Hard Light Productions Forums
Off-Topic Discussion => General Discussion => Topic started by: jr2 on June 02, 2014, 01:53:02 pm
-
http://gizmodo.com/new-method-of-quantum-teleportation-could-bring-us-a-qu-1583771236
-
The obligatory preminder that entanglement transmission does not allow information to move faster than light
-
TBH, this particular application is more interesting in that it's a step towards quantum computing being viable. "Quantum internet" is a distant possibility (in fact, entanglement could allow essentially fusing all the world's quantum computers into a single, mighty computing machine. Think quantum-based, world-wide Grid), but first of all, this has awesome implications regarding building a quantum computer in first place.
-
in fact, entanglement could allow essentially fusing all the world's quantum computers into a single, mighty computing machine
lol
-
Well, you can already sort of do that - there are systems that allow you to essentially use the entire internet (or at least, parts of it that agree to it) to perform a calculation. There are data analysis projects that use this. Now imagine that, except with infinite bandwidth and transmission speeds equal to speed of light. Of course, noone in the right mind would entangle all the computers they produce, but unless there's some bottleneck I don't know of, I'd expect a huge increase in this kind of computing. Quantum transmission is fast.
-
TBH, this particular application is more interesting in that it's a step towards quantum computing being viable. "Quantum internet" is a distant possibility (in fact, entanglement could allow essentially fusing all the world's quantum computers into a single, mighty computing machine. Think quantum-based, world-wide Grid), but first of all, this has awesome implications regarding building a quantum computer in first place.
I am not an expert of quantum, but this basically eliminates the problem that quantums are both 100% one thing and 100% some other thing, yes?
-
Nah, most of the stuff being said in this thread is bull****. Quantum computing won't move at the speed of light unless the classical channel moves at the speed of light (at least if we're talking quantum transmission...), and the bandwidth won't be remotely infinite. Quantum computers are really good at some classes of problems (they will completely **** public key encryption, for example), but they aren't strictly superior to existing classical computers in all domains, and they come with their own set of limitations.
source: I am a quantum computer
-
*decoheres into ****tinesss*
-
(in fact, entanglement could allow essentially fusing all the world's quantum computers into a single, mighty computing machine. Think quantum-based, world-wide Grid)
Skynet!
-
Nah, most of the stuff being said in this thread is bull****. Quantum computing won't move at the speed of light unless the classical channel moves at the speed of light (at least if we're talking quantum transmission...), and the bandwidth won't be remotely infinite.
Unless. :) Remember, the other major thing in computer development are light-based electronics (since we're pretty much at the limit of electron-based ones). As such, I'd expect classical channel to be done through light, which, by it's nature, moves at the speed of light. In fact, I'd imagine light-based classical channel will go perfectly with quantum-based communication. And while bandwidth is not going to be truly infinite, it's probably still going to be astronomical. I'm not sure on exact figures, but it should be possible to make it high enough that it is not a bottleneck anymore. It depends on how good they can make the classical channel, I suppose. As far as I'm concerned, the only BS in this thread is your attempt at calling BS. :)
Also, while quantum computers are indeed limited in certain problems, there's nothing stopping you from having a classic CPU handle stuff quantum one can't (or isn't very good at). In fact, that's how I expect the future computers to work, with each part doing what it's best at, and quantum communication ensuring they're always in sync. Of course, it's not like gaming will be the reason to use quantum computers. Cryptography, scientific calculations and a number of other applications would benefit the most from this invention. In fact, scientific research is why being able to link quantum computers together with very high bandwidth comms is so interesting. That's where this performance will really matter, and that's where the problems that could use quantum computing are. Well, that, and designing cryptographic algorithms that would take millennia to break on a classic computer (I sincerely hope this is done before a bunch of shady characters gets their hands on one of those things).
-
Well, that, and designing cryptographic algorithms that would take millennia to break on a classic computer (I sincerely hope this is done before a bunch of shady characters gets their hands on one of those things).
You know that a "classic" computer is a moving target, right? Also, if you assume a naive application of Moore's law in regards to performance of "classic" computers, you can just use existing algorithms with an increased key size, to guess what a computer X years from now will take Y years to break.
Regarding quantum resistant cryptography algorithms, we have had those much longer than non-resistant ones. They're called symmetric key algorithms. And while they don't have as many interesting applications as asymmetric ones, they are used almost everywhere.
-
No, my bull**** call is straight on; you're just taking it too personally. About three quarters of the stuff you said was accurate, and a quarter of it needs to be chiseled back to a reasonable position with sane qualifications.
infinite bandwidth
Bull****. Information theory provides ceilings on bandwidth. (You've correctly recognized that this is bull****, and stepped back from the claim.)
