|
%I #9 May 11 2019 11:01:58
%S 1,1,2,1,3,1,4,1,5,1,6,1,7,1,8,2,2,3,2,4,2,5,2,6,3,4,3,5,3,6,4,4,5,4,
%T 6,5,5,6,6,7,3,7,4,7,5,7,6,8,3,8,4,8,5,8,6,9,3,9,4,9,5,9,6,10,3,10,7,
%U 7,8,7,9,7,10,8,8,9,8,10,9,9,10,10,11,3
%N Lexicographically earliest sequence of positive terms such that for any distinct m and n, a(m) + dup(a(m+1)) <> a(n) + dup(a(n+1)) (where dup corresponds to A020330).
%H Rémy Sigrist, <a href="/A308073/b308073.txt">Table of n, a(n) for n = 1..10000</a>
%e The first terms, alongside a(n) + dup(a(n+1)), are:
%e n a(n) a(n)+dup(a(n+1))
%e -- ---- ----------------
%e 1 1 4
%e 2 1 11
%e 3 2 5
%e 4 1 16
%e 5 3 6
%e 6 1 37
%e 7 4 7
%e 8 1 46
%e 9 5 8
%e 10 1 55
%o (PARI) s=0; v=1; for(n=1, 84, print1(v", "); for (w=1, oo, if (!bittest(s,x=v+w*(1+2^#binary(w))), s+=2^x; v=w; break)))
%Y See A308057 for other variants.
%Y Cf. A020330.
%K nonn,look,base
%O 1,3
%A _Rémy Sigrist_, May 11 2019
|
