[seqfan] A000959: Lucky numbers

[hv at crypt.org]

hv at crypt.org wrote:
:Given my perl code managed up to 1e6 easily enough, I think rewriting in C
:would let me generate a list of the lucky numbers up to 1e8 fairly easily,
:maybe closer to 1e9.

The file <http://crypt.org/hv/maths/lucky_1e8.bz2> lists the lucky numbers
up to 1e8. The file has 5,286,238 entries; filesize is 46,884,917 bytes,
compressed to 16,490,206 bytes.

I verified the results up to 5e6 with code provided by David Wilson that
used a completely different algorithm, so I have high confidence in these
results. My code is at <http://crypt.org/hv/maths/cgen_lucky.c> (2,762
bytes).

The program has now reached 3e8 after 17 CPU hours; I'll let it run until
1e9 or until I need the processor back, and make the full results available
as a separate file. (I estimate it'll take >6 CPU days to reach 1e9: the
slowdown for large n is rather worse than I'd guessed based on small n.)

Neil, I'm not sure how useful something this size is as a bfile, but
feel free to take all or part of it. I anticipate it will mostly be of use
for extending existing or new derivative sequences.

Hugo