Remember, the other major thing in computer development are light-based electronics (since we're pretty much at the limit of electron-based ones). As such, I'd expect classical channel to be done through light, which, by it's nature, moves at the speed of light. In fact, I'd imagine light-based classical channel will go perfectly with quantum-based communication. A
Here you're conflating two developments: optical communication networks (which we've had for ages, they're fiber optics) and optical computation. You fail to successfully identify the major challenges facing optical computation - shot noise and the weak coupling capabilities of photons compared to electrons. Compounding the error, you're 'imagining' that optical computation will be a great partner for a photon-mediated electron spin entanglement without addressing the OEO problem, which seems like it could be pretty necessary. You need to be more rigorous about delineating communication problems and computation problems.
In all your excitement over global computational meshes firing off Shor's algorithm, you're losing focus on what's really exciting here - which you correctly identified in the very first sentence you posted! One of the biggest hurdles in QC is the problem of quantum error. By achieving full determinism, these researchers have made a huge hurdle towards a scalable quantum processor, which is necessary before we can build useful quantum computers (let alone start networking them). And if the diamond trap substrate is robust and economical, then that would be (finally) a reasonable consensus architecture to start from.
That's what's super cool here, and what you're right to be excited about.
Well, that, and designing cryptographic algorithms that would take millennia to break on a classic computer (I sincerely hope this is done before a bunch of shady characters gets their hands on one of those things).
Designing useful quantum algorithms is one of the major steps that still needs to be taken before QC becomes broadly applicable. However, both the algorithms necessary for QC to factor public-key encryption and the quantum encryption techniques that might make a usable replacement for RSA have been understood mathematically for...at least a decade, I believe.
You know that a "classic" computer is a moving target, right? Also, if you assume a naive application of Moore's law in regards to performance of "classic" computers, you can just use existing algorithms with an increased key size, to guess what a computer X years from now will take Y years to break.
Regarding quantum resistant cryptography algorithms, we have had those much longer than non-resistant ones. They're called symmetric key algorithms. And while they don't have as many interesting applications as asymmetric ones, they are used almost everywhere.
There are some classically difficult algorithms that a quantum computer will munch through - BQP stuff, including factorization. But the majority of algorithms aren't any faster on a quantum computer. So as you said, we've had Shor-resistant security algorithms for a while. One of my computational physics friends pointed me to this website (http://pqcrypto.org/) a while back as a good read.
-
Well, that, and designing cryptographic algorithms that would take millennia to break on a classic computer (I sincerely hope this is done before a bunch of shady characters gets their hands on one of those things).
You know that a "classic" computer is a moving target, right? Also, if you assume a naive application of Moore's law in regards to performance of "classic" computers, you can just use existing algorithms with an increased key size, to guess what a computer X years from now will take Y years to break.
ooh ooh i'm not sure this is even relevant but i love this quote so
One of the consequences of the second law of thermodynamics is that a certain amount of energy is necessary to represent information. To record a single bit by changing the state of a system requires an amount of energy no less than kT, where T is the absolute temperature of the system and k is the Boltzman constant. (Stick with me; the physics lesson is almost over.)
Given that k = 1.38×10-16 erg/°Kelvin, and that the ambient temperature of the universe is 3.2°Kelvin, an ideal computer running at 3.2°K would consume 4.4×10-16 ergs every time it set or cleared a bit. To run a computer any colder than the cosmic background radiation would require extra energy to run a heat pump.
Now, the annual energy output of our sun is about 1.21×1041 ergs. This is enough to power about 2.7×1056 single bit changes on our ideal computer; enough state changes to put a 187-bit counter through all its values. If we built a Dyson sphere around the sun and captured all its energy for 32 years, without any loss, we could power a computer to count up to 2192. Of course, it wouldn't have the energy left over to perform any useful calculations with this counter.
But that's just one star, and a measly one at that. A typical supernova releases something like 1051 ergs. (About a hundred times as much energy would be released in the form of neutrinos, but let them go for now.) If all of this energy could be channeled into a single orgy of computation, a 219-bit counter could be cycled through all of its states.
These numbers have nothing to do with the technology of the devices; they are the maximums that thermodynamics will allow. And they strongly imply that brute-force attacks against 256-bit keys will be infeasible until computers are built from something other than matter and occupy something other than space.
-
This is great, but you should include some formatting corrections (superscript didn't translate from wherever you got this):
One of the consequences of the second law of thermodynamics is that a certain amount of energy is necessary to represent information. To record a single bit by changing the state of a system requires an amount of energy no less than kT, where T is the absolute temperature of the system and k is the Boltzman constant. (Stick with me; the physics lesson is almost over.)
