Experiments
Deterministic scripts with dual CSV+JSON storage. Green checks are the experiment's own validation checks; click a row for data, charts, and caveats.
| ID | Experiment | Checks | Rows | Date |
|---|---|---|---|---|
| E-001 | Window constants R(psi) symbolic+numeric verification (task 0.4) 0.4 |
— | 2 | 08-17-2026 |
| E-002 | First 10^4 zeta zero ordinates (dps 25, multiprocess) 0.6a |
✓ ✓ ✓ | 10000 | 08-17-2026 |
| E-003 | Finite Gtilde verification T=1000/2000/4000, psi0+psiMT 0.6b |
✓ ✓ ✓ ✓ ✓ ✓ | 6 | 08-17-2026 |
| E-004 | Sine-kernel Gram matrix spectral moments Monte Carlo (thread T1) 1.3 warm-up (thread T1) |
✓ ✓ | 1 | 08-17-2026 |
| E-005 | Oversampled third-moment decider (lambda,q) sinc-Gram CMS bounds 1.3 decider (D-011 oversampling hatch) |
— | 8 | 08-17-2026 |
| E-006 | Positivity horizon: eigenvalue floor vs bandwidth 3.1 / Q-3.4b (positivity horizon) |
— | 4 | 08-17-2026 |
| E-007 | Extended-precision positivity floor (T=200, Hann exact FT, dps40) 3.1 / Q-3.4b (extended-precision positivity floor) |
— | 4 | 08-17-2026 |
| E-007b | Extended-precision positivity floor (T=400 full-rank points) 3.1 / Q-3.4b (extended-precision positivity floor) |
— | 3 | 08-17-2026 |
| E-008 | Li coefficients n<=2000 from zero table with tail bounds 3.2 part 1 (Li coefficients from zero table) |
✓ ✓ ✓ | 2000 | 08-17-2026 |
| E-009a | Suzuki screw function g(t): implementation + eq. 2.2 validation + tabulation 3.4c (D-018 step 1): Suzuki screw function g(t) |
✓ ✓ ✓ ✓ ✓ ✓ | 10001 | 08-17-2026 |
| E-009b | Galerkin spectrum of Suzuki A_a (Weil-form operator); validated vs zero-side explicit formula 3.4c (D-018 step 2): Galerkin spectrum of Suzuki A_a |
✓ ✓ ✓ ✓ ✓ | 120 | 08-17-2026 |
| E-010 | Zeros of Suzuki W(a,theta;z) via Galerkin deficiency vectors; EXTENDED sweep (a<=8, lambda=-1 added, h_conv to 2.5e-3) 3.4c cont. (D-019 §4): zeros of Suzuki W(a,theta;z) |
✓ ✓ ✓ ✓ | 48 | 08-17-2026 |
| E-011 | Li coefficients n<=2000 from arithmetic side (Bombieri-Lagarias Thm 2 eta-form; flint Stieltjes + Arb zeta; zero-free); ALL CHECKS PASS 3.2 continuation: Li coefficients from arithmetic side |
✓ ✓ ✓ ✓ | 2000 | 08-18-2026 |
| E-012 | dBN heat-flow toy: two-body collision times (10^4 zeros) + full ODE check (300 zeros) 3.3: de Bruijn-Newman heat-flow toy (D-021 illustration) |
✓ ✓ ✓ | 9999 | 08-18-2026 |
| E-013 | xi-prime certificate constants from D1 diagonal law (Remark 7.1 reproduced to 5e-6) T4: xi'-analogue — why is the method stronger on xi'? (D-022) |
✓ ✓ ✓ ✓ | 3 | 08-18-2026 |
| E-014 | Cor 1.6 diagnostics: theta-count interlacing, v+/- localization, zero-flow to a=10, imag-axis read-off (theta=pi) Q-S3: Corollary 1.6 diagnostics (Suzuki line) |
✓ ✓ ✓ ✓ | 80 | 08-18-2026 |
| E-015 | Li coefficients arithmetic side n<=3000 (Bombieri-Lagarias Thm 2, zero-free) 3.2 continuation: Li coefficients from arithmetic side (extended to n_max=3000) |
✓ ✓ ✓ ✓ | 3000 | 08-18-2026 |
| E-016 | Herglotz fine structure of W at theta=pi vs F (Q-S3c) Q-S3c: Herglotz fine structure of W at theta=pi vs F |
✓ ✓ ✗ ✓ | 3519 | 08-18-2026 |
| E-016b | V3 root cause: winding diagnostic + recon variants 2.6: root-cause E-016 V3 Poisson-reconstruction failure |
— | 5600 | 08-18-2026 |
| E-016c | V3 repair: rescaled analytic + pair-augmented recon, ALL CHECKS PASS 2.6: V3 repair — stable rescaled analytic rho_W + pair-augmented Poisson reconstruction |
✓ ✓ | 18 | 08-18-2026 |
| E-017 | Cor 1.6 thinning remainder R_tau = W/F - R_0 direct measurement 2.7: direct measurement of Cor 1.6 thinning remainder R_tau = W/F - R_0 (D-026) |
✓ ✗ | 320 | 08-18-2026 |
| E-018 | descent profile |M|/|R0| on z=x+iy0, y0 in {1,0.3,0.1} 2.8: descent profile of |M|/|R_0| on z = x + i y0 (D-027 sec.3) |
