Solver App for Android
Strategies for Popular Number Puzzles
Page:2024 2023 2022 2021 2020 2019 2018 2017
2015 2014 2013 2012 :1 2 3 4 5 6 7 8 9 :2011 2010 2009 2008 2007 2006

Feedback and Questions

I've received a lot of interesting comments and questions from Sudoku fans over the last few years and this page is where I try to answer them. I'm also directing Str8ts feedback here. Please feel free to drop me a note on the side of the page. Or you can email me directly at .

Post a Comment or Question here...
Your Name or 'Handle'

Part of the World you're from

Email Address (optional) so I can reply directly if necessary (it will not be displayed here)

Any Sudoku you want to publish here for easy loading into the solver - 81 characters, use '.' or '0' for unknowns.

public - comment/question will be added to feedback column
private - email comment/question directly to Andrew Stuart, don't display here

Comment, question or feedback:

Enter these letters
Please enter the
letters you see:

Many thanks to all the people who have helped improve the solvers and strategies with their feedback!

Thursday 22-Mar-2012

... by: Andrew Stuart, Norfolk, UK

I have to apologise to everyone who has sent feedback in the last two months and not had a reply. I've just discovered where all the mail from the server is sent to and the sending out part was not configured properly. Recent MS Office upgrade complicated things. Real pity as I replied to tons of people. I have re-sent the replies from this month. Also I can now see the comments on the doc pages when they come in so I can respond more quickly. so frustrating....

Sunday 18-Mar-2012

... by: Anna, Germany

Thank you!!!

Sunday 18-Mar-2012

... by: Tom Veil, US

First rate site.. can't imagine the amount of work that goes into it. Makes my head hurt ;)

Any plans to extend the KenKen/KenDoKu solvers out to 9x9??


Andrew Stuart writes:
Thanks for the feedback. I'm still not completely happy with the KenKen solver. I can solve all bar 40% of the diabolicals from KenKen itself (obviously 100% of my own but that doesn't count). It is partly related to Killers where the solver still struggles on a few very hard ones and I get often get sent examples. So when that improves I'll release a 9x9 version.

Saturday 17-Mar-2012

... by: Tim Reynolds, Anderson IN

Load Sudoku:

I know you will not reply. This puzzle shows a bug in the grader. It has three solutions but still grades diabolical. It needs a one at C4 to become unique.
Also the grader drops the grade when 7 is removed from A4.

Andrew Stuart writes:
I don't think it's a bug as such but perhaps it's not informative enough. The solution count is quite separate from the solver and the solver doesn't check the number of solutions before trying to step through. With multiple solutions it usually it 'runs out of known strategies' because there is not completely logical solution. In this case, which is unusual, it takes a logical path and runs it to the end, despite there being multiple solutions.

But your example does inspire me to change the grader to check for solutions since the grade should be meaningless with more than one solution.

On the grade drop - I have found that the number of clues has no relevance to the grade apart from making the puzzle longer - which can accumulate scoring points and boost the grade slightly. It is easy to make 17 clue puzzles that are trivial and 32 clue puzzles that are diabolical, so it it not indicative of anything that adding or removing a clue changes the grade.

Friday 16-Mar-2012

... by: Roggey, Uk

Dear Andrew.

Trying to load a sudoku into your grid on an iPad is a nightmare. You have to tap each square separately then select numbers. Then repeat the whole process for the next number. If you get out of sync and tap the wrong thing the whole entry gets lost and you have to start from scratch.

Is there anything you can do to improve this.

Best regards


Andrew Stuart writes:
I quite agree, it's totally not designed for the iPad. I have built a Sudoku player for iPad which is available here. For the solver we are nearly finished the iPhone version and intend to release an iPad version soon after. More info here.

Thursday 15-Mar-2012

... by: JG, CA

The solver is a wonderful learning tool. Is there a way to have it show multiple possible eliminations within a single strategy/step?

For example, I can turn everything off on the solver except 'X-cycles' and if an X-cycle exists in the puzzle, the solver will only show the first one it finds. It would be very useful for learning to also have the solver optionally show the 2nd, 3rd, etc other X-cycles in the puzzle for a given step if more such possibilities exist.

The strategy overviews are excellent. Great site. JG

Andrew Stuart writes:
This is something I'd love to do. Especially with chaining strategies there are usually lots of alternatives. I have had to design a special rule set to pick out the 'best' chain based on a lot of factors - length, type of link components, how many eliminations it makes, etc. It would be more elegant to allow the user to flick through them and choose. The main problem is User Interface and how to present this information/feature without over complicating the page. Definitely on my wish list

Wednesday 14-Mar-2012

... by: Hondo500, California

i received the following message on your solver:

Run out of known strategies. Use 'Solution Count' to check the puzzle has only one solution.

Should it have said "to check to see if the puzzle has only one solution" ?

Or was it saying that the puzzle has only one solution?

When i did check the solution count, it indicated 27 solutions. So what does that mean, that there are 27 solutions? i thought Sodoku puzzles only had one solution???

Andrew Stuart writes:
Thanks for the grammar check; that change will go in the next release, appreciated.

On multiple solutions - The solver uses logical strategies to take one through the solution path, so a single solution is required. Anything with multiple solutions is not a proper puzzle and will not solve logically - since it is ambiguous what the solution will be. The Solution Count uses a rapid brute force method to calculate the solutions - a very different approach - and can provide all those solutions, however many there are. Some publishers don't know how to check for multiple solutions so they are publishing garbage, to be frank, because a user expects a single solutions. But if you miss off a clue, or put the wrong number in, zero or many solutions are also possible and a logical outcome is not possible.

Saturday 10-Mar-2012

... by: kit_kat_130, usa

you may already be aware of this, but there appears to be something wrong with the jigsaw solver. somewhere in the middle of your jigsaw puzzle 180, i get this message: paint_yes_no: no row called RFXW.

i was very sorry to hear about the hacking of your site, which remains absolutely the best. and thanks so much for adding daily jigsaw puzzles as they are my top favorite.

thanks again. kit

Andrew Stuart writes:
Yes the Jigsaw solver thought finned x-wings were a valid strategy. Pure coincidence if it worked, but I've taken it out now and the puzzle solves in the way it should (AICs in this case).
Thanks for the alert, I hadn't caught the bug. Glad you are enjoying the puzzles!

Thursday 8-Mar-2012

... by: Nick Winter, UK/France

Re Killer Jigsaw monthly 50. Delighted that you're making the early ones harder. I'm finding it much more difficult to do them in pen now! (that is without penciling in the alternatives first)

Tuesday 6-Mar-2012

... by: Joseph, US

I am having a problem understanding 3D Medusa Rule 6. I understand the rule; however, I don't see what makes the 1 in cell E6 the color green? Shouldn't there be a blue 1 in the same unit (row, column, or box) that therefore would then make the 1 in cell E6 green? If that is the case, then I think I understand why the 1 in E6 and the 7 in E5 can be removed, but in order to do this, there must be a blue 1 that can see the green 1. Perhaps I am missing something or just not understanding this?

Andrew Stuart writes:
The 1 in E6 is green because the 6 in E6 is blue. E6 is a bi-value cell which means it can be a link in a chain. The chain (or net) rather got to the 6 in E6 first and turned it blue. Then it continued to the 1 turning it green. You are not taking into account bi-value cells basically, you are just looking in units (which is valid for Singles Chain, the 2D version of Medusa)
Page:2024 2023 2022 2021 2020 2019 2018 2017
2015 2014 2013 2012 :1 2 3 4 5 6 7 8 9 :2011 2010 2009 2008 2007 2006
Thank-you everyone for all your questions and contributions.