Quantum Computing: Exploring Speed Beyond Light Communication

[Johnz414]

Hi,

Does anyone remember a recent article on quantum computing where they describe incredibly fast computing, even faster than light communication. It describes this as a result of quark pairs being made at the same time. It says being made at the same time they come out as if they are relatives, very closely related and relating to each other very quickly. If one of the quarks is affected then the other quark seems to "know" this and know if faster than light communication. The article explains this holds promise for extremely fast computing for computers. It discribes that it should only take about 20 years to convert the find into technology. It was about 2 or 3 months ago that I read it and I'd like to find it again for reference. It was in something like an about.com article or somewhere like that. I did check about.com and I don't seem to be able to find it there. If anyone knows where to find this article I'd really appreciate finding it again. Thank you.

John :uhh:

 

[franznietzsche]

Hi,

Does anyone remember a recent article on quantum computing where they describe incredibly fast computing, even faster than light communication. It describes this as a result of quark pairs being made at the same time. It says being made at the same time they come out as if they are relatives, very closely related and relating to each other very quickly. If one of the quarks is affected then the other quark seems to "know" this and know if faster than light communication. The article explains this holds promise for extremely fast computing for computers. It discribes that it should only take about 20 years to convert the find into technology. It was about 2 or 3 months ago that I read it and I'd like to find it again for reference. It was in something like an about.com article or somewhere like that. I did check about.com and I don't seem to be able to find it there. If anyone knows where to find this article I'd really appreciate finding it again. Thank you.

John :uhh:

hmm, there are some suspicious claims in here that aren't quite right, pertaining to the ftl communication. What you're describing is basically quantum entanglement, however, IIRC it has been clearly demonstrated that you cannot actually transmit useful information by this (and there is some dispute in quantum mechanics whether its even possible at all, depending on whether the copenhagen interpretation of quantum mechanics is correct).

Quantum computing is not necessarily any faster than conventional computing in the normal sense of processing data faster (though it probably would be). The real advantage to quantum computing is that many basic computer science algorithms become much simpler if you aren't limited to binary. Qubits allow to operate in higher bases than base 2, and the resulting simplification of many basic algorithms is the real benefit that we can never get from traditional computing.

 

[Rach3]

As far as physics is known - relativity, QM, QFT - the idea of "FTL" communication is always prohibited and remains so. There are lots of subtleties with quantum entanglement, it is too often that popular sources misrepresent and overhype.

There is nothing in the works involving entangled quarks - we're working with much more manageable things like electrons or photons.

 

[Rach3]

Quantum computing is not necessarily any faster than conventional computing in the normal sense of processing data faster (though it probably would be).

Well, the target of "computing" is not really processing data as much as solving hard, NP problems. Of course no one really knows what it will do to traditational microcomputers and such, any more than Bardeen et. al. in 1947 could predict high-speed graphics is 2006.

The real advantage to quantum computing is that many basic computer science algorithms become much simpler if you aren't limited to binary.

:rofl: :rofl: :rofl: I wish!

 

[Rach3]

Mods - please move this to the QM physics forum.

 

[franznietzsche]

:rofl: :rofl: :rofl: I wish!

let me rephrase: They take fewer steps when you aren't limited to binary.

 

[franznietzsche]

Well, the target of "computing" is not really processing data as much as solving hard, NP problems. Of course no one really knows what it will do to traditational microcomputers and such, any more than Bardeen et. al. in 1947 could predict high-speed graphics is 2006.

Yes, but the normal layman's measure of computing speed is usually 'Ghz' or if they're a notch above the crowd, Gflops. Which is what I was referring to.

 

[Anttech]

There is so much we *dont* know, only the naive would rule it out.

Quantum computing is not necessarily any faster than conventional computing in the normal sense of processing data faster (though it probably would be). The real advantage to quantum computing is that many basic computer science algorithms become much simpler if you aren't limited to binary. Qubits allow to operate in higher bases than base 2, and the resulting simplification of many basic algorithms is the real benefit that we can never get from traditional computing.

Base 2 *is* very simple...

 

[-Job-]

Quantum computing is not necessarily any faster than conventional computing in the normal sense of processing data faster (though it probably would be).

A quantum computer is way faster than a traditional computer. The computational power of a quantum computer increases exponentially with the number of qubits, independently of a clock speed. This is particularly evident with NP-Complete problems. For instance, suppose you have a set of n numbers and you want to know if there is a subset of those numbers whose sum is X.
In a traditional, turing-equivalent, computer, an algorithm for solving this problem can, in the worst case, take 2^n operations. If the set has 1000 numbers we are looking at 2^1000 operations, an infinity for any practical purposes.
On the other hand, a quantum computer with 1000 qubits can solve this problem in an instant. The speed up is connected with quantum superposition. Whereas a traditional bit is either 0 or 1, a qubit can be 0, 1, or in a superposition of both 0 and 1. A set of 1000 qubits, therefore, is in a superposition of 2^1000 states. When a quantum computer performs an operation on n qubits, it's performing an operation on all the 2^n states at once.
This is the power of a quantum computer, and it fits really well with NP-Complete problems which currently have an exponential time upper bound.
Quantum computers won't break the exponential upper bound of these hard problems, but they can be used to scan the whole domain of possible solutions in search of one that works. Its use is primarily for implementing "brute-force" algorithms with a feasible runtime.

The real advantage to quantum computing is that many basic computer science algorithms become much simpler if you aren't limited to binary. Qubits allow to operate in higher bases than base 2, and the resulting simplification of many basic algorithms is the real benefit that we can never get from traditional computing.

What do you mean? It's possible to implement a computer in any numerical base other than binary. Not only that but you can emulate any architecture built on any numerical base on today's binary computers, and there is no simplification that i can see. We don't program with 0's and 1's anymore so, from a programming perspective it doesn't make a difference (apart from maybe reducing loss of precision in numerical computations). Computers built on a higher numerical base might be faster by some constant factor, because there is more paralellism, i.e. rather than f(n) binary operations, to perform the sum of 2 n-bit numbers, we're looking at f(n/2) binary operations, at the hardware level, in a base 4 machine (possibly though the hardware would still be twice as slow, i.e. 1 operation in twice the time, thus no speedup).
The cost of a different numerical base would be more complexity, both on a hardware level and a software level. It's hard to avoid binary, most decisions involve a binary domain. Most questions we ask involve either true or false. Imagine coding with a ternary data type instead of a boolean. Of course you can implement booleans in base 3, for example, but then you're wasting resources. It's human nature to use binary, i don't believe it would make anybody's life any easier.
Finally, not every problem/algorithm can be approached in a way such that it can profit from the use of a quantum computer in a non-sequential manner, and for sequential algorithms we have the sequential Turing-equivalent computers.

 

[michael879]

Ive only read the first two posts cause I am really tired but basically that quark thing is just entanglement which just appears to be information transfer. No information is actually being transferred. However, faster than light communication is possible with quantum effects. I am not sure about the exact conditions that it happens it (it might be all) but electrons and photons can travel faster then the speed of light while tunneling (since their mass is imaginary it makes sense). Some group made a beam of light travel 4xc I read.