Answers are grounded in this paper and cite their sources.
We revisit the self-contained lower bound for Maynard's ratio ρ(F) in Sections 6.8–6.9 of A. Granville's survey "Primes in intervals of bounded length" (Bull. Amer. Math. Soc. 52 (2015), 171–222). (i) The numerator of ρ(F) is rewritten there as L_k − M_k, but in fact k∫(A−B)² ≤ L_k − kM_k with strict inequality: this is an upper bound where the argument needs a lower bound, and a factor k is missing from M_k. (ii) The parameter η in Section 6.9 is misprinted; the final inequality (6.8) is nevertheless correct, by an explicit polynomial identity. (iii) We give two rigorous repairs. Young's inequality gives ρ(F) ≥ (1/2)log k − 2, proving Theorem 6.4 under k > e^(8m+4). Truncating the support in k−1 variables gives ρ(F) ≥ log k − 2 log log k − 1/2 for k ≥ e^7, which recovers Theorem 6.4 and Corollary 6.5 (B_m = e^(8m+5)) exactly as stated. Values of ρ(F) computed by FFT convolution confirm that the gap is in the proof and not in the conclusion. (iv) In Part 2 we correct a Fourier cutoff in Section 9.3 and two misprints, and resolve a notational ambiguity in Section 11.2. The record includes the note (PDF and LaTeX), a verification script (symbolic and numerical), its output and the figure.
No takes yet. Share an insight, caveat, or question.
Ruqing Chen (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: