The New Sudoku Players' Forum

by **Rovers_5** » Wed Feb 01, 2006 10:48 am

Code: Select all: . . . . . . . 1 6 7 . . 1 8 6 . . 5 . . . 5 4 . . 9 2 . 7 . 8 . 3 . . . 5 . . . . . . . 8 . . . 2 . 1 . 6 . 2 1 . . 6 8 . . . 8 . . 7 1 5 . . 3 3 5 . . . . . . .

And I now have it here:

Code: Select all: 4 8 5 9 3 2 7 1 6 7 . . 1 8 6 . . 5 . . . 5 4 7 8 9 2 . 7 . 8 . 3 . . . 5 . . 6 . 4 . . 8 9 . 8 2 . 1 . 6 . 2 1 . 3 6 8 . . . 8 . . 7 1 5 . . 3 3 5 . 4 2 9 . 8 .

by **Crazy Girl** » Wed Feb 01, 2006 11:59 am

What's the source of the puzzle :?:

The puzzle can't be solved using basic/intermediate techniques, so it is likely to need a Chain/an advanced technique to solve.

by **CathyW** » Wed Feb 01, 2006 12:16 pm

Hello and welcome to the Sudoku Forum!

It helps to have candidate lists as well so we know what you've eliminated. Will assume it's as follows:

Code: Select all: {4} {8} {5} {9} {3} {2} {7} {1} {6} {7} {29} {29} {1} {8} {6} {34} {34} {5} {16} {36} {136} {5} {4} {7} {8} {9} {2} {16*} {7} {1246#} {8} {59} {3} {12459} {245} {149} {5} {23} {123} {6} {79} {4} {1239} {237} {8} {9} {34} {8} {2} {57} {1} {345} {6} {47} {2} {1} {479} {3} {6} {8} {459} {457} {479} {8} {469} {469} {7} {1} {5} {2469} {24} {3} {3} {5} {67*} {4} {2} {9} {16} {8} {17*}

Sorry - rest of post deleted as the cells I'd marked are NOT an xy-wing :!:

by **Crazy Girl** » Wed Feb 01, 2006 12:38 pm

It helps to have candidate lists as well so we know what you've eliminated. Will assume it's as follows:

Code: Select all: -------------------+-----------------+-----------------| 4 8 5 | 9 3 2 | 7 1 6 | 7 29 29 | 1 8 6 | 34 34 5 | 16 36 136 | 5 4 7 | 8 9 2 | -------------------+-----------------+-----------------| 16* 7 1246# | 8 59 3 | 12459 245 149 | 5 23 123 | 6 79 4 | 1239 237 8 | 9 34 8 | 2 57 1 | 345 6 47 | -------------------+-----------------+-----------------| 2 1 479 | 3 6 8 | 459 457 479 | 8 469 469 | 7 1 5 | 2469 24 3 | 3 5 67* | 4 2 9 | 16 8 17* | -------------------+-----------------+-----------------|

Cathy,
this is not an xy-wing as the 67 in R9C3 is not connected to R4C1.
the two cells are not in the same block / row / column.

So we are still stuck at the above grid. :!:

by **MCC** » Wed Feb 01, 2006 1:04 pm

I'm wondering if the Filet-O-Fish method couldn't be used here.

Code: Select all: -------------------+-----------------+-----------------| 4 8 5 | 9 3 2 | 7 1 6 | 7 29 29 | 1 8 6 | 34 34 5 | 16 36* 136* | 5 4 7 | 8 9 2 | -------------------+-----------------+-----------------| 16 7 1246# | 8 59 3 | 12459 245 149 | 5 23 123 | 6 79 4 | 1239 237 8 | 9 34 8 | 2 57 1 | 345 6 47 | -------------------+-----------------+-----------------| 2 1 479 | 3 6 8 | 459 457 479 | 8 46*9 46*9 | 7 1 5 | 2469 24 3 | 3 5 67^ | 4 2 9 | 16 8 17 | -------------------+-----------------+-----------------|

Using the x-wing of 6's * and the fin of 6^ to remove the 6#

The 6's are either in r3c3 and r8c2 or
in r3c2 and (r89c3)
So 6 occupies r3c3 or r89c3 therefore the 6 in r4c3 can be eliminated.

MCC

by **Wolfgang** » Wed Feb 01, 2006 1:04 pm

This should solve it.
If r5c8=7:
1. r6c9=4, r6c2<>4
2. r5c5=9, r4c5=5, r4c8=2, r8c8=4, r8c2<>4
Then there is no 4 left for column 2 => r7c8=7

by **tarek** » Wed Feb 01, 2006 1:40 pm

This is how my solver solves it. considerably longer than wolfgang's.

