[seqfan] Re: A057652 has no more terms below 10^6

[Maximilian Hasler]

> I don't know if anyone has studied the equivalent of Gilbreath's conjecture
> for lucky numbers, but it appears that instead of a line of 1s, you get a
> line of alternating 2s and 0s, but there can be two or three consecutive 2s
> or 0s in a row. I haven't been able to discern any pattern to this
> alternation.
>

Indeed the terms preceding the first term >2 are always 0 or 2 since
all lucky numbers are odd
(while the primes start with the even 2 which leads to the initial "1"
of each row).

The equivalent of
oeis.org/A036277 		Position of first term > 2 in n-th row of Gilbreath
array shown in A036262.

for lucky numbers would be :
2,2,5,11,12,14,13,14,25,28,27,26,25,24,23,102,101,105,145,144,143,142,141,140,139,138,137,136,138,422,421,420,419,418,417,416,416,430,429,428,427,426,430,2332,2331,2332,2331,
...
(next term > 100 000).

The growth behaviour seems at least as fast as that of A036277.

Maximilian