Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated

Lebowitz lockard 2025 increasing sequences decreasing prime factors

../


N. Lebowitz-Lockard, Increasing sequences with decreasing prime factors, Notes Number Theory Discrete Math. 31 (2025), no. 3, 635--638; DOI 10.7546/nntdm.2025.31.3.635-638. Received 22 April 2025, revised 14 September 2025, accepted and published online 16 September 2025; MSC 11A05, 11A41.

The retained folder-name PDF is the publisher PDF, 4 pages with a text layer, printed pp. 635--638 (physical p. nn is printed p. 634+n634+n). Provenance: retained from the repository's survey download set of September 2026; the download URL was not recorded, but the identifier is the DOI https://doi.org/10.7546/nntdm.2025.31.3.635-638 printed on the first page. 183,395 bytes. The file prints "Copyright © 2025 by the Author. This is an Open Access paper distributed under the terms and conditions of the Creative Commons Attribution 4.0 International License (CC BY 4.0). https://creativecommons.org/licenses/by/4.0/" on its first page, the Creative Commons Attribution 4.0 license.

Read status: claims checked. The abstract and Theorems 1.3 and 1.4 were read clause by clause on the text layer, and the four-line proof of Theorem 1.3 was read; the proof of Theorem 1.4 was not checked.

Contents

For an arithmetic function ff let gf(x)g_f(x) be the largest kk for which there is a sequence a1<⋯<ak≤xa_1<\cdots<a_k\le x with f(a1)>⋯>f(ak)f(a_1)>\cdots>f(a_k); the note writes g−(x)=gP−(x)g^-(x)=g_{P^-}(x) for the smallest prime factor P−P^-.

  • Pages 635--636 recall Erdős's question for the largest prime factor, with Cambie's bounds as Theorem 1.1 (p. 635), 2x/log⁡x≲g(x)≲22x/log⁡x2\sqrt{x/\log x}\lesssim g(x)\lesssim2\sqrt2\sqrt{x/\log x}, and Pollack, Pomerance and Treviño's bounds for Euler's function as Theorem 1.2 (p. 636), x0.19≤gφ(x)≤xexp⁡(−(12+o(1))log⁡xlog⁡log⁡x)x^{0.19}\le g_\varphi(x)\le x\exp(-(\tfrac12+o(1))\sqrt{\log x\log\log x}); it cites Tao (its [12]) and others for the variants in which φ\varphi is constant or increasing.
  • Theorem 1.3 (p. 636): g−(x)≲2x/log⁡xg^-(x)\lesssim2\sqrt x/\log x. Proof (p. 636): every term after the first is composite, so the smallest prime factors are distinct primes at most x\sqrt x, and k≤π(x)+1k\le\pi(\sqrt x)+1.
  • Conjecture 1.1 (p. 636): the largest gap between consecutive primes up to xx is ≪(log⁡x)2\ll(\log x)^2 (a weak form of Cramér's conjecture, with Granville's papers cited for the discussion).
  • Theorem 1.4 (p. 636; proof p. 637): under Conjecture 1.1, g−(x)≫x/(log⁡x)2g^-(x)\gg\sqrt x/(\log x)^2, by products qipiq_ip_i with pip_i the first primes above x/2\sqrt x/2 and primes qi≥piq_i\ge p_i. Page 637 notes that the Pollack--Pomerance--Treviño sequence gives g−(x)≫x0.19g^-(x)\gg x^{0.19} unconditionally, and page 638 that an unconditional g−(x)=x1/2+o(1)g^-(x)=x^{1/2+o(1)} would follow from prime gaps of size xo(1)x^{o(1)}.

Compiled scope

The whole four-page note was read on the text layer; only the proof of Theorem 1.3 was followed. Nothing here is independently reviewed.

Bears on. #49, as the 2025 paper the page records as citing Tao without improving the totient maximum: it concerns sequences whose smallest prime factors decrease and quotes the Euler function bounds only as context, so it says nothing about the problem's strictly φ\varphi-increasing sequences.