|
I was hoping for something more symmetrical, so I ran it again...
The second time produced the second image. It's slightly better, 2.9159 vs 2.9215.
The limiting factor appears to be the three squares in a row at the top
of the image. They are the same in both solutions while the rest of the
arrangement appears to have some freedom.
The third attempt came up with a different arrangement. This one
actually has a ratio of 2.9122, so it is about 0.3% better. Too close
to call really.
Non of them are particularly symmetrical though.
The octagons don't line up because one of the degrees of freedom in
the simulated annealing process is to rotate the container leaving the
contents in place.
|