Each experiment's shared phase is centred on one of the two labs' own current settings, via a coin flip and a half-cosine distribution (a cosine-weighted lobe). Below 100% setting dependence (λ), the rest of the draws are a uniform phase.
Experiments (Random Setting Each Time)
Trials completed: 0
Live CHSH S: — (classical ≤2, Tsirelson ≤2.828)
Setting
n
E
Correlation vs. Separation (S-curve)
Lab 1 fixed at 0°; Lab 2 stepped across 0-180° separation. STEP adds trials to every separation point at once, so you can watch the curve refine as it builds up.
0 trials per separation point so far.
Medium-bound model (instant maths, shown in the lab)
White arrow = the medium's draw, centred on the primary lab's pick (E). Coloured arrow = each lab's rotation (setting). A lab flips its sign (+ ↔ −) when the angle between the two arrows is over 90°; the primary never does. Arrow turns red on a flip, green if not.
Trials kept: 0 Attempts: 0 Set aside: 0 Live CHSH S: — |S|: —
Pair
n
E
|S| as trials build up (grey = 2, cyan = 2.828)
Medium-bound model
One shared medium. H is fixed when the pair is emitted. The pair leaves with opposite signs at Lab A 0° and Lab B 45°, or Lab A 90° and Lab B 135° (50:50), and the lab whose start angle is closest to the pair's hidden plane is primary; the medium's draw is a half-cosine lobe around the primary's own angle, so the primary never flips and the secondary flips with chance (1 − cos gap)/2. A flip makes the secondary read the other sign, so the pair then reads the same. The herald is a sync check: only pairs with both particles in step with the frame are kept. Waves cross space at c; the constraint that binds the pair is far faster and carries no signal. This view runs the phase-clock engine, which gives the same table; the exact rule is in the foundation lab.