Code: Select all: *-----------------------------------------------------------------* | 4 8 5 | 9 3 2 | 7 1 6 | | 7 29 29 | 1 8 6 | 34 34 5 | | 16 36 136 | 5 4 7 | 8 9 2 | |---------------------+---------------------+---------------------| | 16 7 1246 | 8 59 3 | 12459 245 149 | | 5 23 123 | 6 79 4 | 1239 237 8 | | 9 34 8 | 2 57 1 | 345 6 47 | |---------------------+---------------------+---------------------| | 2 1 479 | 3 6 8 | 459 457 479 | | 8 469 469 | 7 1 5 | 2469 24 3 | | 3 5 67 | 4 2 9 | 16 8 17 | *-----------------------------------------------------------------* Candidates in r4c9 will force r6c7 to have only 35 as valid Candidates r4c9=1: r4c9=1 => r9c9=7 => r6c9=4 => r6c7<>4 => r6c7=35 r4c9=4: r4c9=4 => r6c7<>4 => r6c7=35 r4c9=9: r4c9=9 => r4c5=5 => r6c5=7 => r6c9=4 => r6c7<>4 => r6c7=35 Threfore r6c7=35 *-----------------------------------------------------------------* | 4 8 5 | 9 3 2 | 7 1 6 | | 7 29 29 | 1 8 6 | 34 34 5 | | 16 36 136 | 5 4 7 | 8 9 2 | |---------------------+---------------------+---------------------| | 16 7 1246 | 8 59 3 | 12459 245 149 | | 5 23 123 | 6 79 4 | 1239 237 8 | | 9 34 8 | 2 57 1 | 35 6 47 | |---------------------+---------------------+---------------------| | 2 1 479 | 3 6 8 | 459 457 479 | | 8 469 469 | 7 1 5 | 2469 24 3 | | 3 5 67 | 4 2 9 | 16 8 17 | *-----------------------------------------------------------------* Candidates in r4c8 will force r2c7 to have only 4 as valid Candidates r4c8=2: r4c8=2 => r8c8=4 => r2c8=3 => r2c7=4 r4c8=4: r4c8=4 => r2c8=3 => r2c7=4 r4c8=5: r4c8=5 => r6c7=3 => r2c7=4 Threfore r2c7=4 *--------------------------------------------------------* | 4 8 5 | 9 3 2 | 7 1 6 | | 7 29 29 | 1 8 6 | 4 3 5 | | 16 36 136 | 5 4 7 | 8 9 2 | |------------------+------------------+------------------| | 16 7 1246 | 8 59 3 | 1259 245 149 | | 5 23 123 | 6 79 4 | 1239 27 8 | | 9 34 8 | 2 57 1 | 35 6 47 | |------------------+------------------+------------------| | 2 1 479 | 3 6 8 | 59 457 479 | | 8 469 469 | 7 1 5 | 269 24 3 | | 3 5 67 | 4 2 9 | 16 8 17 | *--------------------------------------------------------* Eliminating 4 From r4c8 (2 & 7 in r5c8 form an XY wing with 4 in r8c8 & r6c9) Eliminating 4 From r7c9 (2 & 7 in r5c8 form an XY wing with 4 in r8c8 & r6c9) *--------------------------------------------------------* | 4 8 5 | 9 3 2 | 7 1 6 | | 7 29 29 | 1 8 6 | 4 3 5 | | 16 36 136 | 5 4 7 | 8 9 2 | |------------------+------------------+------------------| | 16 7 1246 | 8 59 3 | 1259 25 149 | | 5 23 123 | 6 79 4 | 1239 27 8 | | 9 34 8 | 2 57 1 | 35 6 47 | |------------------+------------------+------------------| | 2 1 479 | 3 6 8 | 59 457 79 | | 8 469 469 | 7 1 5 | 269 24 3 | | 3 5 67 | 4 2 9 | 16 8 17 | *--------------------------------------------------------* Candidates in r6c7 will force r9c3 to have only 7 as valid Candidates r6c7=3: r6c7=3 => r6c2=4 => r6c9=7 => r9c9=1 => r9c7=6 => r9c3=7 r6c7=5: r6c7=5 => r7c7=9 => r7c9=7 => r9c9=1 => r9c7=6 => r9c3=7 Threfore r9c3=7 *--------------------------------------------------------* | 4 8 5 | 9 3 2 | 7 1 6 | | 7 29 29 | 1 8 6 | 4 3 5 | | 16 36 136 | 5 4 7 | 8 9 2 | |------------------+------------------+------------------| | 16 7 1246 | 8 59 3 | 1259 25 49 | | 5 23 123 | 6 79 4 | 1239 27 8 | | 9 34 8 | 2 57 1 | 35 6 47 | |------------------+------------------+------------------| | 2 1 49 | 3 6 8 | 59 457 79 | | 8 469 469 | 7 1 5 | 29 24 3 | | 3 5 7 | 4 2 9 | 6 8 1 | *--------------------------------------------------------* Candidates in r6c7 will force r7c3 to have only 4 as valid Candidates r6c7=3: r6c7=3 => r6c2=4 => r6c9=7 => r7c9=9 => r7c3=4 r6c7=5: r6c7=5 => r7c7=9 => r7c3=4 Threfore r7c3=4

by **CathyW** » Wed Feb 01, 2006 4:44 pm

Crazy Girl wrote:Cathy,
this is not an xy-wing as the 67 in R9C3 is not connected to R4C1.
the two cells are not in the same block / row / column.

