Answers are grounded in this paper and cite their sources.
Bombieri and Ghosh computed the supremum of the real parts of the zeros of the Davenport–Heilbronn function and of its sieved versions: 1.120362 for the function itself, and 1.046712, 1.063679, 1.093482 and 1.106200 for q = 2, 3, 7, 13. We recover these numbers from the coefficients of −f′/f alone, computed exactly to n = 6·107, without computing any zero. A first registered test, comparing raw growth exponents with the suprema at tolerance 0.002, failed: the raw exponents overshoot by 0.009 to 0.023. The overshoot is described for all five functions at once by one empirical correction, exp(−K/((σ−1) log x)) with K ≈ 0.27–0.31; fitted on x ≤ 2·107 and tested blind on 2·107 < x ≤ 6·107, the corrected model held 4 of 4 registered criteria. The principle (the growth exponent is the rightmost singularity) is classical; we did not find the measurement or the correction law stated in the literature searched. K is empirical. Written jointly with Claude, an AI system made by Anthropic, which carried out much of the computation, the derivations and the drafting; the human author is responsible for the content. Not peer reviewed. The package contains the PDF, its LaTeX source, and the scripts, registered predictions and result files it cites.
No takes yet. Share an insight, caveat, or question.
Nathan Berzai (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: