cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A382357 Lexicographically earliest sequence of distinct positive integers such that the 2-adic valuations of adjacent terms differ exactly by one.

This page as a plain text file.
%I A382357 #11 Mar 26 2025 16:17:03
%S A382357 1,2,3,6,4,8,12,10,5,14,7,18,9,22,11,26,13,30,15,34,17,38,19,42,20,24,
%T A382357 16,32,48,40,28,46,21,50,23,54,25,58,27,62,29,66,31,70,33,74,35,78,36,
%U A382357 56,44,72,52,82,37,86,39,90,41,94,43,98,45,102,47,106,49
%N A382357 Lexicographically earliest sequence of distinct positive integers such that the 2-adic valuations of adjacent terms differ exactly by one.
%C A382357 The first term with a given 2-adic valuation, say k, is necessarily 2^k.
%C A382357 Empirically, powers of two appear as pairs of consecutive terms.
%C A382357 We cannot have three consecutive powers of 2: if a(n) = 2^k and a(n+1) = 2^(k+1) then a(n+2) <= 3*2^k < 2^(k+2).
%C A382357 All powers of two appear in the sequence:
%C A382357 - by contradiction: if 2^m is missing, then the 2-adic valuation of the terms of the sequence is bounded by m,
%C A382357 - by necessity, we have some k < m such that all the integers with 2-adic valuation k appear in the sequence,
%C A382357 - hence all integers with 2-adic valuation k+1 (and k-1 provided k > 0) will appear in the sequence,
%C A382357 - gradually, all integers with 2-adic valuation k+2, k+3, etc. and eventually 2^m, will appear, a contradiction.
%C A382357 Conjecture: this sequence is a permutation of the positive integers.
%C A382357 The fact that A007814 contains every positive integer infinitely many times is not sufficient to guarantee that the present sequence is a permutation of the positive integers (the variant based on A003602 instead of A007814 contains only finitely many even numbers, and so is not a permutation of the positive integers, although A003602 contains every positive integer infinitely many times).
%H A382357 Rémy Sigrist, <a href="/A382357/b382357.txt">Table of n, a(n) for n = 1..10000</a>
%H A382357 Rémy Sigrist, <a href="/A382357/a382357.png">Colored scatterplot of the first million terms</a> (blue pixels for even n's, red pixels for odd n's)
%H A382357 Rémy Sigrist, <a href="/A382357/a382357.gp.txt">PARI program</a>
%e A382357 The initial terms are:
%e A382357   n   a(n)  A007814(a(n))
%e A382357   --  ----  -------------
%e A382357    1     1              0
%e A382357    2     2              1
%e A382357    3     3              0
%e A382357    4     6              1
%e A382357    5     4              2
%e A382357    6     8              3
%e A382357    7    12              2
%e A382357    8    10              1
%e A382357    9     5              0
%e A382357   10    14              1
%e A382357   11     7              0
%e A382357   12    18              1
%e A382357   13     9              0
%e A382357   14    22              1
%e A382357   15    11              0
%o A382357 (PARI) \\ See Links section.
%Y A382357 Cf. A003602, A007814, A073675, A266089, A382360.
%K A382357 nonn,base
%O A382357 1,2
%A A382357 _Rémy Sigrist_, Mar 22 2025