> On 21/09/2011 21:00, RichD wrote: > >> On Sep 20, Willem<wil...@toad.stack.nl> wrote: >> >>> )> >>>The REAL issue is that to solve a Sudoku grid, there are >>> actually TWO >>> )> >>>parts: (a) find a solution (i.e. prove that one >>> )> >>>exists), but also (b)prove that it is the ONLY >>> )> >>>solution. >>> )> >>> )> I have done sudoku puzzles in which I have gotten >>> )> to a point from which I could produce two >>> )> solutions. >>> ) >>> ) What is the smallest number of initially >>> ) fixed cells such that the solution is >>> ) guaranteed unique? >>> >>> 78, I think. Or 17. I'm not sure what you wanted to ask exactly. >> >> >> It's a bit tricky to phrase precisely, though a simpe question. >> >> What is the smallest number of initially determined cells, >> such that there is at least one puzzle of such configuration, >> with a unique solution? >> >> And you claim there is a known example of 17? > > > http://mapleta.maths.uwa.edu.au/~gordon/sudokumin.php claims to list > about 50,000 known examples. >
I remember hearing long ago that, empirically, 17 was the smallest number amogst sudokus prodcued to date. Has it actually been proven that there can be no fewer than 17?
-- Stephen J. Herschkorn firstname.lastname@example.org Math Tutor on the Internet and in Central New Jersey and Manhattan