[seqfan] Re: A102421; Start with 2n+1, multiply by 3 and add 1 and divide out any power of 2; ....

Neil Sloane njasloane at gmail.com
Sun Aug 23 20:03:10 CEST 2020

```Since I created that sequence, let me defend it.

Here is Maple:

f:=proc(n) local j; j:=3*n+1; while j mod 2 = 0 do j:=j/2; od: j:=3*j-1;
while j mod 2 = 0 do j:=j/2; od: j; end;

and that generates the sequence, look:

[seq(f(2*n+1),n=0..60)];

[1, 7, 1, 1, 5, 25, 7, 17, 19, 43, 1, 13, 7, 61, 1, 35, 37, 79, 5, 11, 23,
97,

25, 53, 55, 115, 7, 31, 1, 133, 17, 71, 73, 151, 19, 5, 41, 169, 43, 89,

91, 187, 1, 49, 25, 205, 13, 107, 109, 223, 7, 29, 59, 241, 61, 125,
127,

259, 1, 67, 17]

Looks OK to me.

Best regards
Neil

Neil J. A. Sloane, President, OEIS Foundation.
11 South Adelaide Avenue, Highland Park, NJ 08904, USA.
Also Visiting Scientist, Math. Dept., Rutgers University, Piscataway, NJ.
Email: njasloane at gmail.com

On Sun, Aug 23, 2020 at 9:57 AM David Corneth <davidacorneth at gmail.com>
wrote:

> Hi,
>
> Can I get an extra pair of eyes over at A102421? It seems data was
> inconsistent with a comment even before I got there.
> It describes a function that, well, takes a number and then returns another
> number.
> Like after some magic, 17 becomes 19... according to a comment. But data
> says 17 becomes 19.
> Then 19 becomes 43.
>
> Upon inspection,
> a(8) = a((17 - 1)/2) = 19.
> a(9) = a((19 - 1)/2) = 43
>
> Could that be it?
> There's no reference to (n-1)/2 so that'd be guessing. Or should that be
> read from take 2n+1 in name?
> One could say if x = 2n + 1 then n = (x-1)/2 but to me that doesn't follow
> from the name.
>
> Best,
> David
>
> --
> Seqfan Mailing list - http://list.seqfan.eu/
>

```