Skip to content
This repository was archived by the owner on Jul 7, 2026. It is now read-only.

ligo_tests_findings.md

Latest commit

 

History

History
176 lines (125 loc) · 5.9 KB

ligo_tests_findings.md

File metadata and controls

176 lines (125 loc) · 5.9 KB
Raw

SSZ-LIGO Tests — Repository Findings

Repository: E:\clone\ssz-ligo-tests
GitHub: error-wtf/ssz-ligo-tests
Last Updated: 2026-05-19
Task: SSZ_LIGO_DERIVE_MISSING_FORWARD_EQUATIONS_FROM_SOURCES
Status: PHASES 1–9 COMPLETE

Overview

This repository implements an anti-circular, exploratory SSZ strain pipeline for GW240925. No physics claims are made. All formulas are source-locked or explicitly marked BLOCKED.

Derived V0 Formulas

deltaPsi_SSZ(f) — DERIVED_V0_PROXY

Source: formula_compendium.md §C.1, SSZ Book Ch.31, ssz_inspiral.py

r(f)     = (G*M / (pi*f)^2)^(1/3)       [Kepler 3rd law]
rdot_SSZ = rdot_GR * D^2 / s^4          [locked: formula_compendium §C.1]
deltaPsi_V0 = integral [1/D(r') - 1] * Omega(r') / rdot_GR(r') dr'

Regime LIGO band (20–800 Hz, 8.9 Msun): r/rs ≈ 100–1000 (weak field)
Branch: g2_decay (operative per FORMULA_BRANCH_LOCK.md)
FORMULA_STATUS: DERIVED_V0_PROXY
Implementation: src/ssz_ligo_tests/derived_phase.py

deltaA_SSZ(f) — DERIVED_V0_PROXY

Source: formula_compendium.md §B.4

P_GW_SSZ = P_GW_GR * D^2 / s^2
deltaA(f) = D(r(f))^2 - 1   in (-1, 0]

Implementation: src/ssz_ligo_tests/derived_amplitude.py

h_SSZ(f) — DERIVED_V0_PROXY

h_SSZ(f) = h_GR(f) * [1 + deltaA(f)] * exp(i * deltaPsi(f))

Implementation: src/ssz_ligo_tests/derived_waveform.py

epsilon_220 Registry — BLOCKED

Three conflicting branches from corpus. No canonical branch selected.

Branch Value Source Status
v51_ch30_3_percent 0.03 SSZ Book V51 Ch.30 PARTIAL_EXPLORATORY
dmin_squared_scale 0.308 formula_compendium §B.7 SUPERSEDED_OR_DIFFERENT_REGIME
photon_sphere_39_percent 0.39 qnm_spectrum.md (r*/rs=1.387) DISCARDED_FOR_LIGO_STRAIN

Note: The 39% value is a source-frame QNM frequency ratio from qnm_spectrum.md, not a LIGO strain amplitude observable. Using it as a strain observable is a category error.
Implementation: src/ssz_ligo_tests/epsilon_220_registry.py
FORMULA_STATUS: BLOCKED_BRANCH_CONFLICT

Pipeline Run Results (GW240925, H1, 2026-05-18)

Item Value
Strain source GWOSC LIGO HDF5
Anti-circularity VALID_INDEPENDENT
deltaPsi band (20–800 Hz) min=5.98e-02, max=1.08e+01 rad
deltaA band (20–800 Hz) min=−0.692, max=−0.065
lnL GR control (0PN) −3.24e+07
lnL SSZ V0-proxy −3.24e+07
delta_lnL +6.34e-06
Interpretation INDISTINGUISHABLE (delta_lnL < 1)
MF-SNR GR 39.92
MF-SNR SSZ 14.23 (V0 proxy not amplitude-matched)

SSZ_SUPPORT_CLAIM_MADE: NO
SSZ_FALSIFICATION_CLAIM_MADE: NO
READY_FOR_REAL_LIGO_SSZ_CLAIM: NO

Test Suite

Test File Tests Status
test_derived_delta_psi_v0.py 13 PASS
test_derived_delta_a_v0.py 11 PASS
test_h_ssz_v0_waveform_application.py 13 PASS
test_epsilon_220_branch_registry.py 11 PASS
test_no_data_fitting.py 13 PASS
Full suite 330 330 PASS, 1 xfail (expected)

xfail: test_d_min_squared_vs_3_percent — ε₂₂₀ ambiguity (31% vs 3%) cannot be resolved without author decision.

