login

Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).

Record n for the sequence A006666(n)/log(n), for n > 1; i.e., the n at which the number of steps to 1 for the halving-Collatz iteration divided by log(n) sets a new maximum.
0

%I #25 Jul 21 2017 15:36:56

%S 2,3,7,9,27,230631,626331,837799,1723519,3732423,6649279,8400511,

%T 63728127,3743559068799,100759293214567,104899295810901231

%N Record n for the sequence A006666(n)/log(n), for n > 1; i.e., the n at which the number of steps to 1 for the halving-Collatz iteration divided by log(n) sets a new maximum.

%C These numbers (for n > 2, at least) are called "Abby-Normal" numbers by Dominici in the referenced paper (definition 21); 2 is included here to agree with the RECORDS transform convention. - _D. S. McNeil_, Mar 26 2009

%C Three further terms are known (see Roosendaal) but they may not be the next three terms. - _N. J. A. Sloane_, Oct 21 2012

%H D. Dominici, <a href="http://arXiv.org/abs/0704.1057">Working with 2s and 3s</a>, arXiv:0704.1057

%H Eric Roosendaal, <a href="http://www.ericr.nl/wondrous/comprecs.html">3x+1 Completeness and Gamma Records</a> - From _N. J. A. Sloane_, Oct 21 2012

%K nonn

%O 1,1

%A Diego Dominici (dominicd(AT)newpaltz.edu), Apr 16 2007

%E Definition edited by _D. S. McNeil_, Mar 26 2009

%E Extended by _D. S. McNeil_, Mar 26 2009

%E a(13) from _Donovan Johnson_, Feb 16 2011

%E a(14)-a(16) from Roosendaal's table added by _N. J. A. Sloane_, Oct 21 2012