Science a GoGo's Home Page Forums General Discussion General Science Discussion Forum Progress in Quantum Computing

An error-checking method that could prove crucial to the development of a practical quantum computer has been developed by US researchers.

Quantum computers process information in the form of quantum bits, or qubits. These act like the bits in conventional computers but, instead of existing in one of two states, a qubit can exist in both states simultaneously.

This means a quantum computer can perform multiple calculations simultaneously. So far, only a handful of qubits have been used at a time to perform calculations in the laboratory. But if quantum computers can be scaled up, they should be able to perform incredibly tricky calculations in an instant.

Physicists at the University of California in Santa Barbra, US, have discovered a new way to check how much the information stored inside a quantum computer has decayed. This is an impressive feat since measuring the state of a qubit normally destroys its quantum properties.

For more:
http://www.newscientisttech.com/article/dn9301-errorcheck-breakthrough-in-quantum-computing.html

Interesting. But unless we can understand the meaning of what we compute it is useless. I've given up playing Chess against the computer. I now accept a computer can do any challenge, I set for it, better than I. It only proves there's a meaning to life, but also that I have not found it. I've decided friendship is above principle. For principle has no meaning except in that context. I should delete this post as off topic but it's what it says to me.

Quantum computation, like controlled hydrogen fusion, is the wave of the future! Always has been; always will be.

Decoherence during input, processing, and output places serious restrictions on the number of qubits and the duration of computation. The Second Law of Thermodynamics is only statistical, but I don't see Maxwell's Demon for sale in any Wal-Mart. You'd think folks would be motivated.

A quantum CPU will be an esoteric concatentation of nanofabrications in exotic materials cooled hard by absolute zero - and those are the easy parts.

I can't understand why many Quantum Computing researchers are trying use qubits that are in Quantum Superposition states, as this would create a number of difficulties.

QS States would hold any usable information for computing purposes. The old computing saying of "Garbage in, garbage out" can be restated as "No information in, no information out".

Boolean logic would be difficult or impossible to use with Quantum Superposition States. E.g. If you AND a known "1" qubit e with a QS State qubit, what result would you get?
The possible results would be:
1. A ?1? - because the "1" of the known qubit would match the "1" component of the QS State qubit.
2. A ?0? - because the "1" would not match the "0" component of the QS state qubit.
3. Both a ?1? and a ?0? - i.e. a QS State.
4. Any of the above based on some random function.

How can you test if a qubit is in Superposition or not?
Even if you are able to test for a QS state, how should it be interpreted? The only useful way I can think of would be to use it as a third value - i.e. "on", 'off" and "both". However the Russians were experimenting with trinary logic circuits (using +1, 0, -1), but nothing seems to have come of it.
The closest thing to the successful use of a third state is in statistical computing where they have the concept of a ?null? value, which is distinct from a zero value (or blank or false). However this applies to a field rather than a single bit, and indicates that there is no value to process as no information was received.

Current computer hardware, software and communication protocol designs use redundancy to detect and correct errors ? usually limited detecting and fixing errors where only a few bits have changed value. With QS States this would be impossible (unless you can detect and use a QS State as a third value ? however doing this would require the development of completely new computer hardware, software architectures and communications protocols).

Given all the problems and uncertainties associated with processing qubits in Quantum Superposition states, researchers should be seeking to avoid QS states instead of trying to encourage them. This may not require any effort as the act of processing qubits should ensure that they are not no in a QS State.

Very good.
In fact the idea of QC is interesting, but all the attempts to develop it have failed miserably.

ES

- Good job Edison didn't think like you.

Blacknad.

Looks like some of my words went into a quantum superposition, lost all information vanished into the bit bucket - or maybe its just my bad typing.

Blacknad, I agree with you. I can see no fundamental problems with Quantum Computing - as long as you don't get qubits in Quantum Superposition.
QC seems to be in a similar state to that of the electric light during the period when Edison was still trying to get it to work. While there are technical problems to be resolved, and there are none which seem to be insoluble, it is still possible that a problem may be encountered which makes QC impractical.

Peter.

Quantum computing works precisely because of these superpositions. It has already been demonstrated in the lab. Shor's factorization algorithm and Grover's search algorithm have been implemented using a few qubits.

Do you have a reference for that?

You are smarter than this.
It is not that algorithm can not be implemented,
it is that it is super SLOW, contrary to the claims it being super fast.

ES

extrasense wrote:
"it is that it is super SLOW, contrary to the claims it being super fast."

Of course it is. That's why all of the smartest PhDs in computer science are working on making it commercially viable.

Do you have reference to support this statement ?

Listen,
There is NO scientific work, that PROVES any of the claims about speed of "quantum computation".

And "computer science" is not an area to discuss this issue, it is PHYSICS.

All attempts to "correct", are attempts to run around the quantum uncertainty - a pretty stupid effort.

The evaluation of a "quantum function" with a PARAMETER, was never a part of the quantum theory.

You need new physical system for each N, and the time to create such a system grows dramatically with N.

The CS PhDs are not qualified to handle this issue, they are totally misguided.

ES

extraNONsense wrote:
"There is NO scientific work, that PROVES any of the claims about speed of "quantum computation"."

Please drink some coffee. You claimed it was slow. Now you say there is no proof it is fast: These two statements are not equivalent. So don't change the subject but rather consider the ethical course which would be to retract your statement.

BTW: Your second statement too is incorrect. So, yet again, either provide a reference or retract your statement. Quantum computing has been done. And we do know how fast it is. You may not know but that is your personal issue not a matter for science or engineering.

Every "work" on quantum computing is invoking the "projection postulate", and adds that the "measurement" has no time. Wishful thinking!

ES

Thank you for posturing and spin-doctoring. But where is that single reference that supports either of your statements?

As unreachable in our universe as the interior of a black hole.

The pseudoscience worships references.
I doubt you know what the "projection postulate" is, and how it is different from the real quantum theory.

QC was invented by a bunch of ignoramuses, who never had a clue what the real quantum theory is.

ES

The lack of references that support your statements proves two things beyond a shadow of a doubt.

1. You just made up what you posted.

2. Not a single person has ever published anything that supports or agrees with your statements.

Conclusion: You are wrong!

That is how science works.

You have been relegated to being another Aristotle ... long on opinion ... short on fact.

There is a number of publications, that agree with my statements.
One of the fairly wide accepted approaches is, that QC can be useful only as a sort of analog computing, with all the advantages and disadvantages of analog computers - plus the conundrum of uncertainty.

All the claims beyond that are demonstrably false.

ES

"There is a number of publications, that agree with my statements."

We went through this before when you cited references you said would support your argument - which they did not. I don't know if QC will ever be anything more than a curiosity. But I do know this, extrasense - that you are abjectly ignorant of the subject.

extraNONsense wrote:
"There is a number of publications, that agree with my statements."

Name one! That's all I've been asking you to do for days ... Just name one!

Have you no shame?