Talk:Simulated optics experiment/Data analysis

I want to clarify that there is no quantum mechanics in the simulation, and there is no presumption that the light bursts are "photons". (I use scare quotes because the term comes with all kinds of presumptions that should be discarded here.) They could be "photons", or they could be a polarized light beam and a shutter. The simulation and the analysis both employ classical physics. Nevertheless we have finite bursts of fixed magnitude light at one end, and detection events at the other, which is supposed to be (or so I thought) all that quantum mechanics cares about. To quote Niels Bohr: "Physics is not about how the world is, it is about what we can say about the world." And QM, though merely trying to give the correct answer, without presuming how that answer came about, does come to the correct answer, if our simulation is to be believed.

It is good, no matter what, that we can use as an example of physics simulation something that actually provokes thought, rather than some something anodyne. An anodyne physics simulation is not one actually worth publishing, but this one I got from a published paper. --Chemoelectric (talk) 12:50, 30 May 2023 (UTC)

Also, one might wish to examine the "Extra credit" problem. The "credit" is for animating an explanation of where the orthodox reasoning goes awry. "Bell’s Theorem", unless I am wrong, contains an error so great that in most any other field of science, and in all of engineering until recently, it would be laughed out of the room. --Chemoelectric (talk) 12:54, 30 May 2023 (UTC)

Let me make one point more succinctly. It belongs in the Talk section rather than the task, because the task is about using computer programming in the sciences. (A summary of the point of the paper on which the simulation was based was necessary, of course.)

The simulation is a purely classical model. To argue that I am introducing some (illicit) step to attain "quantum magic" is to miss the point that there is no "quantum magic". The correlation coefficients are those of classical optics, not quantum optics. I have done what any physicist would do to analyze such a classical model. Yet we all get a CHSH contrast that supposedly proves there is "quantum magic". The supposition must be wrong, because we Rosetta Codeniks have written a simulation in which there obviously is no "magic".

(This is a tangential matter, but the reason for this is simple: "Bell's Theorem" is wrong. It must be wrong, because there is this counterexample. But, in fact, it is quite easy to show where Bell made an error that would get you a big Zero in, say, an electrical engineering Random Processes class: he correctly says things I will paraphrase as "A does not depend on b and B does not depend on a", but he mistakenly assumes that a and b are independent variables rather than functions of the "hidden variables".) --Chemoelectric (talk) 14:44, 30 May 2023 (UTC)

-- I will take the quantum reference out of the example. The point is the extra correlation, even in a fully classical simulation, happens because of the left-right switch. --Wherrera (talk) 17:13, 30 May 2023 (UTC)

-- There is no "extra correlation"! And this experiment has actually been done with radio waves (which are basically the same as light waves), and come up with the predicted results. (Evdokimov N V, Klyshko D N, Komolov V P and Yarochkin V A, Bell’s Inequalities and EPR-Bohm correlations: working classical radio frequency model, Physics-Uspekhi 39 (1969) 83–98.)

The left-right switch is necessary to compute the correlation coefficient correctly. How is one supposed to compute the correlation coefficient of electromagnetic fields at the outputs of polarizers, without accounting for the geometry of the EM fields coming into the polarizers? It cannot be done. If I did not account for geometry, I would be doing the calculation not only incorrectly, but obviously incorrectly. And we are doing the calculation using only information that plainly could be available in a fully classical experiment.

Ask an antenna engineer, who would have no urge to deny their knowledge of electromagnetism, merely to defend prevailing doctrine in other people's field. And who would be a tangible failure at their job, rather than published in PRL and Science, if they did deny it.

However, I am aware that nothing can get adherents to change their minds, much as I could not possibly convince the Pope to convert to Judaism, even if I had a letter from the Eternal One, signed in glowing script that floated a centimeter above the page. --Chemoelectric (talk) 12:55, 1 June 2023 (UTC)

-- I just waiting to see it claimed that correlation coefficient of light detections and the correlation coefficient of the EM fields responsible for those detections cannot be approximately equated. Or that approximate equation is not valid in physics experiments. Or anything of that kind. Seriously, I am waiting for it. --Chemoelectric (talk) 13:01, 1 June 2023 (UTC)

-- BTW has it occurred to anyone that they could prove or disprove whether Bell was right, by taking actual data sets from experiments and comparing them with the data sets from these simulations or similar ones? If the data sets are similar, then Bell was wrong.

I have never even read of that being done. I do not have such data sets, and I suspect no one has ever recorded such raw data and made it available.

And, even so, I am certain it would not change minds of the strong adherents. --Chemoelectric (talk) 13:06, 1 June 2023 (UTC)