Feedback and Questions

Theoretically it should be possible to use logic on all sudoku puzzles, but I don’t possess every Sudoku strategy, although I do try to incorporate new ideas as soon as possible. Currently, if I produce a large amount of random stock, about 0.5% will still be unsolvable, so it’s possible to produce many of these ‘unsolvables’. I don’t think there is a general solving strategy that is entirely logical, but people do post solutions which combine several strategies to get past bottlenecks and the solvers on the weekly 'unsolvable' do come up with some pattern based method for these examples. My job, when I get the time, is to work out what they have done and try and make a general principle I can add to the solver armory.



The problem is candidate density. If you look at my solver when it comes to 'Run out of known strategies' you will see most cells contain 3 or more candidates. Most of the advanced strategies and all the chaining ones require bi-value (2 in one cell) or bi-location (2 in one unit) to get anywhere. So there is plenty of room for more thought and ideas, which is the attraction of Sudoku – its very deep.

Great to hear you are enjoying the site. I don't have a Mac to test it on but I do have Safari for the PC which does work for me. Sounds like a browser bug if Take Step isn’t working...both mouse and keyboard ?

Yes there are many routes possible - unchecking some of my strategies will reveal alternatives, but even within a strategy it only returns the first or shortest example. I have tried to use logical and statistical techniques to order the strategies in terms of complexity and then return the easiest/shortest solve routes so I hope it is optimal, but humans also have intuition which I can't program. So you might find another way.

Hi Kate
This is a nice example of Simple Colouring in Sudoku X - short because of the diagonal. The 7 in B2 can see two green cells - and importantly the blue 7 in G9 because of the diagonal. This is an excellent example of taking advantage of thee extra constraint.

It's A589 is neither a Naked Single of {1,3,7} because of the 5 in A5, nor a Hidden Single because of the extra 7 in A3. It can work out, however, if you strike lucky.

Complicated but perhaps doable. The user interface if already fairly complex but I'll give it a good think over

Not if you take into account the Killer Cage Convention, which I do. Have a read of this:
The_Killer_Cage_Convention

I think you're right, that one is too hard to be just diabolical. I believe the issues is that the spectrum is so wide at the tail end diabolical are in too great a range. I did some recalibrartion recently on the jigsaws and I wanted to extract the extremes by picking the hardest 10% or so, but that still leaves too many like these in the diabolical. What I need to do is tweak the heuristics to promote puzzles that rely on too many 'extreme' strategies. I'll give that a go in the next week or so

There is a problem with the solution count on the jigsaw solver. Its not working for me but is working offline. I'll have to figure that out first.

Bad timing. The stock ran out just as I went away for a week. However, I have uploaded a new stock with four grades throughout the week and a link to the KenKen solver.
Share and Enjoy