Given that k = 1.38×10-16 erg/°Kelvin, and that the ambient temperature of the universe is 3.2°Kelvin, an ideal computer running at 3.2°K would consume 4.4×10^-16 ergs every time it set or cleared a bit. To run a computer any colder than the cosmic background radiation would require extra energy to run a heat pump.
Now, the annual energy output of our sun is about 1.21×1041 ergs. This is enough to power about 2.7×1056 single bit changes on our ideal computer; enough state changes to put a 187-bit counter through all its values. If we built a Dyson sphere around the sun and captured all its energy for 32 years, without any loss, we could power a computer to count up to 2192. Of course, it wouldn't have the energy left over to perform any useful calculations with this counter.
But that's just one star, and a measly one at that. A typical supernova releases something like 1051 ergs. (About a hundred times as much energy would be released in the form of neutrinos, but let them go for now.) If all of this energy could be channeled into a single orgy of computation, a 219-bit counter could be cycled through all of its states.
These numbers have nothing to do with the technology of the devices; they are the maximums that thermodynamics will allow. And they strongly imply that brute-force attacks against 256-bit keys will be infeasible until computers are built from something other than matter and occupy something other than space.
You should probably clarify that an erg is equal to 10-7 joules, too. That clears the numbers up, I hope. Still, those are theoretical limits, practical ones are, as always, lower. Classic computers are running into thermodynamic barriers right now. High-performance gaming systems, for instance, run into incredible heat dissipation problems. The latest Intel's CPU line, in particular. I recall it being not that much of an improvement, but a real devil to keep cool.
Also, there is an entire class of problems which could only be solved by quantum computing. They were called "P-NP problems" (at least IIRC), and this was mathematically proven. I don't know if you could design a cryptographic algorithm that would use a such a problem, but if you could, trial and error would probably be the only way to approach this kind of problem with a classical computer (and, as shown in the quote, brute-forcing this would be futile). Also, by "taking millennia to solve" I referred to "a classic computer would need to run for a millennia before solving it", not that someone wouldn't be able to break it anyway. Either way, quantum computing will someday shake cryptography up quite a bit, and would probably have some interesting influence on mathematics in general. It's not just a matter of Shor's algorithm and resistance to it (you don't need a QC to theorize about it and it's algorithms), quantum computers would open up completely new alleys in algorithm design.
No, my bull**** call is straight on; you're just taking it too personally. About three quarters of the stuff you said was accurate, and a quarter of it needs to be chiseled back to a reasonable position with sane qualifications.
infinite bandwidth
Bull****. Information theory provides ceilings on bandwidth. (You've correctly recognized that this is bull****, and stepped back from the claim.)
Every time I write "infinite" replace that with "infinite for all intents and purposes" (more than we could possibly use/not a bottleneck/not gonna run out anytime soon). In the same vein, "infinitely small" means "finitely small, but irrelevant to the calculations". :) Few things in the universe are truly infinite, but the word makes for a nice rhetoric (works well enough for laymen, at least). I should've been more specific.
Here you're conflating two developments: optical communication networks (which we've had for ages, they're fiber optics) and optical computation. You fail to successfully identify the major challenges facing optical computation - shot noise and the weak coupling capabilities of photons compared to electrons. Compounding the error, you're 'imagining' that optical computation will be a great partner for a photon-mediated electron spin entanglement without addressing the OEO problem, which seems like it could be pretty necessary. You need to be more rigorous about delineating communication problems and computation problems.
In all your excitement over global computational meshes firing off Shor's algorithm, you're losing focus on what's really exciting here - which you correctly identified in the very first sentence you posted! One of the biggest hurdles in QC is the problem of quantum error. By achieving full determinism, these researchers have made a huge hurdle towards a scalable quantum processor, which is necessary before we can build useful quantum computers (let alone start networking them). And if the diamond trap substrate is robust and economical, then that would be (finally) a reasonable consensus architecture to start from.
That's what's super cool here, and what you're right to be excited about.
Well, I thought the matter of implications for the very existence of an usable QC has been settled already. :) Yes, it's nice that we can actually make a quantum processor now. But I see little to discuss in that matter.
