The New Sudoku Players' Forum

by **wapati** » Thu Nov 30, 2006 12:52 am

This can be done. (It is very hard.)

Think fins, sushi.

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

by **re'born** » Thu Nov 30, 2006 4:38 pm

wapati wrote:This can be done. (It is very hard.)

Code: Select all: *--------------------------------------------------------------------* | 458 6 258 | 249 23489 1 | 2348 24 7 | | 9 7 28 | 246 2348 3468 | 12348 5 146 | | 148 12 3 | 5 7 468 | 2489 2469 469 | |----------------------+----------------------+----------------------| | 6 8 1 | 24 5 9 | 24 7 3 | | 35* 359* 4 | 1 23 7 | 6 29 8 | | 7 239- 29 | 8 6 34 | 5 1 49 | |----------------------+----------------------+----------------------| | 358 359* 5689 | 469 1 2 | 7 469 4569 | | 15* 4 569 | 7 89 568 | 19 3 2 | | 2 159* 7 | 3 49 456 | 149 8 14569 | *--------------------------------------------------------------------*

A little subset counting (ALS xz, if you like) on the *'d cells implies r6c2<>3, and it is all singles from here.

by **daj95376** » Thu Nov 30, 2006 6:59 pm

wapati wrote:This is about the ugliest pattern based puzzle I have seen. I haven't tried it yet, I print it out now.

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

Code: Select all: # after SSTS, which includes eliminations for [r3c6]<>2 & [r8c5]<>2 # # [r5c1]=1 => [r4c6],[r6c9]=1 => [r9c169]<>1 contradiction! => [r5c1]<>1 # [r5c1]=1=[r5c89]-1-[r6c9] => [r6c9]<>1 *--------------------------------------------------------------------* | 4567 9 56 | 57 2457 3 | 1 8 24 | | 3 478 2 | 1789 1479 148 | 47 5 6 | | 4578 478 1 | 6 2457 458 | 9 247 3 | |----------------------+----------------------+----------------------| | 1567 127 8 | 4 3 15 | 27 267 9 | | 45-1 3 9 | 2 6 7 | 48 14 1458 | | 14567 1247 56 | 15 8 9 | 3 267 45-1 | |----------------------+----------------------+----------------------| | 2 18 7 | 3 14 6 | 5 9 148 | | 9 6 3 | 158 145 12458 | 248 124 7 | | 18 5 4 | 79 79 128 | 6 3 128 | *--------------------------------------------------------------------* # - also - # finned Jellyfish [r2358C2678] in <4> # ???: [r5] Jellyfish cells locked in [b6] => [r6c9]<>4 *--------------------------------------------------------------------* | 4567 9 56 | 57 2457 3 | 1 8 24 | | 3 *478 2 | 1789 1479 *148 |*47 5 6 | | 4578 *478 1 | 6 2457 *458 | 9 *247 3 | |----------------------+----------------------+----------------------| | 1567 127 8 | 4 3 15 | 27 267 9 | | 15-4 3 9 | 2 6 7 |*48 *14 1458 | | 14567 #1247 56 | 15 8 9 | 3 267 15-4 | |----------------------+----------------------+----------------------| | 2 18 7 | 3 14 6 | 5 9 148 | | 9 6 3 | 158 145 *12458 |*248 *124 7 | | 18 5 4 | 79 79 128 | 6 3 128 | *--------------------------------------------------------------------*

Thus, [r5c1]=5 & [r6c9]=5 and puzzle solves with Singles.

by **wapati** » Thu Nov 30, 2006 11:49 pm

Bug sword xy and Turbot is what I find.

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

by **wapati** » Thu Nov 30, 2006 11:54 pm

Fins and Sashimi, keep that in mind.

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

by **wapati** » Thu Nov 30, 2006 11:58 pm

This has many patterns, x-wing, xy and sky are enough.

XYZs are here and Bug, if you care to find them.

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

by **wapati** » Fri Dec 01, 2006 12:11 am

sword, xyz and some more.

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

by **wapati** » Fri Dec 01, 2006 1:16 am

Testy, this one. Sword and a bunch more.

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

by **Carcul** » Fri Dec 01, 2006 3:12 pm

Wapati wrote:Bug sword xy and Turbot is what I find.

Code: Select all: *------------------------------------------------------* | 5 167 16 | 2 4 9 | 178 38 1378 | | 37 8 24 | 36 1 356 | 9 45 27 | | 139 19 24 | 7 8 35 | 6 45 12 | |----------------+------------------+------------------| | 4 5 7 | 1368 23 2368 | 18 68 9 | | 16 2 9 | 4 5 68 | 3 7 168 | | 8 3 16 | 16 9 7 | 5 2 4 | |----------------+------------------+------------------| | 127 17 3 | 5 6 4 | 278 9 78 | | 269 469 5 | 38 7 238 | 24 1 36 | | 267 467 8 | 9 23 1 | 247 36 5 | *------------------------------------------------------*

[r5c1]=1=[r5c9]-1-[r3c9]-2-[r2c9]-7-[r2c1]-3-[r2c4]-6-[r6c4]=6=[r6c3]=1=[r5c1], => r5c1=1 solving the puzzle.

Carcul

by **Carcul** » Fri Dec 01, 2006 3:29 pm

Wapati wrote:Fins and Sashimi, keep that in mind.

