by mchassoc » Thu Aug 06, 2015 11:57 pm
Jason, thank you so much: the Sudopedia explanation is both succinct, complete in explanation, and the reasoning is clearly presented. It is undoubtedly my lack of completely following previously read descriptive reasoning behind the method that did not allow consistent results. I thought only two of the three had to "see" one another (for which XY wings would be much easier to spot), and that this would allow deleting the common candidate that could see the two pincers. Its now obvious why that is not feasible.
My wife and I were introduced to Sudoku about 5 years ago and have made good progress in solving some of the most difficult puzzles. We have completed solutions in Steve Soto and Monica Jones "Soduko Puzzles, Hard to Very Hard", and are presently more than 1/3 through Frank Longo's "Beyond the Black Belt", encountering very few puzzles that we cannot solve. Most of these I have been able to solve using only Chad Barker's "1 to 9" technique over and over. But on a few a more sophisticated "trick" was required. I am just now starting to investigate solutions based on use of the SAS method.
Early in learning Sudoku I wrote a short computer macro to populate all candidates for all cells in any given puzzle. Once a cell candidate, or group of candidates, is determined the macro can be rerun, eliminating unnecessary candidates from all cells. These are eliminated based on "no legitimate candidate can occupy more than one cell on the board". Whereas we don't use this "Sudoku Aid" in the solving process, it is comforting to use it as a check.
Thanks again for solving my quandary re XY wings.
M.C. Horton (mchassoc)