A357762 Decimal expansion of -Sum_{k>=1} A106400(k)/k.

1, 1, 9, 6, 2, 8, 3, 2, 6, 4, 3, 2, 5, 2, 5, 6, 4, 3, 7, 2, 2, 2, 2, 9, 1, 6, 3, 3, 2, 0, 0, 8, 1, 9, 1, 8, 1, 0, 1, 0, 4, 2, 6, 7, 4, 6, 4, 0, 1, 5, 9, 4, 3, 8, 1, 8, 9, 8, 7, 2, 3, 3, 3, 7, 3, 0, 7, 8, 3, 7, 5, 1, 6, 1, 0, 9, 1, 5, 8, 0, 8, 7, 7, 7, 9, 1, 1, 9, 6, 4, 5, 4, 6, 2, 1, 1, 0, 7, 4, 8, 9, 6, 3, 3, 3

Offset: 1

Author

Amiram Eldar, Oct 12 2022

Keywords

• nonn
• cons

Comments

The asymptotic mean of the excess of the number of odious divisors over the number of evil divisors (A357761, see formula).

The convergence of the partial sums S(m) = -Sum_{k=1..2^m-1} A106400(k)/k is fast: e.g., S(28) is already correct to 100 decimal digits (see also Jon E. Schoenfield's comment in A351404).

Examples

			1.19628326432525643722229163320081918101042674640159...
		

Crossrefs

Cf. A106400, A357761.

Similar constants: A215016, A351404

Programs

Mathematica

sum = 0; m = 1; pow = 2; Do[sum -= (-1)^DigitCount[k, 2, 1]/k; If[k == pow - 1, Print[m, " ", N[sum, 120]]; m++; pow *= 2], {k, 1, 2^30}]

PARI

default(realprecision, 150);
sm = 0.; m = 1; pow = 2; for(k = 1, 2^30, sm -= (-1)^hammingweight(k)/k; if(k == pow - 1, print(m," ",sm); m++; pow *= 2))

Formula

Equals -2 * Sum_{k>=1} A106400(2*k-1)/(2*k-1).

Equals lim_{m->oo} (1/m) * Sum_{k=1..m} A357761(k).