FrediFizzx wrote:FrediFizzx wrote:Hmm... seems like I never posted the 3D code. I will have to find it. But I am going to tell you now that it is QM mapped to geometric algebra for validation of the product calculation.
Here is a link to the GHZ code to amuse yourself with in the meantime.
viewtopic.php?f=6&t=324.
I guess I never did the 3-sphere with 3D vectors code so I will have to do that. In the meantime,
viewtopic.php?f=6&t=388.
GAviewer code would be fine. But please make sure that it is code which recognisably implements your current two formulas defining A(a, lambda) and B(b, lambda). We would like to see what you really do mean by your non-standard limit notation and your non-standard linking of the dummy variables in two separate equations. If I try to implement it, I will end up getting the wrong answer, *and* you guys will accuse me of doing a straw-man attack, and it will all have been a waste of time.
Example. Already some years ago I tried to understand Python code of Michel and wrote "my own" version of his algorithm in R, but of course, he could not accept what I then wrote about "his" algorithm. He had published Python code only. He gave no literature reference and gave no clue where he had got the ideas from. If he had published a mathematical article using conventional mathematical notations, then the Python code could have been an implementation of his published mathematical formulas. And other people could have made other implementations. And everyone could openly discuss the maths and the implementations and the interpretations.
I think we need to come to some minimal requirements of simulation programs. The important thing is that one can experiment with them. And experiments must be reproducible. And for these reasons, I think the programs must have the following minimal requirements:
1) "save random seed" and "restore random seed" facilities
2) the user can freely submit their own *settings*
3) the user can demand that only one "trial" is done (one trial = two settings and two outcomes)
Any comments on these requirements? I think they are reasonable. If you think they are unreasonable, why?
I want to be able to test such software on a single computer, and not have to go to the trouble of setting up a computer network. But the computer program must in principle be so modular with modules for the source, for Alice's detector, and for Bob's detector.