pjb wrote:I saw this puzzle somewhere given as a great example of POM analysis. Following basics, POM first finds two 3's at 1,1 and 4,4, and then no less than 20 9's. After this the puzzle solves easily.
As you may have guessed from my previous posts, what I'm interested in is comparing various ratings of puzzles based on various coherent families of rules.
All the consistent rating systems I've studied, give the same rating to almost all (in the sense of unbiased stats) the puzzles.
In this exceptional example, it appears that the following two ratings both give the same result, 5:
1) the gW rating
2) a tentative rating that would be based only on basic interactions, generalized Subsets (with Finned, Franken and Kraken Fish) and XYZ-Wing .
I'm curious: what's the maximum size of the patterns used in your POM solution?
As for the number of candidates eliminated by a single pattern, this has never impressed me much: often, people like to exhibit artificially big patterns by aggregating smaller ones; as a result, most of, if not all, their eliminations could be done by simpler patterns. Notice that this is the case in the Hodoku solution (but Hodoku is a great solver anyway !)