GAViewer Simulation No Hidden Variable

[FrediFizzx]

Jeezus. Alpha and beta are angles. When the settings are opposite (differ by pi, or 180 degrees), does this mean that alpha = -beta?

Whatever. Let's say you are right but forget this nonsense on stackexchange. On page 2 of Gull's original so-called "proof" he says, "Your task is to ensure:



over long term performance." Well, the only time you get ++ or -- is when . Of course that means that and the RHS is zero. Well, my GAViewer program ENSURES that is true for EVERY single trial. So, the rest of Gull's "proof" must be nonsense. I'm still waiting for Gill to explain why that is not the case.
.

Gull is talking about two completely separate programs. They are each given the same initial “set of outcomes of all to be needed hidden variables”. Then each is given its own stream of settings, and each produces its own stream of outcomes. After that, we look at the statistics of + +, + -, - +, and - - for each theta1, theta2. What GAViewer can do when run once, on one computer, is irrelevant.

Yep, as expected. More waffling. You obviously don't understand Gull's TASK.
.

You are supposed to be explaining the rest of Gull's "proof" starting at number (2) on page 3. You already agreed that the GAViewer program passes the test on number (1) which is very obvious that it does no matter whether is is run on one or two computers. The result is the same. It never fails test (1). Now stop waffling and get to it.
.

[Heinera]

On page 2 of Gull's original so-called "proof" he says, "Your task is to ensure:



over long term performance." Well, the only time you get ++ or -- is when .
.

No, this is completely wrong. According to QM some of the outcomes will be ++ or -- even when .

[FrediFizzx]

On page 2 of Gull's original so-called "proof" he says, "Your task is to ensure:



over long term performance." Well, the only time you get ++ or -- is when .
.

No, this is completely wrong. According to QM some of the outcomes will be ++ or -- even when .

No problem. Just said it backwards. When , you will only get outcomes of ++ or --. IOW, the correlation is always +1.
.

[Heinera]

No problem. Just said it backwards. When , you will only get outcomes of ++ or --. IOW, the correlation is always +1.
.

Yes. And that is of course trivial to achieve. Gull does not say that this is difficult or impossible.

[gill1109]

You are supposed to be explaining the rest of Gull's "proof" starting at number (2) on [Gull’s] page 3. You already agreed that the GAViewer program passes the test on number (1) which is very obvious that it does no matter whether is is run on one or two computers. The result is the same. It never fails test (1). Now stop waffling and get to it.
.

Maybe this weekend. I have some grandchildren coming to visit, very soon. We have to socially distance while enjoying a big meal together. That means a lot of home engineering is needed rather fast.

[FrediFizzx]

No problem. Just said it backwards. When , you will only get outcomes of ++ or --. IOW, the correlation is always +1.
.

Yes. And that is of course trivial to achieve. Gull does not say that this is difficult or impossible.

Duh! And that was wrong also that you agreed with. When , you will never get outcomes of ++ or --. IOW, the correlation is always -1.
.

[Heinera]

No problem. Just said it backwards. When , you will only get outcomes of ++ or --. IOW, the correlation is always +1.
.

Yes. And that is of course trivial to achieve. Gull does not say that this is difficult or impossible.

Duh! And that was wrong also that you agreed with. When , you will never get outcomes of ++ or --. IOW, the correlation is always -1.
.

Ah, now I see that you were wrong again. When , the RHS becomes zero. So according to Gull (and QM) you will never get outcomes of ++ or -- in that case.

I don't understand why you have a problem with this equation. It's not something Gull has invented; it's taken straight from the QM predictions.

[FrediFizzx]

No problem. Just said it backwards. When , you will only get outcomes of ++ or --. IOW, the correlation is always +1.
.

Yes. And that is of course trivial to achieve. Gull does not say that this is difficult or impossible.

Duh! And that was wrong also that you agreed with. When , you will never get outcomes of ++ or --. IOW, the correlation is always -1.
.

Ah, now I see that you were wrong again. When , the RHS becomes zero. So according to Gull (and QM) you will never get outcomes of ++ or -- in that case.

I don't understand why you have a problem with this equation. It's not something Gull has invented; it's taken straight from the QM predictions.

Why do you think I have a problem with that equation? I don't have any problem with it. It is also a Bell condition that the correlation is always -1 when theta is zero.
.

[Heinera]

Why do you think I have a problem with that equation? I don't have any problem with it. It is also a Bell condition that the correlation is always -1 when theta is zero.
.

Good!

[FrediFizzx]

You are supposed to be explaining the rest of Gull's "proof" starting at number (2) on [Gull’s] page 3. You already agreed that the GAViewer program passes the test on number (1) which is very obvious that it does no matter whether is is run on one or two computers. The result is the same. It never fails test (1). Now stop waffling and get to it.
.

Maybe this weekend. I have some grandchildren coming to visit, very soon. We have to socially distance while enjoying a big meal together. That means a lot of home engineering is needed rather fast.

Hmm... didn't think you could do it. But you have the time to continue with your obsession with Joy. But enjoy your grandchildren just the same.

At the top of page 2, Gull writes; "Write a computer program which is to run on two independent personal computers which mimics the QM predictions for the EPR setup." No problem. My GAViewer program at the beginning of this thread can easily do that. So, I still contend that the rest of Gull's "proof" is nonsense. Now, you added the part about having to simulate the experiments event by event into Gill's "theorem". Of course that is NOT a QM prediction because QM can't do it either!
.

[gill1109]

You are supposed to be explaining the rest of Gull's "proof" starting at number (2) on [Gull’s] page 3. You already agreed that the GAViewer program passes the test on number (1) which is very obvious that it does no matter whether is is run on one or two computers. The result is the same. It never fails test (1). Now stop waffling and get to it.
.

