I began a table which, as far as it goes, looks like this:
Yellow rows are solvable (I think), blue are not, and green are undecided.
The number of configurations is given by 2 raised to the power of the number of slots in the grid. Unless, of course, we are ignoring rotations/reflections. In which case, the number of configurations is something else. (Starting with 1x1 has 2, 1x2 has 3, 1x3 has , and so on) Actually, that way of counting seems to make a bit more sense; I'll add that to the table later.
So far, it seems that unsolvable ones can be found in the form of 2s and 3s with 2s on the "outside." It also seems that one dimension has to be even, or a power of two but I don't have enough to see if this is a pattern yet.

Good start! Can you figure out the number for an nxm grid?
ReplyDeleteIn your example column, you're stating the numbers in the puzzle. Do you still have the x's and o's that lead to these unsolvable puzzles? Those might be good as well. Perhaps as another column...
I believe you have 1 yellow error and 1 blue error in the solvable or not column.
(Also, I don't think you're going to want to finish this table, figuring out 7x4 one way or the other sounds really exhausting if you're doing it by trial and error, focus on the littler ones, as I think you've been doing...)
Great work!
~Susa