Anyone know how a quantum comp gets answers?

I get the idea qubits can have lots of positions between zero and one, say a million of them and such, like 0, then 0.000001 and 0.000002 and on the other end, 1.0000 or 0.999999 or 0.999998 and so forth.
So the first question, how do you pose a question in that kind of computer and how does the qubits come up with an answer and how is that answer read out?

@sonhouse said
I get the idea qubits can have lots of positions between zero and one, say a million of them and such, like 0, then 0.000001 and 0.000002 and on the other end, 1.0000 or 0.999999 or 0.999998 and so forth.
So the first question, how do you pose a question in that kind of computer and how does the qubits come up with an answer and how is that answer read out?

I once read that to programme a quantum lawnmower to cut a lawn for example would take a lot of time and effort.
However, once done the lawnmower would cut each blade to the exact same height in minutes
As to how you would go about it I've no idea

very quickly...

@mlb62
Now new algorithms are showing classical comps beating theoretical quantum computers.

@sonhouse said
@mlb62
Now new algorithms are showing classical comps beating theoretical quantum computers.

Quantum computers are better at some computational problems, but not all, compared to classical computers. The difference is manifest where the problem has multiple parameters which could yield multiple acceptable solutions within tolerance; in such cases, quantum computers are more efficient than classical ones. However, where there are fewer parameters and only one correct solution, classical computers (with enough horsepower) may be more efficient.

For example, suppose you have 20 guests to invite to a dinner party, and you have to arrange the seating in such a way that no two people who can't converse civilly with each other are seated next to each other on either side. So you can't have a So. Baptist seated next to either an atheist or a classical pianist refugee from Ukraine, but you could have a Texas oil magnate seated next to a So. Baptist, and you could have a New England avant grade artist seated next to either a UC Berkeley history professor or a man who spent the last five years sailing round the world without using any fossil fuel (don't put the sailor next to the Texas oil magnate), and so on and so on -- i.e., there could be many acceptable seating arrangements, and the time to calculate the one perfect seating arrangement would be prohibitive for a classical computer, but a quantum computer might find an acceptable seating arrangement in a few microseconds.

Classical computers consists of "circuits" which can take one or the other of only two determinate states, closed/open, on/off, electricity is flowing or not-flowing, mapping to either one or zero, true or false. Quantum computers consist of "gates" which can be in an indeterminate state, like a Schroedinger-cat-box. Multiple "gates" can take multiple values simultaneously (both true and false, and possibly other values such as 'maybe', i.e., "fuzzy logic" (google that) ), and the multiple gates settle into fixed values at once, once certain other gates take fixed values. That is, once one gate or a certain number of gates settle into fixed states, then a cascade is triggered which settles other gates; this phenomenon is called "entanglement", and may seem to defy classical models of causality.

I'll see if I can find an article for you which explains in more detail how quantum computers actually process information internally.

@moonbus said
Quantum computers are better at some computational problems, but not all, compared to classical computers. The difference is manifest where the problem has multiple parameters which could yield multiple acceptable solutions within tolerance; in such cases, quantum computers are more efficient than classical ones. However, where there are fewer parameters and only one correct ...[text shortened]... for you which explains in more detail how quantum computers actually process information internally.

Nice explanation. Was reading up on this recently and Google was not able to explain it so clearly...

@sonhouse

<<
First, I’ll let you into a secret. Quantum computers will always be hybrid devices, partly quantum and partly normal. The latter is required to handle inputs and outputs, in order to interface with the lumbering apes who want to use the device.

For this reason, quantum SDKs are typically embedded in a standard programming language. QISKit uses Python, so we can use a Python program to deal with both the normal parts of the program, and to construct and run jobs for the quantum part.
>>

Quoted from:

https://medium.com/qiskit/how-to-program-a-quantum-computer-982a9329ed02



The idea of NOT (negation, or flip from one to zero or zero to one) is nicely described in the above-linked article, especially as it applies to quantum computing, where fractions of a NOT can be applied (i.e., half a NOT or a third of a NOT) in a game of battleship where, for example, it takes three hits (NOTs) to sink an aircraft carrier, two hits to sink a heavy cruiser and one hit to sink a destroyer.

@sonhouse said
@mlb62
Now new algorithms are showing classical comps beating theoretical quantum computers.

I'm sure that information transfer by telegraph was faster and more accurate than telephone for awhile when new.

@sonhouse said
I get the idea qubits can have lots of positions between zero and one, say a million of them and such, like 0, then 0.000001 and 0.000002 and on the other end, 1.0000 or 0.999999 or 0.999998 and so forth.
So the first question, how do you pose a question in that kind of computer and how does the qubits come up with an answer and how is that answer read out?

This video on the Veritasium channel attempts to describe how a quantum computer would go about finding the factors of a number that is the product of two enormous prime numbers, which is a matter relevant to data encryption and security. It's hand-wavey in its presentation, but still somewhat informative.

YouTube

https://phys.org/news/2024-04-team-qubits-ultrasensitive-thermal-detectors.html

There is a new paper out, as reported e.g. here about the way to read the data.

@Ponderable
So far, 92% accuracy. They say going to graphite instead of steel might get that up to 99.9 %, I wonder what the accuracy has to be if you have a million qubits?

@Suzianne
Well telegraph was pretty error free except for readers mistakes, say ..-. mistaken for .... F for H but that is a measure of the training code readers got back in century 19. But it is considered the first internet. I had to learn code for my first ham license but they have cut that out of the test now it is just for bragging rights and getting the best bang for the power and bandwidth buck, code can be transmitted with something like 200 Hertz as opposed to single side band speech at 3,000 Hertz. We can go a bit below that but try to get decent audio out of say 1,000 Hertz muddies up the result. Communicating at say 300 Hertz V 3000 Hertz gives you a ten db gain without any other change, ten db is like having a 1000 watt amp but using 100 watts total, so to get the same readability out of SSB would take 1000 watts over the 100 watts needed for code.