5D Library › Dyadic Phase Coupling, Bilinear Compatibility Tensors and Spatiotemporal Grid Games · 中文

Preface: Writing “Compatible or Not” as Something One Can Compute

Note

This is not divination and not relationship advice; please read this first. The book studies the mathematical structure behind rules for matching two people’s charts, choosing dates and choosing directions on the nine palaces: bilinear forms, coupled oscillators, first-passage times, permutation groups and games. Its models are taken from traditional divinatory arts, but the book neither claims nor tests that any divinatory prediction holds — whether two people suit each other, when they will part, when and in which direction to act. “Relationship”, “separation” and “good door” are symbols inside models; wherever a concept from physics or game theory is applied to people, it is marked as an analogy. All data are generated by the programs printed in the book.

A chart-matching engine, whether it uses the Eight Characters, a Western horoscope or a Zi Wei chart, must in the end answer three questions: what kind of quantity “compatibility” between two people is; how long it lasts in time and when it is broken; and, on a grid of directions that turns with time, when and where it is best to act. The traditional answers are mnemonic verses. This book writes them as equations, runs the equations, and asks what the equations themselves can and cannot say.

Like the other books of 5D Library, it tries to refute before it sets bounds, and it reports negative results as they are. Several run directly against the outline and are listed below.

What the book contains

Chapter 1 writes a match score as a bilinear form s(A,B)=xA⊤MxBs(A,B) = x_A^{\top} M x_B, splits the matrix MM into a symmetric and an antisymmetric part, and asks what data can see “one-sidedness”. Chapter 2 uses the locking equations of Kuramoto and Adler for two rhythms, and the Arnold tongues of the sine circle map for other ratios. Chapter 3 adds amplitude, asks when “one side gains while the other dissipates” holds in a model, and treats coupling in which one side is attracted and the other repelled. Chapter 4 writes the breaking of a lock as a noise-driven first-passage time, adds a yearly outside kick, and asks where a “danger window” comes from and how much data it would take to test one. Chapter 5 writes the Luoshu square and the eight doors as permutation groups, and Chapter 6 solves for the best time and direction on the turning nine-palace grid and for a two-person game on it. Chapter 7 reads the Tuan commentary on the hexagrams Xian and Heng of the Zhouyi, the Rhapsody of the Old Fisherman of Mist and Waves and Yang Hui’s magic squares alongside the models. Chapter 8 gives a four-step check and a pre-registered test.

The book borders on several others in 5D Library and recalls them briefly only where they overlap: Book 21 treated attachment theory and the dynamics of couples; Book 44 the synchronisation transition of many coupled oscillators (this book deals with two people only); Book 33 falsifiability and multiple testing; Book 37 the modular arithmetic of the sexagenary calendar; Book 55 games of imperfect information.

Claims corrected after checking

  1. A shared outcome cannot see one-sidedness. When only an outcome common to both people is observed — how long the pair lasted — the antisymmetric part of MM never enters the data: only 10 of 16 parameters are identified, however many pairs are seen. To see who gives and who gains, each person must be measured separately (p02).
  2. Asymmetry does not change how easily a pair locks, only who adapts. The width of the locking tongue depends only on the sum of the two couplings; with the same sum, the common rhythm is 1.2700 at (0.45, 0.05) and 1.0300 at (0.05, 0.45) (p03).
  3. “One side gains, the other dissipates” does not hold between equal oscillators. Diffusive coupling costs both sides amplitude, the follower more; only when the natural amplitudes differ does the weaker side gain (+0.0123 and +0.0517 in the examples) while the stronger pays (p05).
  4. Attraction against repulsion can still lock, with a common rhythm faster than either. When one side pursues and the other flees, the common rhythm in the example is 1.7500, above both natural rhythms of 1.0 and 1.3 (p06).
  5. Noise alone makes no “danger window”. With noise only, the wait until a lock breaks is close to exponential and the hazard is constant; separations bunch in one month of the year entirely because of the assumed yearly kick, and more so the stronger the kick, the hazard ratio rising from 1.09 to 11.22 (p07, p08).
  6. Testing a “yearly window” needs many dated separations. A hazard ratio of 2 in a named month needs about 130 separations with exact dates; choosing the month after seeing the data makes a 5 % test “find” a window in 0.432 of studies (p09).
  7. The Luoshu square is unique up to symmetry, and the flying and turning moves generate every arrangement. There are only 8 magic squares of order three, images of one another under the symmetries of the square; together the flying-palace and turning-ring moves reach all 362,880 arrangements, of which only 8 keep the square magic (p10, p11).
  8. When the direction is free, timing adds nothing. The three good doors stand together and the four palaces next to the centre alternate around the ring, so a good door is always one step from the centre: the mean value is 0.900 when time and direction are free and also 0.900 when only the direction is. Only with a fixed direction does waiting raise the mean, from −0.1500 to +0.6625 (p12).
  9. Against a rival, the value of the game does not change with the hour, and neutral doors are used more than good ones. The turning of the doors only permutes the values over the palaces, so the value is 2 in every slot; in equilibrium each neutral palace is played with 0.200 and each good palace with only 0.133 — a predictable “best direction” is exactly what a rival blocks (p13).
  10. Xian and Heng are each other turned upside down. Read upside down, Xian is Heng: the same two roles with giver and receiver exchanged, which in the model of Chapter 1 is MM against its transpose (p14).
  11. Choosing the convention can manufacture any “match effect”. On data with no real effect, picking the best of twenty scoring conventions after the fact gives a “significant” result in 0.517 of studies, of which only 0.024 replicate in a fresh sample (p15).

How to read the boxes

The boxes separate published evidence, runnable Python calculations (16 programs, standard library only, fixed random seeds, each actually run at build time with its real output printed beneath it), classical sources, notes and the author’s hypotheses; every hypothesis states what result would show it to be wrong.

Disclosure

The author uses the works of 5D Library, including this book, as the methodological basis of his community courses, the 5D Ascension Cards (Dimensional Shift Card Decks) and the Yafu AI knowledge base, and develops and runs paid online reading software and services built on traditional divinatory symbol systems (chart matching, the six-line hexagram method, Zi Wei Dou Shu and Qimen date selection); the models of this book are the mathematical basis of those programs’ algorithms, so he has a direct commercial interest in its subject. The book promotes no software, course, card deck or service, uses no user data from the author’s businesses, and accepts no sponsorship. All its models are offered for mathematical study and teaching only and constitute no promise of divinatory prediction and no advice on relationships, date selection or investment. The test in Chapter 8 must be carried out with pre-registered code by an independent analyst who has no financial relationship with the author and does not know what the scores mean, and its results will be published whether positive or negative.

Fan, Y. (2026). Dyadic Phase Coupling, Bilinear Compatibility Tensors and Spatiotemporal Grid Games: Asymmetric Kuramoto Locking, Stopping-Time Separation Hazards and Magic-Square Permutation Operators [双人相位耦合、双线性相容张量与时空网格博弈] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090172 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).