The New Sudoku Players' Forum

by **tarek** » Fri Jan 11, 2008 9:23 pm

I was wondering if this pattern has been posted somewhere before.....

it is a cross between windoku & 9-cell constraint (sort like Clover).....

To me it resembles Girandola fireworks. If anyone is interested, I could also post some puzzles following the rules of this sudoku variant

tarek

[Edit: Fixed broken link]

by **Ruud** » Fri Jan 11, 2008 9:56 pm

The 5-cell overlap between the R159C159 Windoku group and the additional constraint forces 2 quads for the same 4 digits in R5C1289 and R1289C5. These should be taken into account when rating the difficulty.

tarek wrote:If anyone is interested, I could also post some puzzles following the rules of this sudoku variant

Bring 'em on!

[edit]

Just figured out that the quads would also force R5C46 to contain the same digits as R37C5 and R5C37 the same as R46C5. This is a very restrained puzzle format.

Ruud

by **tarek** » Fri Jan 11, 2008 10:20 pm

There is no overlap that i can see Ruud....
The Windoku groups (in blue) and the extra constraint in red

The clover has some overlap but not this one........

Could you elaborate a bit !!

tarek

by **Ruud** » Fri Jan 11, 2008 10:46 pm

hidden constraints for the Windoku variant:

2 boxes in rows 234 leaves 9 cells containing digits 1-9 (R234C159)
2 boxes in rows 678 leaves 9 cells containing digits 1-9 (R678C159)
2 boxes in columns 234 leaves 9 cells containing digits 1-9 (R159C234)
2 boxes in columns 678 leaves 9 cells containing digits 1-9 (R159C678)

4 windoku boxes + 4 hidden constraints leaves 9 cells containing digits 1-9 in R159C159.

This last constraint has 5 overlapping cells with your red cells. The remaining 4 cells have a LOL relationship, but since they are distributed 2 by 2 in R5 AND C5, we have the quads.

Ruud

by **tarek** » Fri Jan 11, 2008 11:02 pm

thanx,

I'll try & generate some puzzles to see how this works out.

tarek

by **HATMAN** » Sat Jan 12, 2008 1:08 pm

Tarek

If you are interested in Restrained Sudokus, I recently posted a very limited one on Ruud's site at

http://www.sudocue.net/forum/viewtopic.php?t=887

Maurice

by **Smythe Dakota** » Sat Jan 12, 2008 1:17 pm

Clarification, please.

I'm not quite sure what all the nonets are. For example, is R1C2345678 R3C5 R4C5 one of them?

Bill Smythe

by **Ruud** » Sat Jan 12, 2008 2:59 pm

My first Girandola

Code: Select all: . . .|. . .|. . . . . 7|. . .|. . . . . .|5 . .|. . 3 -----+-----+----- . . .|. . .|. . . . . .|. . .|. . 7 . . .|. 5 8|. 1 . -----+-----+----- . . .|. . 4|. . . 3 1 .|8 . .|. 6 . . . .|6 9 1|. . 5

A light appetizer...

Ruud

by **HATMAN** » Sat Jan 12, 2008 10:47 pm

Bill

I do not think so - but it is an obviuos likelyhood.

Maurice

by **HATMAN** » Sat Jan 12, 2008 10:59 pm

Light as in "para"glider - all singles? (I mean the sport that I do occasionally - not him from {I've been chasing back on "para" - I thought he was from) the Netherlands but nothing is clearly stated,)

by **tarek** » Sat Jan 12, 2008 11:15 pm

[r5c8]: 9 is a hidden single for the Girandola group does it

Solution below:

193487652547263189268519743731946528685132497924758316856324971319875264472691835

will post some puzzle soon

tarek

by **tarek** » Sun Jan 13, 2008 4:04 am

thanx HATMAN for the link,

Here is my 1st Girandola ........ More to come in the future.

Code: Select all: 901000605000000200240001009002000000000000000000000100100500062003000000408000903 9 . 1 | . . . | 6 . 5 . . . | . . . | 2 . . 2 4 . | . . 1 | . . 9 -------+-------+------ . . 2 | . . . | . . . . . . | . . . | . . . . . . | . . . | 1 . . -------+-------+------ 1 . . | 5 . . | . 6 2 . . 3 | . . . | . . . 4 . 8 | . . . | 9 . 3

It should be very easy

tarek
[Edit: fixed broken link]

by **HATMAN** » Sun Jan 13, 2008 7:35 pm

tarek

With Restained Sudokus you have to limit the clues more severely
this one is human solvable with just the values in N1379

by **Ruud** » Sun Jan 13, 2008 7:48 pm

An optimized version of tarek's Girandola:

Code: Select all: . . .|. . .|. . 5 . . .|. . .|. . . . 4 .|. . .|. . 9 -----+-----+----- . . 2|. . .|. . . . . .|. . .|. . . . . .|. . .|1 . . -----+-----+----- 1 . .|. . .|. 6 2 . . 3|. . .|. . . 4 . 8|. . .|9 . 3

Although it has a unique solution, I wonder if this is still human solvable...

by **tarek** » Sun Jan 13, 2008 10:58 pm

Ruud,

you mentioned the LOL relationship between r28c5,r5c27 & r19c5,r5c19 which I understand....
But what is the quad you are talking about? (is it a LOLLOL

)

regarding Minimal puzzles & Beyond human level:

Code: Select all: . . 1 | . . . | . . . . 9 3 | . . . | . 4 . . . . | . . . | . 8 2 -------+-------+------ . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . . -------+-------+------ 7 1 . | . . . | . . . . 3 . | . . . | 5 1 . . . . | . . . | 2 . .

Another question, What do you call the regions .....
W1-4 is easy what about the other 5 hidden (do they have a name)?

tarek

[Edit: the above grid has multiple solutions produced due to a bug in my program which has been fixed]