Maybe this weekend. I have some grandchildren coming to visit, very soon. We have to socially distance while enjoying a big meal together. That means a lot of home engineering is needed rather fast.

Hmm... didn't think you could do it. But you have the time to continue with your obsession with Joy. But enjoy your grandchildren just the same.

At the top of page 2, Gull writes; "Write a computer program which is to run on two independent personal computers which mimics the QM predictions for the EPR setup." No problem. My GAViewer program at the beginning of this thread can easily do that. So, I still contend that the rest of Gull's "proof" is nonsense. Now, you added the part about having to simulate the experiments event by event into Gill's "theorem". Of course that is NOT a QM prediction because QM can't do it either!
.

Yes, of course you can run any computer program on any number of computers.
On the middle of page 2 of http://www.mrao.cam.ac.uk/~steve/maxent2009/images/bell.pdf, Gull says:
Each computer sets up a *dialogue*.
He means that a loop now starts, the computers both initialise n = 1

Both dialogues now start and look like this:

Computer: I am now processing particle number n. Please input an angle
Alice or Bob: the angle for Alice or Bob in the nth rial
Computer: the outcome + or - of the n'th trial for Alice of Bob

The computer now increments "n" and goes to the beginning of the loop again.

This is the part where says that the experiment is simulated event by event. QM is not a deterministic theory, and as it stands it is a non-local theory, since it does not tell us how to do that simulation on separate computers. Of course it is easy to do on one computer - you first simulate Alice's output, the wave function collapses, then you simulate Bob's output. But how to do it on two?

[FrediFizzx]

More freakin' waffling again! Number (2) on page 3. Start there! Stop wasting our time with obvious stuff.
.

[local]

Stop wasting our time with obvious stuff.

Amen. So boring.

[gill1109]

More freakin' waffling again! Number (2) on page 3. Start there! Stop wasting our time with obvious stuff.
.

Amen. So boring.

That was not "freakin waffling", I pointed out a serious error in Fred's thinking. He needs to think about it. OK, We can come back to it later.

Right then, let's go on to (3), so that "local" will be happy too.

The programs must be deterministic, and moreover identical to one another, apart from a minus sign. So there's a function p of the submitted angle and of the trial number, such that Alice's program result at the n'th trial is p(theta,n) when her input is theta, and Bob's program result at the n'th trial is p(phi, n) when his input is phi.

Hope I'm not going to fast for you, Fred, and hope I'm not going to slow for you, "local".

[FrediFizzx]

... So there's a function p of the submitted angle and of the trial number, such that Alice's program result at the n'th trial is p(theta,n) when her input is theta, and Bob's program result at the n'th trial is p(phi, n) when his input is phi. ...

Well, that is not how I was reading first sentence of number (2). But if that is right, let's use his notation. So, we will have for Alice and for Bob. I take it that these are probability functions. So, what are the actual functions? What is the probability that will be a certain angle on the nth trial? Does this even make sense to know that? Maybe I am missing what he is saying here.
.

[FrediFizzx]

... So there's a function p of the submitted angle and of the trial number, such that Alice's program result at the n'th trial is p(theta,n) when her input is theta, and Bob's program result at the n'th trial is p(phi, n) when his input is phi. ...

Well, that is not how I was reading first sentence of number (2). But if that is right, let's use his notation. So, we will have for Alice and for Bob. I take it that these are probability functions. So, what are the actual functions? What is the probability that will be a certain angle on the nth trial? Does this even make sense to know that? Maybe I am missing what he is saying here.
.

Oh for heaven's sake. I guess he is just talking about the measurement functions and . In the case of the GAViewer program, lambda is c the singlet spin vector direction.
.

[gill1109]

More freakin' waffling again! Number (2) on page 3. Start there! Stop wasting our time with obvious stuff.
.

Amen. So boring.

That was not "freakin waffling", I pointed out a serious error in Fred's thinking. He needs to think about it. OK, We can come back to it later.

Right then, let's go on to (3), so that "local" will be happy too.

The programs must be deterministic, and moreover identical to one another, apart from a minus sign. So there's a function p of the submitted angle and of the trial number, such that Alice's program result at the n'th trial is p(theta,n) when her input is theta, and Bob's program result at the n'th trial is p(phi, n) when his input is phi.

Hope I'm not going to fast for you, Fred, and hope I'm not going to slow for you, "local".

Sorry, typo, Bob’s output is -p(phi, n).

Alice’s p(theta, n) is the output her program would give on the n’th call of her program, if the n’th input were theta. Actually, I suppose that that output could depend also on the previous n-1 inputs. We will have to see if that is a problem.

[FrediFizzx]

...
Sorry, typo, Bob’s output is -p(phi, n).

Alice’s p(theta, n) is the output her program would give on the n’th call of her program, if the n’th input were theta. Actually, I suppose that that output could depend also on the previous n-1 inputs. We will have to see if that is a problem.

If that is the case, what are the functions?
.

[FrediFizzx]

...
Sorry, typo, Bob’s output is -p(phi, n).

Alice’s p(theta, n) is the output her program would give on the n’th call of her program, if the n’th input were theta. Actually, I suppose that that output could depend also on the previous n-1 inputs. We will have to see if that is a problem.

If that is the case, what are the functions?
.

So we have,




What are he question marks?
.

[FrediFizzx]

I suspect you are misreading Gull's number (2). I think for "answer", he is talking about the correlation. And the correlation for each event will be 0 or 1. But I don't understand what he says, "Can be different between trials". ???

If anyone else has a clue about this, go ahead and chip in. Here is Gull's original proof.

http://www.mrao.cam.ac.uk/~steve/maxent ... s/bell.pdf
.