The New Sudoku Players' Forum

[herschko]

There is a new (at least to me) sudoku variant in the current issue of MAA Focus. (Mathematical Association of America), Called a meandoku, the only information is the about the average of the entries in pairs adjacent cells within nonets. One is told whether this mean is greater than, less than or equal to five. The author is David Nacin of William Paterson University, New Jersey, USA.

I am addicted to greater-than-killers. However, having been doing the for years, I know how to approach those. It is always nice to have develop a new strategy.

Sorry I don't know how to download the image, or even if this is legal. I did copy it to an Excel file. (I couldn't do it unless I could save my work and go back after mistakes.) Is it kosher to post it?

[Pat]

MAA Focus -- 2017 April/May
Meandoku puzzle is on page 20

[herschko]

Can people who are not members of the MAA access this?

[HATMAN]

MeanDoku NC 2

A very pleasant puzzle - but easy to make mistakes, it took me three attempts.

To me It naturally links to Non-Consecutive so here is one:

I found the trapezoids difficult to see clearly so I have gone for colours:
Green: the average of the two adjacent cells is below five
Blue: the average of the two adjacent cells is five
Red: the average of the two adjacent cells is above five

NC horizontally and vertically adjacent cells are non-consecutive.

[HATMAN]

I've posted a harder version of this puzzle on the assassin site with the solution:

www.rcbroughton.co.uk/sudoku/forum/viewtopic.php?p=7383#p7383

Hidden Text: Show

Note: it is easier to think of the clues as sum = 9-, sum = 10 and sum = 11+ hence to use JSudoku to solve it

[HATMAN]

MeanDoku NC X 11

I found the trapezoids difficult to see clearly so I have gone for colours:
Green: the average of the two adjacent cells is below five
Blue: the average of the two adjacent cells is five
Red: the average of the two adjacent cells is above five

NC horizontally and vertically adjacent cells are non-consecutive.
X no repeats on the two main diagonals.

I created this one based on a pretty solution and was surprised to find that there were two solutions.
Note I've posted a few on the assassin site.

[Quadrata]

This is David Nacin, author of the Meandoku discussed here in this thread.  I was really excited when I found this thread.

First of all I really wanted to thank Hatman for mentioning that it was tricky distinguishing the trapezoids in the version I made for the MAA Focus.  It totally hadn't occurred to me to color in the Trapezoids with different colors.  Now I do it every time!  That suggestion made a big difference in all the Meandoku puzzles I've made since.

Really excited to see Hatman making some of these!  I wanted to toss out a few more places where you can find Meandoku in case anyone wanted to try some more.  
There's another Meandoku at my blog:

https://quadratablog.blogspot.com/2018/08/meandoku.html

And a Meandoku was posted on Twitter (skip the link to the article and look below):

https://twitter.com/wpunj_edu/status/11 ... 3987743752

Always excited to see people doing my Sudoku variants and also looking forward to checking out some of the other puzzles here, including Hatman's!

[Quadrata]

This is David Nacin, the author of the Meandoku discussed here in this thread.  Was excited when I found this post.

First of all I really wanted to thank Hatman for mentioning that it was tricky distinguishing the trapezoids in the version I made for the MAA Focus. It totally hadn't occurred to me to color in the trapezoids with different colors.  Now I do it every time!  That suggestion made a big difference.

Really excited to see Hatman making some of these!  I wanted to toss out a few more places where you can find Meandoku in case anyone wanted to try some more.  

There's another Meandoku at my blog: https://quadratablog.blogspot.com/2018/08/meandoku.html

And a easier Meandoku was posted as part of a (now over) contest on Twitter which might be good for a first Meandoku: https://twitter.com/wpunj_edu/status/1191764565836603393 (Ignore the media article, puzzle is below.)

[Wecoc]

I find this variant quite interesting. Any reason for not including the cells between different boxes?
I tried to make one myself, with unique solution
It's not minimal (the entire grid is filled) but it's still difficult enough in my opinion.

meandoku: Show

I guess the existence of a mean-doku implies there's also a kind-doku, somewhere to be find...

[HATMAN]

No real reason but the interactions in a nonet help in the solution.
Alternatively you can put them in a row or column as I do with my Mean ORCs, for my Christmas last post see the assassin site: http://www.rcbroughton.co.uk/sudoku/for ... =13&t=1424

I like the kind-doku or KinDoku idea. I'll see what I can come up with.