The New Sudoku Players' Forum

[Sue De Coq]

I've started to experiment with puzzles that possess double-diagonal symmetry. (Such puzzles necessarily possess the rotational symmetry of order 2 that is a feature of Pappocom puzzles). It's quickly become apparent that these puzzles are often very tough to solve because they contain remarkably long chains. I'm yet to come up with a good reason why this should be.

An example, which permits logical solution with chains:

Code: Select all

 3 . . | . . . | . . 2
 . 9 . | . 4 . | . 1 .
 . . 5 | 7 . . | 8 . .
-------+-------+------
 . . 1 | 5 . . | . . .
 . 8 . | . 9 . | . 4 .
 . . . | . . 3 | 7 . .
-------+-------+------
 . . 2 | . . 1 | 5 . .
 . 4 . | . 8 . | . 9 .
 6 . . | . . . | . . 3


The log from my solver for the third move reads like War And Peace.