So we are still stuck at the above grid.

Oh dear! Another "happy coincidence"? I can see I shall have to re-read the xy-wing descriptions at Angus J's and Simes' sites.

Will try again later with colouring and see where I get to.

by **tso** » Wed Feb 01, 2006 5:59 pm

[duplicate post deleted]

by **tso** » Wed Feb 01, 2006 6:26 pm

MCC wrote:I'm wondering if the Filet-O-Fish method couldn't be used here.

Code: Select all
-------------------+-----------------+-----------------| 4 8 5 | 9 3 2 | 7 1 6 | 7 29 29 | 1 8 6 | 34 34 5 | 16 36* 136* | 5 4 7 | 8 9 2 | -------------------+-----------------+-----------------| 16 7 1246# | 8 59 3 | 12459 245 149 | 5 23 123 | 6 79 4 | 1239 237 8 | 9 34 8 | 2 57 1 | 345 6 47 | -------------------+-----------------+-----------------| 2 1 479 | 3 6 8 | 459 457 479 | 8 46*9 46*9 | 7 1 5 | 2469 24 3 | 3 5 67^ | 4 2 9 | 16 8 17 | -------------------+-----------------+-----------------|

Using the x-wing of 6's * and the fin of 6^ to remove the 6#

The 6's are either in r3c3 and r8c2 or
in r3c2 and (r89c3)
So 6 occupies r3c3 or r89c3 therefore the 6 in r4c3 can be eliminated.

MCC

This doesn't work -- there are two indistinguishable fins.
There is no logical way to decide whether to exclude r4c3 or r9c3.

Why...
"Using the x-wing of 6's * and the fin of 6^ to remove the 6#"

...instead of...
"Using the x-wing of 6's * and the fin of 6# to remove the 6^"

The statement:

"The 6's are either in r3c3 and r8c2 or in r3c2 and (r89c3)"

... is unsupported.

Here's the situation with everything filtered-out but the 6s: (Bracketed 6s are already solved.)

Code: Select all: . . . | . . . | . . [6] . . . | . . [6] | . . . 6 6 6 | . . . | . . . -------------------+-----------------+----------------- 6 . 6 | . . . | . . . . . . |[6] . . | . . . . . . | . . . | . [6] . -------------------+-----------------+----------------- . . . | . [6] . | . . . . 6 6 | . . . | 6 . . . . 6 | . . . | 6 . .

There's nothing that the 6s can tell us that rules out this:

Code: Select all: . . . | . . . | . . [6] . . . | . . [6] | . . . 6 . . | . . . | . . . -------------------+-----------------+----------------- . . 6 | . . . | . . . . . . |[6] . . | . . . . . . | . . . | . [6] . -------------------+-----------------+----------------- . . . | . [6] . | . . . . 6 . | . . . | . . . . . . | . . . | 6 . .

... in which neither r3c2 nor r3c3 is 6. And ... r4c3 IS 6.

by **CathyW** » Wed Feb 01, 2006 10:33 pm

I give up on this one!

According to Sudoku Susser, it requires Trebor's Tables:

"* Made progress using Trebor's Tables to find inferences about the puzzle. A total of 9574 implications about the puzzle were generated and examined in order to find these inferences - you'd run through several pencils working them out by hand!"

by **tso** » Wed Feb 01, 2006 10:50 pm

CathyW wrote:I give up on this one!

According to Sudoku Susser, it requires Trebor's Tables:

"* Made progress using Trebor's Tables to find inferences about the puzzle. A total of 9574 implications about the puzzle were generated and examined in order to find these inferences - you'd run through several pencils working them out by hand!"

You have an out-of-date version of Susser. The current version is 2.5.0 -- and it solves this puzzles using "only" comprehensive forcing chains.

by **CathyW** » Wed Feb 01, 2006 10:51 pm

Thanks Tso - I'll go get the most recent one now!

by **Wolfgang** » Thu Feb 02, 2006 10:57 am

tso wrote:You have an out-of-date version of Susser. The current version is 2.5.0 --

Its a pity, but the new susser version does not run under debian linux like the old one (my library versions are ok). Does someone know, it it runs with other linux distributions?

by **MCC** » Thu Feb 02, 2006 12:33 pm

Thanks Tso
I realised after posting that I might have made a mistake and have been waiting for someone to comment on it.

I saw something that wasn't there.
It wasn't there again today.
Slippery things, fish.

MCC

The New Sudoku Players' Forum

I could use some help=)

I could use some help=)