Validation checkpoints · 2026-10-09
Where confidence actually comes from
October research update: scalar calculation repaired; independent trace agreement at r=6M under an empirical numerical allowance.
A later narrow pass does not rewrite an earlier FAIL. The 20 September discovery grades stay. Mτ remains paused and unconfirmed. Infinite-support Abel is not validated. Full independent RSET is false. GATE = NO.
Why this tree exists
The instrument is not trusted until the gates fire
B1 is calibration of a numerical engine, not a portal test. A tensor that merely produces interesting numbers is not enough. It must conserve, converge, obey the appropriate trace relation, and reproduce independent benchmarks before it can be aimed at an unknown geometry. That is why this ledger is a tree of separate claims rather than a single green light.
The rule
01
Human visual confirmation
02
Independent source corroboration
03
Derivation certification
04
Numerical reproduction
05
Physical validation
LIVE FRONTIER · EMPIRICAL BUDGET · NOT FULL RSET
Repaired scalar trace at r = 6M
VAL-B1-SCALAR-TRACE-R6
The massless scalar subtraction was repaired and compared with the stored PMR trace at one radius, r = 6M, under minimal coupling. Status TRACE_R6_NUMERICAL_PASS_UNDER_EMPIRICAL_BUDGET. The residual exceeds the scalar-only allowance and is covered only by a combined empirical allowance dominated by the inherited PMR estimate. This website did not re-run the calculation.
EMPIRICAL PASSSCALAR ALLOWANCE DOES NOT COVERINHERITED PMR ALLOWANCE DOMINATESNOT A TAIL BOUNDNOT NINE TRACE TESTSE NOT VALIDATEDFULL RSET FALSEGATE NOCLOSED NARROW WINDOW · NOT GLOBAL · NOT THE LIVE FRONTIER
CNG confirmation window, narrow pass
VAL-B1-PMR-CNG-CONFIRMATION
On 21 September 2026 this pipeline’s predeclared oscillatory structure passed on the unseen window ω ∈ [10.40, 10.90], 51/51 angular-accepted, ε₁ = 37.5 fixed and never fitted. The accepted record reports C1–C6 PASS. C5 was corrected from sequential residual fitting to joint-H1 source A/B transfer plus a fresh target cubic. This is not global asymptotics and does not erase the 20 September discovery grades.
NARROW PASS 2026-09-21C1–C6 PASS ON ACCEPTED RECORDC5 CORRECTEDGLOBAL NOT ESTABLISHEDD FAIL RETAINEDF CRITERIA GAP RETAINEDNOT HH RSETGATE NONARROW COMPARISON · FULL CHAIN STILL INVALID
Digitized published Hartle–Hawking curves
VAL-B1-PUBLISHED-HH-CURVES
October 3 records, reconciled on 7 October: normalized multi-frequency self-cancellation was already implemented. Digitized published Hartle–Hawking component values agree 12/12 at r = 5.5M, 6M, 6.5M, and 7M within stated tolerances. The original full-chain label remains MRX_R_INVALID because r = 5 failed tt n-convergence under analysis v1.0. A supplementary v1.1 pass does not replace that label.
12/12 NARROWMRX_R_INVALIDR=5 ANALYSIS_R_INVALIDV1.1 DOES NOT REPLACEWARD STILL HOLDNOT FULL RSETGATE NODERIVATION ONLY · TARGET-EXPOSED · E_VALIDATED FALSE
Hadamard e derivation candidate
VAL-B1-HADAMARD-E-CANDIDATE
A separate general-Hadamard derivation matches a, b, and e exactly, with c = d = 0. Sixteen checks are recorded as passing. Status E_DERIVATION_ONLY_CANDIDATE. The derivation was target-exposed, not blind. It is not a second Christensen/AHS implementation. E_VALIDATED stays false.
CANDIDATE16 CHECKS RECORDED PASSNOT BLINDE_VALIDATED FALSENOT A TENSOR VALIDATIONGATE NOHISTORICAL FREEZE · 2026-09-20 · NOT THE LIVE FRONTIER
PMR CNG Discovery Scan
VAL-B1-PMR-CNG-DISCOVERY
As written on 2026-09-20, before the unseen window was computed: substantial numerical evidence for a predeclared ε₁=37.5 oscillation, surrogate INVALIDATED, offline re-score UNCLASSIFIED_CRITERIA_GAP, confirmation still running. On 2026-09-21 that window received a narrow pass. This page keeps the discovery grades. D = FAIL and F = CRITERIA_GAP are not erased. The live frontier is now the 9 October scalar trace.
