[seqfan] Re: 3D version of A000938: 3-in-line inside the nXnXn cube

[Douglas McNeil]

On Sun, May 23, 2010 at 11:03 PM, Ron Hardin <rhhardin at att.net> wrote:
> I've set up a data dump at
> http://rhhardin.home.mindspring.com/current.txt

I'm curious how you're computing the numbers so efficiently.  My first
thought was to fix a corner, iterate over all possible second cells,
and then extend along all multiples. Then translate the triples
through the cube, and finally impose the octahedral symmetry, but I'd
have to be careful about that.  Then I thought of doing it recursively
(as already mentioned in the thread) and computing only the extra
lines introduced by the "next" cube layer.  But the first way didn't
seem practical, and the second hurt my head.


Doug

-- 
Department of Earth Sciences
University of Hong Kong