bit of editing wanted : watch out for n>=12

[wouter meeussen]

dear seqfans

here comes a potential trap,
kind of 'lure of the small numbers',

Table[
Numerator[2*BernoulliB[2*n]*(4^n - 1)/(2*n)]/
Numerator[SeriesCoefficient[
    Series[1/(1 + Cosh[x]) , {x, 0, 2n}], 2n - 2]],
{n, 1, 128}]

{1,1,1,1,1,1,1,1,1,1,1,17,1,1,1,1,1,1,1,527,1,1,1,1,
31,1,1,731,1,41,1,1,1,37,1333,17,1,1,1,31,1,1,1,17,73,1,
1,1,43,1271,59,629,1,73,2759,43,1,1,1,17,1,67,7519,1,31,
89,1,289,1,29020032511,1,10573,1,1,1,2227,486029,1,1,1829,73,
1,1,24640667,31,1,1,37,1,49261787,5461,17,1,1,2077,1,1,
617093,6497,527,1,137,1,157,19177,37,1,17,9263,113119,1,5461,
1,10609,31,17,73,1,5461,14647,89,1,1,629,31,899141,1,67}

so, watch out for that 12 th term.

Q: what other seqs start to differ beyond n=12 or abouts?

Neil,
[[I am sorry, but the terms "1,1,17,1,1,1,1,1,1,1,527,1 "
do not match anything in the table.]]

I'll submit asap.

W.


----- Original Message -----
From: "wouter meeussen" <wouter.meeussen at pandora.be>
To: "Brendan McKay" <bdm at cs.anu.edu.au>; <seqfan at ext.jussieu.fr>; "benoit" <abcloitre at wanadoo.fr>
Sent: Sunday, December 07, 2003 1:18 PM
Subject: Re: bit of editing wanted : are these equivalent?


> a spot of trouble:
>

*** clip ***

> Close, but no cigar,
> 'cause the ratios between latter and former are:
>
> {1,1,1,1,1,1,1,1,1,1,1,17,1,1,1,1,1,1,1,527,1,1,1,1,31}
> differences for n= 11,19 and 24 (offset 0)
> a Bernoulli-effect?
>
> W.
>