FREEZE 2026-09-20AS WRITTEN THAT DAY: SCAN NOT YET COMPUTEDLATER NARROW PASS 2026-09-21OLD SURROGATE INVALIDATEDF CRITERIA GAPD FAIL RETAINEDGLOBAL NOT ESTABLISHEDGATE NOCLOSED ANALYTIC LANE · WAS A LIVE FRONTIER
PMR Asymptotic Expectation Audit
VAL-B1-PMR-ASYMPTOTIC-AUDIT
After Levi–Ori F_sing subtraction, F_reg is not expected to decay before the generalized integral. Sparse ω=10,12 passed the frozen adaptive gate. |F_reg^tt| grew. A decaying high-ω tail is not demonstrated. That growth is the literature pre-GI remainder, not by itself a missing F_sing term. Not Abel. Not HH RSET.
GROWTH EXPECTED PRE-GIASYMPTOTIC STRUCTURE MIXEDGLOBAL CONFIRM FALSESPARSE 10/12 PASSDECAYING TAIL FALSEASYMPTOTIC REGIME FALSEFREQUENCY SUPPORT FALSEABEL PRODUCTION FALSEHH RSET NOT CLAIMEDCLOSED INSTRUMENT GATE · WAS A LIVE FRONTIER
PMR Adaptive Angular-Convergence Gate
VAL-B1-PMR-ANGULAR-GATE
Insufficient angular-mode support is the supported cause of an earlier apparent high-ω PMR subtraction failure. An adaptive angular-convergence gate was implemented and prospectively validated at unused ω=7. Sparse ω=10,12 later passed the same frozen gate. This is numerical-method progress, not an HH RSET reproduction or gateway result.
ADAPTIVE GATE PROSPECTIVE PASSSPARSE 10/12 PASSANGULAR PLATEAU PASSℓ-SUPPORT CAUSE SUPPORTEDABEL PRODUCTION FALSEHH RSET NOT CLAIMEDGATE NOPAUSED · SOURCE / LITERATURE REPRODUCIBILITY QUESTION · NOT THE LIVE FRONTIER
Mixed-Time Derivative Sign Reproducibility Check
VAL-B1-MTAU-SIGN-REPRO
Independently reproducible potential sign inconsistency in a displayed mixed-time derivative formula. Externally reviewable. Not confirmed. Internal derivations 3/3 PLUS. Public archaeology UNKNOWN. Lane paused pending external clarification. Production sign not flipped.
EXTERNALLY REVIEWABLENOT CONFIRMEDPAUSED3/3 PLUSARCHAEOLOGY UNKNOWN23/23 DERIVEDPRODUCTION SIGN NOT FLIPPEDPUBLIC REPRO READYACTIVE PRODUCTION-DEVELOPMENT ROUTE · RSET HOLD
B1 Extended-Coordinate Hartle–Hawking Clean-Room RSET
VAL-B1-EXTCOORD-HH
Pre-RSET stack remains a historical pass under the implementation and tolerances of that date. October 9 found a massless-tail omission and a wrong static-mode boundary coefficient; those corrections are recorded on the scalar-trace page and do not silently rewrite this ladder. Ward, Howard 4pt, and four-point DEV remain HOLD. The 20 September confirmation was not yet computed; it later received a narrow pass. B1 is not claimed. The current trace claim is only the r=6M empirical comparison.
B1 PARTIALTRACE HISTORICAL · NOT UNIVERSALWARD HOLDCNG DISCOVERY 2026-09-20CNG WINDOW NARROW PASSSCALAR TRACE EMPIRICALMTAU PAUSEDRSET HOLDGATE NOCLOSED GATE · WAS A LIVE FRONTIER
⟨φ²⟩ derivative nested-h certification
VAL-B1-EXTCOORD-PHI2-DERIV
Nested radial-spacing refinement passes for both first and second derivatives at all four certification radii. That gate remains closed. Four-point RSET DEV is now HOLD because of the separate Mτ source/sign question.
PHI2 DERIVATIVES = PASSNESTED h = PASSFOUR_POINT DEV = HOLDRSET HOLDHOLD · NOT A B1 PASS
Four-Point Hartle–Hawking RSET Development Certification
VAL-B1-EXTCOORD-RSET-4PT
Initial four-point assembly machinery exists. RSET_ASSEMBLY_DEV and FOUR_POINT_DEV remain HOLD. The historical trace machinery is not the 9 October minimal-coupling comparison at r=6M. Production sign is not flipped. The 35-row grid stays locked. Abel production is not validated. Tensor components stay HOLD.
RSET ASSEMBLY DEV = HOLDFOUR_POINT DEV = HOLDWARD HOLDABEL FALSEPRODUCTION GRID LOCKEDHISTORICAL BENCHMARK / NOT CURRENT ENGINE SPOF
Howard 1984 large-n asymptotic equations
VAL-B1-HOWARD-242-246
Human source-transcription checkpoint for Howard equations (2.42)–(2.46). Modern-engine progress does not retroactively repair Howard's undocumented intermediate algebra.
GPT_VISUALLY_RESOLVEDHUMAN OPENSECOND_INDEPENDENT_SCAN = NOT_FOUNDDERIVATION OPEN