================================================================================================================================================================
FROSTYVERSE TEST H2-082 — TWO-BODY MASS CLOSURE SEARCH
================================================================================================================================================================
version:  2.0.82-two-body-mass-closure-search
patch:    h2-081-target-aware-sun-planet-mass-function-extra-dispersion-blind-transfer-1
baseline: Frostyverse-Test-H2-081.zip sha256=e2f48b305c7021bb799dbf5ad02319aba56147401a43b2a56d2991fd9d5ae9f4
python:   3.10.6
profile:  full; R=(20, 40, 80); quadrature samples=32768; workers=6

TEST 01 - CONTRACT / EXACT QUESTION
----------------------------------------------------------------------------------------------------------------------------------------------------------------
R01  [D-UNCHANGED]  D remains the inherited mechanically supported dispersion unit.
R02  [TOTAL-FORM]  Test H_total = H_stable_FV + C(M_sun,m_planet)*D; do not redefine D as 2D plus something.
R03  [TARGET-AWARE]  This package deliberately opens H2-081's ~4D missing response and its Mercury direct coefficient.
R04  [TWO-MASS-ONLY]  Every mass-dependent law receives only q=m_planet/M_sun or equivalent eta/reduced/log functions of those same two masses.
R05  [MERCURY-ANCHOR]  Predeclared normalized mass laws use only Mercury q and H2-081 C_M=4.0498328559103465; no per-planet refit.
R06  [HOLDOUTS]  Venus, Earth, Mars, synthetic orbit x mass grid, mass swaps, and resolution transfer are scored after equations are fixed.
R07  [NULL-BASELINES]  C=4, H2-081 constant C, and 4*(1.011961...) are retained so mass dependence must outperform a constant, not merely fit Mercury.
R08  [POSTFIT-SEPARATE]  A continuous power exponent search is reported separately as exploratory POST-HOLDOUT fitting and never counted as blind evidence.
R09  [NO-NEW-FORCE]  No G change, kick, thrust, spin, master-law change, or new energy source is inserted.
R10  [PHENOMENOLOGICAL]  Even a successful mass law is an empirical effective closure, not microscopic FrostTransit proof.

TEST 02 - REBUILD INHERITED FV STABLE + D LEDGER
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  building inherited FV field/profile R=20 ...
    Mercury stable/D=+2.002165376 missing/D=+3.997834624
  building inherited FV field/profile R=40 ...
    Mercury stable/D=+2.001840588 missing/D=+3.998159412
  building inherited FV field/profile R=80 ...
    Mercury stable/D=+2.002348001 missing/D=+3.997651999
  SEALED stable+D+target ledger: 871ed14a22aa27aa497379928c2365d74a1c3d1df9bfb3320d7573a98ee671c6

TEST 03 - TWO-BODY MASS CATALOG / PREDECLARED EQUATIONS
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  MERCURY    q=m/M=1.660120825459e-07  M/m=6.023658e+06  eta=1.660120274259e-07
  VENUS      q=m/M=2.447838287797e-06  M/m=4.085237e+05  eta=2.447826304016e-06
  EARTH_EMB  q=m/M=3.003489615465e-06  M/m=3.329460e+05  eta=3.003471573647e-06
  MARS       q=m/M=3.227156082932e-07  M/m=3.098704e+06  eta=3.227154000026e-07
  predeclared models=23; all frozen before named/synthetic scoring
  H2-081 direct reference C_M=4.049832855910
  4 x nq0 narrative coefficient=4.047844334160

TEST 04 - NAMED PLANET + RESOLUTION TRANSFER
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  R80  MERCURY    required missing/D=+3.997651999
  R80  VENUS      required missing/D=+3.986821296
  R80  EARTH_EMB  required missing/D=+3.999740457
  R80  MARS       required missing/D=+3.999283199

TEST 05 - SYNTHETIC ORBIT x SYNTHETIC MASS FALSIFICATION GRID
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  synthetic orbits=6 mass factors=13 model rows=1794
  mass ladder relative to Mercury=(0.01, 0.03, 0.1, 0.3, 1.0, 3.0, 10.0, 30.0, 100.0, 300.0, 1000.0, 3000.0, 10000.0)
  Note: this is intentionally a test-particle-ruler falsification control; mass is varied while orbit/target are held fixed.

TEST 06 - NAMED MASS-SWAP CONTROL + FINITE-TOTAL-MASS BOOKKEEPING
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  largest q-only finite-total-mass bookkeeping shift in required C=0.000005%
  This control is not promoted to reciprocal FV two-body mechanics.

