olymmath

ISL 2020 N4

    For any odd prime pp and any integer nn, let dp(n)∈{0,1,…,p−1}d_{p}(n) \in\{0,1, \ldots, p-1\} denote the remainder when nn is divided by pp. We say that (a0,a1,a2,…)\left(a_{0}, a_{1}, a_{2}, \ldots\right) is a pp-sequence, if a0a_{0} is a positive integer coprime to pp, and an+1=an+dp(an)a_{n+1}=a_{n}+d_{p}\left(a_{n}\right) for n⩾0n \geqslant 0. (a) Do there exist infinitely many primes pp for which there exist pp-sequences (a0,a1,a2,…)\left(a_{0}, a_{1}, a_{2}, \ldots\right) and (b0,b1,b2,…)\left(b_{0}, b_{1}, b_{2}, \ldots\right) such that an>bna_{n}>b_{n} for infinitely many nn, and bn>anb_{n}>a_{n} for infinitely many nn ? (b) Do there exist infinitely many primes pp for which there exist pp-sequences (a0,a1,a2,…)\left(a_{0}, a_{1}, a_{2}, \ldots\right) and (b0,b1,b2,…)\left(b_{0}, b_{1}, b_{2}, \ldots\right) such that a0<b0a_{0}<b_{0}, but an>bna_{n}>b_{n} for all n⩾1n \geqslant 1 ? (United Kingdom)

    Official text Transcription from NuminaMath olympiads, checked word for word against the official paper. Quoted with source.

    Official text: IMO 2020 Shortlist booklet, page 12.

    Solutions

    Official solution Solution 1 by the IMO Problem Selection Committee

    The terms of the source let olymmath link to this solution but not copy it. Read the solution (IMO official site).

    Locator isl/2020/n4 · Tag Z4FNHXM0TC · Edit this page on GitHub · Keys: j next, k previous