✓ ✓ | 1944 | 08-18-2026 |
| E-019 | backend v2 validation + zero-flow at a=12 (e^24 blockwise sieve) 2.8(ii): backend v2 validation + zero-flow at a=12 |
✓ ✓ ✓ | 2129 | 08-18-2026 |
| E-020 | xi'-world ceiling adversary LP (pool approach, documented limitation) 1.6: xi'-world bandwidth-one ceiling (LawN256 analogue) |
✗ ✓ ✓ | 6 | 08-18-2026 |
| E-020b | window-class optimum kernel D1 (quartic is optimal; control recovers MT) 1.6: window-class optimum for kernel D1 |
✓ ✓ | 2 | 08-18-2026 |
| E-021 | Exact S_inf(n) to n=5000 + |S_f(n)| curve + D-036 contour constants (A=0.9,B0=1.0); ALL CHECKS PASS (V1 4.8e-25 vs E-015, V2, V3 c0=-0.25) 3.7: exact S_inf(n) + |S_f(n)| curve + D-036 constants |
✓ ✓ ✓ | 5000 | 08-18-2026 |
| E-022 | spectral-measure convergence (D-035 D2): Clark weights 1/K(z_k,z_k) for Suzuki D_{a,pi} 2.10 (D-035 D2): spectral-measure convergence |
✓ ✓ ✓ ✓ | 3815 | 08-19-2026 |
| E-022b | valley analysis: K-diagonal dips at gammas, saturation Delta_inf=3.33 2.10 analysis: K-diagonal valleys + mass concentration |
✓ ✓ ✓ | 4600 | 08-19-2026 |
| E-023 | finite-height assembly (M0 short-segment, Sinf lower-bound validation, margin/N0/N1) 3.8 support: D-037 assembly (short-segment max, S_inf lower bound validation, margin/N0/N1) |
✓ ✓ ✓ | 20000 | 08-19-2026 |
| E-023b | MZeroProp interval-rigorous box-cover certification MZeroProp interval-rigorous certification (ii'-7 follow-up, user directive session 043) |
✓ ✓ ✓ ✓ | 8 | 08-26-2026 |
| E-024 | theta-sweep: theta-independence of consolidation stats; linear-K law 2.12: theta-flow numerics (theta-independence test) |
✓ ✓ ✓ | 14949 | 08-19-2026 |
| E-025 | lambda-flow: discretization floor found; locking correction; channel response confirmed but machinery-limited 2.13: lambda-flow numerics (ground-state channel) |
✗ ✓ ✗ ✓ | 16560 | 08-19-2026 |
| E-026 | inverse-spectral reconstruction: Jacobi coefficients of Clark measures (moment round-trip 3e-14); signal diffuse inverse-spectral reconstruction (Jacobi/Stieltjes coefficients of the Clark measures) |
✗ ✓ ✓ | 720 | 08-19-2026 |
| E-027 | valley-detector stress test: bare onset immediate; tuned-carrier threshold pi/(2a) validated to <1%; pencil onset measured; localization at gamma* | ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ | 100 | 08-19-2026 |
| E-028 | traced R-bound verification (15 pts, 17x slack) + margin rerun A=6.0/B0=40: N0=58998, adopted 60000, margin(N1)=2.458e26 4.5a support: D-056 traced R-bound verification + margin rerun with A=10.5, B0=48 (honest C0 incl. log log term) |
✓ ✓ ✓ | 19991 | 08-19-2026 |
| E-029 | convexity proof verification: sympy table match, psi_i<0, margin-tilde convex, N0_tilde=75621, margin~(8e4)=16020>0 4.5b support: D-057 convexity proof verification |
✗ ✓ ✓ ✓ ✓ ✓ | 19991 | 08-19-2026 |
| E-030 | honest-Bn-fold | ✓ ✓ ✓ ✓ | 14 | 08-24-2026 |
| E-031 | right-tail-stepping | ✓ ✓ ✓ | 8 | 08-25-2026 |
| E-032 | F-D side thread 6-R-0: TRUE complex right-tail horizontal integral of Lagarias Thm 6.1 at T=sqrt(n) vs the E-031 envelope | ✓ ✓ ✓ ✓ | 9 | 08-25-2026 |
| E-033 | F-D side thread 6-R-1: cross-validation of repaired-stepping constants (Z0 certified, q-table, R(c) vs E-032 measured, kernel charge, fold fit, margins at c=1.5 and 2) | ✓ ✓ ✓ ✓ ✓ ✓ ✓ | 29 | 08-25-2026 |
| E-034 | D-074 dog-leg DL-0: exact deformation identity verified (resid 2-5e-5), piece majorants with 4-13x slack, temporary dog-leg Bn_DL margin > 0 to N1 = 9e24 (margin(N1) = 2.458e26) | ✓ ✓ ✓ ✓ ✓ ✓ ✓ | 24 | 08-25-2026 |
| E-035 | Grid check of Paper A Appendix C: three zero-counting perimeter Props derived from Rosser Thm 19 + Trudgian-II (1.2)/Cor.1/Thm 1 | ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ | 10 | 09-01-2026 |