Wiki
Wiki

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

Updated


Claim. W. van Doorn, On the length of an interval that contains distinct multiples of the first nn positive integers, Integers 26 (2026), #A7 (received 19 February 2025, accepted 27 November 2025, published 5 January 2026, the page name's date; also arXiv:2601.16972v1). Its Theorem 1 states that for all large enough nn,

max⁡mf(n,m)−f(n,n)>0.36 nlog⁡nlog⁡log⁡n,\max_mf(n,m)-f(n,n)>0.36\,n\frac{\log n}{\log\log n},

with f(n,m)f(n,m) the least integer such that (m,m+f(n,m)](m,m+f(n,m)] contains distinct a1,…,ana_1,\ldots,a_n with i∣aii\mid a_i; the site's open interval (m,m+f(n,m))(m,m+f(n,m)) makes every f(n,m)f(n,m) one larger and leaves the difference unchanged. So the second question of Problem 711, Erdős's unproved (6) of 1992, max⁡m(f(n,m)−f(n,n))→∞\max_m(f(n,m)-f(n,n))\to\infty, is answered yes in this quantitative form, and the theorem's second sentence restates it: for all large nn there is an interval of length 0.36 nlog⁡n/log⁡log⁡n0.36\,n\log n/\log\log n that contains no distinct multiples of 1,2,…,n1,2,\ldots,n. The proof is half a page: an elementary shift lemma, kn+f(kn,kn)≤k2n+f(n,k2n)kn+f(kn,kn)\le k^2n+f(n,k^2n) for all k,nk,n, combined with the Erdős–Pomerance bounds (2/e+o(1)) nlog⁡n/log⁡log⁡n<f(n,n)<(2+o(1)) nlog⁡n(2/\sqrt e+o(1))\,n\sqrt{\log n/\log\log n}<f(n,n)<(2+o(1))\,n\sqrt{\log n} at k=⌈0.6log⁡n/log⁡log⁡n⌉k=\lceil0.6\sqrt{\log n/\log\log n}\rceil. The library's result page records the statement, the two lemmas and the proof pointer; nothing is independently reviewed by this project.

Covers. The second question only. The first question, the conjectured max⁡mf(n,m)≤n1+o(1)\max_mf(n,m)\le n^{1+o(1)}, is untouched: the theorem gives the lower bound max⁡mf(n,m)≫nlog⁡n/log⁡log⁡n\max_mf(n,m)\gg n\log n/\log\log n, which is consistent with the conjecture and far below the published upper bound n3/2n^{3/2}.

Acceptance. Refereed: published in Integers, a refereed journal, with the dates above. The site's curator's commentary (page last edited 11 January 2026) records the paper's affirmative answer to the second question, after a thread comment of 6 January 2026 announcing the journal reference; the site's label OPEN attaches to the pair of questions and is not acceptance evidence. Nothing here is this project's own review.

Depends on. Erdős and Pomerance (1980), Theorem 2 and Erdős and Pomerance (1980), Theorem 3.