Locked Constants (from FORMULA_BRANCH_LOCK.md)

Constant Value Source
Xi_strong 1 − exp(−φ·rs/r) [decay form] formula_compendium.md §A.3
BLEND_START 1.8 rs FORMULA_BRANCH_LOCK.md
BLEND_END 2.2 rs FORMULA_BRANCH_LOCK.md
D_min at rs 0.555 formula_compendium.md
Xi at rs 0.802 formula_compendium.md
s at rs 1.802 radial_scaling.md

Open Blockers

  1. BLOCKED_MISSING_EQUATION: deltaPsi exact RSG phase integral (SSZ Book Ch.31 not yet final)
  2. BLOCKED_BRANCH_CONFLICT: epsilon_220 (3 conflicting sources: 3%, 31%, 39%)
  3. GR_CONTROL_LIMITED: TaylorF2 0PN only — no spin, no merger, no ringdown

Phase 9 — H1/L1 Time-Delay Replication (2026-05-19)

Script: scripts/run_h1_l1_time_delay_replication.py
Output: reports/H1_L1_TIME_DELAY_REPLICATION_REPORT.md, data_manifest/h1_l1_delay_scan.csv (2208 rows)

Method

FFT cross-correlation lag scan ±50 ms coarse / ±10 ms fine. 6 subbands (20-40, 40-80, 80-120, 120-160, 160-210, 20-210 Hz). Off-source comparison window at t−500 s to detect persistent systematics. Sign-flip test for L1.

Results

Window Band Verdict Peak xcorr
TRIGGER 20-210 Hz PHYSICAL_DELAY_COHERENT −0.9959
TRIGGER 20-40 Hz PHYSICAL_DELAY_COHERENT −0.9998
OFF_m500 20-210 Hz PHYSICAL_DELAY_COHERENT −0.9999
OFF_m500 20-40 Hz PHYSICAL_DELAY_COHERENT −0.9998

H1_L1_REPLICATION_STATUS: COHERENT_WITH_SIGN_FLIP (updated 2026-05-19)

The full-band trigger shows abs(xcorr) = 0.991 at dt = 0 ms. The negative sign is physically expected (H1/L1 opposite arm orientations). The previous PERSISTENT_SYSTEMATIC classification was a peak-detection artifact from using argmax(corr) instead of argmax(|corr|).

The L1 20–40 Hz non-Gaussianity (kurtosis +44.9) remains separately unresolved. Subband abs-correlation run is the next required step.

Gate

H1_L1_REPLICATION_STATUS:        COHERENT_WITH_SIGN_FLIP
BEST_DELAY:                       0 ms
BEST_ABS_CORR:                    0.991
PREVIOUS_DT_MINUS_10MS_RESULT:    CLASSIFICATION_ARTIFACT
TRIGGER_SPECIFIC:                 PENDING (subband abs-corr required)
L1_20_40_STATUS:                  DQ_CONTEXT_REQUIRED
READY_FOR_REAL_LIGO_SSZ_CLAIM:    NO
SSZ_SUPPORT_CLAIM_MADE:           NO
SSZ_FALSIFICATION_CLAIM_MADE:     NO

Absolute Rules Compliance

All 9 absolute rules confirmed: no fitting, no tuning, no posterior verdict, no SSZ claims, no silent GR substitution, no source deletions, all formulas documented, blocked items marked, conflicts preserved.