[seqfan] Re: More rows of triangle in A271100?

[Bruce Leenstra]

Hi David et al,

The good news: my search works for ever larger prime limits. The bad news:
It's an O(n^2) problem and the arc connections are thinning out fast.
Here's a table of results:

_Limit Primes Cycle Rows Terms CPU_time
_30000 _3248 _ 101 _ 39 _ 780 _129.80 s
_60000 _6057 _ 146 _ 50 _1275 _488.75 s
120000 11301 _ 163 _ 57 _1653 1581.49 s
180000 16342 _ 164 _ 57 _1653 5460.76 s

*Cycle: primes in cyclic paths.

You'll notice that only one of the last 5041 primes had both in-arcs and
out-arcs and it didn't form a 58-tuple. I'm working on saving/reloading the
digraph and then extending it in batches. And rather than resubmitting the
b-file over and over I'm keeping the largest here:
https://drive.google.com/file/d/0B3Gc9XcQbCj-QzR1T0xFaEQ0eG8/view

Best Regards,
Bruce