Other implications are a bit more far-fetched. I must say that I'm not that much into optics right now, but it's definitely a thing. We'll definitely have to go optical to overcome the aforementioned thermodynamic barriers. Oh, and I was talking about computation the whole time, which requires communication between computer's subsystems. Perhaps I wasn't clear, but I wasn't referring to bandwidth as in "internet bandwidth", but rather as bandwidth between computer subsystems (at least I think so. I posted that late at night). Ordinary electricity moves at the speed of light (according to one interpretation), and we've got fiber optics, so light-speed comms are nothing new in that field. As for the OEO problem, my whole idea was to sort of go around it by having an all-optical classic system directly interfacing with the quantum PMESE system. Again, I "imagine", because I'm not sure if that's possible. I should've stated that.
Or perhaps it was late at night and I got those two subject confused/tried to discuss them both at the same time. I've been sleep-deprived lately (in no small part thanks to learning physics). :)
-
Quantum computers can't provably solve NP-complete problems in polynomial time. NP-complete is probably disjoint from BQP, the set of problems quantum computers are really good at.
Shor's algorithm is a quantum algorithm, it requires a quantum computer to operate.
Electricity doesn't actually move at the speed of light in vacuum unless it's in a vacuum. This is not a huge problem for telecommunications, though, since we can get almost arbitrarily close to lightspeed in very modern fiber optics (merely vanilla modern fiber optics are limited to something around 70% lightspeed).
-
Don't tell me quantum computing won't bring Quantum Healing. Sure it will. Quantum healing is healing the bodymind from a quantum level. That means from a level which is not manifest at a sensory level. Our bodies ultimately are fields of information, intelligence and energy. Quantum healing involves a shift in the fields of energy information, so as to bring about a correction in an idea that has gone wrong. So quantum healing involves healing one mode of consciousness, mind, to bring about changes in another mode of consciousness, body.
-
imo, yeah
-
Quantum computers can't provably solve NP-complete problems in polynomial time. NP-complete is probably disjoint from BQP, the set of problems quantum computers are really good at.
Shor's algorithm is a quantum algorithm, it requires a quantum computer to operate.
Electricity doesn't actually move at the speed of light in vacuum unless it's in a vacuum. This is not a huge problem for telecommunications, though, since we can get almost arbitrarily close to lightspeed in very modern fiber optics (merely vanilla modern fiber optics are limited to something around 70% lightspeed).
Ah, of course. Well, technically it does move at the speed of light - just not in vacuum, but in conductor of your choice. :) Slower than c, but still a pretty brisk pace, and rather irrelevant on earthly distances. As for the maths, I've never been delving into this particular branch. I do recall reading about a subset of NP-complete problems (because in general, they can't be provably solved in polynomial time, IIRC, that's the whole deal with being NP-complete) that could be solvable with a quantum computer, but it was a long time ago and I don't remember it well. Probably had something to do with BQP (perhaps those sets are not disjoint?), but again, work on quantum computers wasn't nearly as advanced as it is now back when I read it. Theoretical mathematics isn't my strong suit. It's nice to dip in once in a while, but that's it. :)
-
Also, there is an entire class of problems which could only be solved by quantum computing. They were called "P-NP problems" (at least IIRC), and this was mathematically proven.
(http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-845-quantum-complexity-theory-fall-2010/6-845f10.jpg)
hth
-
Ah, this is the very diagram I was thinking about, thanks. NP-complete are indeed separate, and it's all a subset of NP problems. It seems that I've got NP and NP-complete mixed up. As I said, theoretical math isn't my strong suite. :)
-
Every time I hear Quantum, Relativity must be groaning XD.
Regardless, I'm not so hot on the "Quantum" bit with computers.
But Quantum Porn from the Quantum Internets? Now that's something to either behold or fear. XD
-
I'm gonna entangle you like crazy, come spinning while superpositioning on me, decohere my buckleballs, bring me to a glorious eigenstate faster than light.
-
Don't tell me quantum computing won't bring Quantum Healing. Sure it will. Quantum healing is healing the bodymind from a quantum level. That means from a level which is not manifest at a sensory level. Our bodies ultimately are fields of information, intelligence and energy. Quantum healing involves a shift in the fields of energy information, so as to bring about a correction in an idea that has gone wrong. So quantum healing involves healing one mode of consciousness, mind, to bring about changes in another mode of consciousness, body.
Like doctor who!
-
I'm gonna entangle you like crazy, come spinning while superpositioning on me, decohere my buckleballs, bring me to a glorious eigenstate faster than light.
Glorious.
-
The diagram has its problems.
A conventional computer is able to solve the most of these problems but it takes a lot of time.
We will reach the final stage of micro processor performance in the coming twenty years.
Because the processores are getting smaller and more comact so hall-effect and quantum tunneling become more and more a problem.
By the way the NASA already has a working quantum computer but it's not free programmable and has only 128 q-bits.