TEST 07 - PREDECLARED MODEL RANKING
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  #01 FOUR_PLUS_ETA                worst=  0.228% named=  0.220% synMass=  0.228% res=  0.039% mass=True 
  #02 FOUR_PLUS_Q                  worst=  0.228% named=  0.220% synMass=  0.228% res=  0.039% mass=True 
  #03 CONST_4                      worst=  0.228% named=  0.220% synMass=  0.228% res=  0.039% mass=False
  #04 SYMMETRIC_ETA_PLUS           worst=  0.245% named=  0.220% synMass=  0.245% res=  0.039% mass=True 
  #05 TOTAL_MASS_PLUS              worst=  0.245% named=  0.220% synMass=  0.245% res=  0.039% mass=True 
  #06 TOTAL_MASS_MINUS             worst=  0.339% named=  0.219% synMass=  0.339% res=  0.039% mass=True 
  #07 CONST_4X_NQ0                 worst=  1.017% named=  1.017% synMass=  0.932% res=  0.837% mass=False
  #08 CONST_H2_081_DIRECT          worst=  1.050% named=  1.050% synMass=  0.965% res=  0.870% mass=False
  #09 LOG_GAP_DIRECT_NORM          worst=  1.210% named=  0.907% synMass=  1.210% res=  0.870% mass=True 
  #10 Q_POWER_M1_8                 worst=  1.611% named=  0.813% synMass=  1.611% res=  0.870% mass=True 
  #11 LOG_GAP_INVERSE_NORM         worst=  2.160% named=  1.223% synMass=  2.160% res=  0.870% mass=True 
  #12 Q_POWER_M1_4                 worst=  2.761% named=  0.715% synMass=  2.761% res=  0.870% mass=True 
  #13 Q_POWER_P1_8                 worst=  2.761% named=  1.382% synMass=  2.761% res=  0.870% mass=True 
  #14 Q_POWER_M1_3                 worst=  3.989% named=  0.677% synMass=  3.989% res=  0.870% mass=True 
  #15 Q_POWER_M1_2                 worst=  8.440% named=  0.608% synMass=  8.440% res=  0.870% mass=True 
  #16 Q_POWER_P1_4                 worst=  8.440% named=  1.847% synMass=  8.440% res=  0.870% mass=True 
  #17 Q_POWER_P1_3                 worst= 18.028% named=  2.256% synMass= 18.028% res=  0.870% mass=True 
  #18 ETA_SQRT_NORM                worst= 83.052% named=  3.537% synMass= 83.052% res=  0.870% mass=True 
  #19 Q_POWER_M1                   worst= 83.189% named=  0.439% synMass= 83.189% res=  0.870% mass=True 
  #20 Q_POWER_P1_2                 worst= 83.189% named=  3.537% synMass= 83.189% res=  0.870% mass=True 
  #21 ETA_LINEAR_NORM              worst=8278.105% named= 15.031% synMass=8278.105% res=  0.870% mass=True 
  #22 REDUCED_LINEAR_NORM          worst=8291.847% named= 15.031% synMass=8291.847% res=  0.870% mass=True 
  #23 Q_POWER_P1                   worst=8305.610% named= 15.031% synMass=8305.610% res=  0.870% mass=True 

TEST 08 - EXPLORATORY POST-HOLDOUT POWER SEARCH (NOT BLIND)
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  POSTFIT_Q_POWER        p=-2.000000000 named-C-fit-rms=2.782134e-02 worst transfer=8305.605%
  POSTFIT_ETA_POWER      p=-2.000000000 named-C-fit-rms=2.782134e-02 worst transfer=8305.610%
  These exponents used the named-body required coefficients and therefore are exploratory fits, not blind validation.

  SEALED H2-082 mass-closure fingerprint: ad21991ddd94a870732acff03fd359aa0b9f5555d370c8e5fb7e988637e4bcc7

TEST 09 - DECISION
----------------------------------------------------------------------------------------------------------------------------------------------------------------
  constant H2-081 reference worst primary error = 1.050%
  best PREDECLARED mass-dependent law          = FOUR_PLUS_ETA
  best mass-dependent worst primary error       = 0.228%
  mass-law improvement vs constant              = +0.822 percentage points
  H2-082 verdict: MASS-DEPENDENT EMPIRICAL CLOSURE SURVIVES AND IMPROVES ON THE CONSTANT REFERENCE

Interpretation rule:
  - If a mass law beats the constant on named bodies AND the synthetic mass grid without refit, record it as an empirical two-body closure candidate.
  - If only a postfit exponent wins, that is curve fitting and needs a genuinely new holdout before promotion.
  - If the constant wins, the current evidence says mass dependence is already carried inside D / the orbit state rather than its ~4 multiplier.
  - No outcome here explains the microscopic FrostTransit mechanism.

H2-082 complete. Return h2-082results.txt first; CSVs are optional unless a row needs verification.
