013-06-28 04:17:05 PM

Quantum Apostrophe: Any Pie Left: The Quantum computer is designed to automatically troll 3-d printing and space threads with uncanny efficiency.

I may or may not get that.

Stibium: But as far as programability, we went from punch cards and vacuum tubes to transistors and EPROMs in only a few decades

But once we're at the atom by atom level, you'll never see that kind of progress again.

http://www.youtube.com/watch?feature=player_embedded&v=NGFhc8R_uO4


We will then start having different models of computations, much like this quantum computer. Right processing power has gotten so far the bottlenecks are in storage and transportation of information. Processors and primary cache as getting fast enough to not matter, getting the information to them on the other hand.....
 
2013-06-28 05:11:15 PM

LrdPhoenix: 512 qubits?  Wow.  That's 2512 normal bits worth of information.  That's about 1.34 * 10154.  That's many, many times larger than the estimated number of all the atoms in the universe (1080), which means if you wanted to build a hard drive that could store all the potential information in 512 qubits as bits, you'd need all the matter in the universe about 1074 times over if you wanted to store the individual bits as individual atoms of two different types.


That's still not enough for Drew's porn collection.
 
2013-06-28 06:03:41 PM
I'll believe it when I see it.  Wake me when someone builds a 2048-qubit quantum computer; the original Xbox's public key and I have some unfinished business.
 
2013-06-28 06:29:40 PM

LrdPhoenix: 512 qubits?  Wow.  ...


img.gawkerassets.com
/ I'd much rather see 512 q*berts... stupid snakes and gremlins wouldn't have an f-ing  chance.
 
2013-06-29 09:19:52 AM
Yeah yeah. More extraordinary claims without evidence.
 
Displayed 43 of 43 comments

View Voting Results: Smartest and Funniest


This thread is closed to new comments.

Continue Farking
Submit a Link »






Report