|
%I #15 Feb 14 2019 12:28:55
%S 34,46,62,74,86,94,142,188,202,214,218,231,243,244,262,278,302,356,
%T 375,394,422,423,428,436,446,459,478,494,584,596,604,621,628,634,664,
%U 698,716,837,861,885,903,916,922,944,956,982,1004,1017,1018,1028,1042,1052
%N Numbers n such that tau(n) = (tau(n+1)+tau(n-1))/2.
%H Harvey P. Dale, <a href="/A074757/b074757.txt">Table of n, a(n) for n = 1..1000</a>
%H Vaclav Kotesovec, <a href="/A074757/a074757.jpg">Plot of a(n)/n for n = 1..1000000</a>
%H Vaclav Kotesovec, <a href="/A074757/a074757_1.jpg">Plot of a(n)/(n*log(log(n))) for n = 1..1000000</a>
%F a(n) seems to be asymptotic to c*n with c around 23. [This conjecture is false. The limit a(n)/n, if it exists, is > 29, see graphs. - _Vaclav Kotesovec_, Feb 14 2019]
%t okQ[{a_,b_,c_}]:=(a+c)/2==b; Flatten[Position[Partition[DivisorSigma[0, Range[1100]],3,1],_?okQ]]+1 (* _Harvey P. Dale_, Aug 17 2013 *)
%Y Cf. A090502.
%K easy,nonn
%O 1,1
%A _Benoit Cloitre_, Sep 28 2002
|
