That's helpful, thanks. What I was after is how the apparatus in this experiment maps onto the CHSH description given by Scott (if indeed, it does). It seems

That's helpful, thanks. What I was after is how the apparatus in this experiment maps onto the CHSH description given by Scott (if indeed, it does). It seems clear that the x / y values are used to determine the basis of measurement of the respective detectors, the a / b values are output based on whether each detector fires, and it's the correlations based on difference between the bases that allows you to beat the game. The basis of measurement here is analogous to the angle of a polarising filter in a photon detector, but the detectors in this experiment are directly measuring the spin of the nitrogen vacancies (so what is the physical implementation of the different basis of measurement?).