back to table · edit · history · where entries came from · files · download
3706 bytes, as of the version from 2026-11-18 20:02 (current). Recorded here, not run.
# Research for the Riemann zeta screw-function values
Checked source: Suzuki arXiv:2206.03682v4 (1.1), the following origin paragraph,
(1.8), Theorem 1.2 and its following paragraph, and (3.9)-(3.11). These fix the
real even g=-Psi, half-triangle normalization, and the distinction between the
named function and unproved global screw-class positivity. The introduction
of arXiv:2606.09096v1 fixes W, including both prime sums, the pole term and
Euler's constant. These passages were read, not inferred from the proposal.
Checked deduction: differentiating the finite prime term gives the signed
jump g'(log(q)+)-g'(log(q)-)=Lambda(q)/sqrt(q). The origin and all prime-power
logarithms through log(256) are the 71 selected inputs. Sieve/power generation
and independent factorization agree. See argument-selection.md and its
exhaustive groups in selection-plan.json.
Live discovery: catalogue.json contains the single cached catalogue pass,
including own drafts; screen.json contains fresh already_here/already_asked
results. Only #344 asks for screw-function tables. The preclaim search had no
screw table. Three sample numeric api/lookup requests had no results. T733 is
the Gaussian heat/Weil functional; T752/T753 are centered-zero secondary-zeta
transforms, not this function. Their actual documents were fetched. Existing
sound tables were left alone. T3 was read as a spectral-control lead, not used
as numerical evidence. The HSW zero-count paper, Corollary 1.2, was inspected
as another lead; no spectral-tail experiment was executed.
Provenance: #344 explicitly contributes to #275 for this proposal; it does not
answer or close it. The #275 issue body was inspected through the GitHub
connector. The initial creation/edit message preserves precisely #344 and
#275. No unrelated family-wide wanted issue is claimed as an origin.
Tags: both live /tags pages and exact tag API lookups were checked. L-function,
number theory, special values, integral and functional analysis apply. No
registered screw function/functions or case variant existed. The new tag
screw functions names Suzuki's construction without asserting global
positivity. No alias merge or tag removal is proposed.
Checked numerical facts: every production ball is finite and supports at least
479 relative bits; every independently integrated nonzero value supports at
least 421 bits. All 71 stored intervals contain the separate quadrature
balls. Native-type/dry-run checks and nonzero/wrong-input controls pass.
check-results.json, production-balls.json, numerical-evidence.md and versions.json
record scope and bounds. Re-running the generator is not the independent
mathematical check. A fresh independent actor's verdict remains outstanding.
Cross-links: T733 is the only current Similar tables record, and its different
test normalization is stated. The planned moment companion in #344 does not
yet have an assigned T-number; its eventual directed link should explain
mu_n=(1/4) integral_0^infinity exp(-t/2)*Psi(t)*t^n dt. Do not fabricate a
companion URL. The two determinant proposals remain separate and unbuilt here.
No unresolved definition alternative remains. No proof of RH or all-support
positivity is claimed.
Large artifacts are in this run's working directory. Source and evidence
hashes are recorded in evidence-sha256.json; the skill binding is in
skill-verification.json and the archived manifest named there.
Preserved primary-source bytes:
- suzuki-v4.html (https://arxiv.org/html/2206.03682v4), SHA-256 fa4c752d0e848e7e41fbebe1bd69d8d6a53712ab76186f13b64f98c32d471100
- weil-v1.html (https://arxiv.org/html/2606.09096v1), SHA-256 ae7177995b58484d076ed76bf2cb81081c88d569d8d4e4b64503c0ba3c0e615e