Code: Select all: *----------------------------------------------------------------* | 4 27 12578 | 158 6 1578 | 9 178 3 | | 18 9 6 | 3 17 2 | 5 4 178 | | 158 3 1578 | 9 4 1578 | 2 6 178 | |---------------------+--------------------+---------------------| | 2 1 3 | 4 5 789 | 6 78 789 | | 9 47 4578 | 6 17 3 | 18 1578 2 | | 568 67 578 | 18 2 1789 | 4 3 15789 | |---------------------+--------------------+---------------------| | 3 5 14 | 7 9 6 | 18 2 148 | | 16 8 9 | 2 3 4 | 7 15 156 | | 7 246 24 | 15 8 15 | 3 9 46 | *----------------------------------------------------------------*

[r8c8]-1-[r1c8|r8c1]=1=[X-Wing:r23c19]-1-[r2c5|r3c6]=1=[X-
-Wing:r19c46]-1-[r6c46]=1=[r5c5]-1-[r5c7]=1=[r7c7]-1-[r8c8], => r8c8<>1 solving the puzzle.

Carcul

by **Carcul** » Fri Dec 01, 2006 3:53 pm

Wapati wrote:This has many patterns, x-wing, xy and sky are enough.
XYZs are here and Bug, if you care to find them.

Code: Select all: *--------------------------------------------------------* | 69 1 8 | 4 3679 2 | 39 379 5 | | 5 2 367 | 67 36789 38 | 1 379 4 | | 4 39 37 | 1 379 5 | 6 2 8 | |-------------------+------------------+-----------------| | 1 8 2 | 3 4 6 | 7 5 9 | | 67 67 4 | 2 5 9 | 8 1 3 | | 3 5 9 | 8 1 7 | 2 4 6 | |-------------------+------------------+-----------------| | 678 4 1 | 9 3678 38 | 5 36 2 | | 2679 3679 5 | 67 2367 4 | 39 8 1 | | 28 369 36 | 5 28 1 | 4 369 7 | *--------------------------------------------------------*

[r2c8]=9=[r2c5]-9-[r3c5]=9=[r3c2]-9-[r1c1]-6-[r5c1]=6=[r5c2]=7=
=[r8c2]-7-[r8c4]=7=[r2c4](-7-[r2c8])-7-[r3c5]-3-[r1c5]=3=[r1c78]-3-
-[r2c8], => r2c8=9 solving the puzzle.

Carcul

by **Carcul** » Fri Dec 01, 2006 6:36 pm

Wapati wrote:Testy, this one. Sword and a bunch more.

Code: Select all: *-----------------------------------------------------------* | 23 4 5 | 7 8 9 | 23 6 1 | | 6 1789 1789 | 3 2 5 | 789 89 4 | | 2379 3789 2789 | 6 4 1 | 23789 5 2389 | |--------------------+-----------------+--------------------| | 479 79 6 | 8 1 3 | 5 2479 279 | | 1 2 89 | 5 7 4 | 89 3 6 | | 457 578 3 | 2 9 6 | 1 478 78 | |--------------------+-----------------+--------------------| | 579 579 4 | 1 3 2 | 6 789 789 | | 8 139 129 | 4 6 7 | 239 29 5 | | 237 6 27 | 9 5 8 | 4 1 237 | *-----------------------------------------------------------*

[r8c7]=2|8=[r3c2](=3=[r8c2]-3-[r8c7])-8-[r6c2]=8=[r5c3]=9=[r5c7]-9-
-[r8c7], => r8c7<>3,9 solving the puzzle.

Carcul

by **ArkieTech** » Fri Dec 01, 2006 8:01 pm

Carcul,
I am trying to learn some advanced solving techniques. I have been studying your recent response to Wapati:

Code: Select all: *--------------------------------------------------------* | 69 1 8 | 4 3679 2 | 39 379 5 | | 5 2 367 | 67 36789 38 | 1 379 4 | | 4 39 37 | 1 379 5 | 6 2 8 | |-------------------+------------------+-----------------| | 1 8 2 | 3 4 6 | 7 5 9 | | 67 67 4 | 2 5 9 | 8 1 3 | | 3 5 9 | 8 1 7 | 2 4 6 | |-------------------+------------------+-----------------| | 678 4 1 | 9 3678 38 | 5 36 2 | | 2679 3679 5 | 67 2367 4 | 39 8 1 | | 28 369 36 | 5 28 1 | 4 369 7 | *--------------------------------------------------------*

[r2c8]=9=[r2c5]-9-[r3c5]=9=[r3c2]-9-[r1c1]-6-[r5c1]=6=[r5c2]=7=
=[r8c2]-7-[r8c4]=7=[r2c4](-7-[r2c8])-7-[r3c5]-3-[r1c5]=3=[r1c78]-3-
-[r2c8], => r2c8=9 solving the puzzle.

First question. When I load the puzzle I get:

Code: Select all: *-----------* |.18|4.2|..5| |52.|...|1.4| |4..|1.5|628| |---+---+---| |182|346|759| |..4|259|813| |359|817|246| |---+---+---| |.41|9..|5.2| |..5|..4|.81| |...|5.1|4.7| *-----------*

Which has these candidates followed by yours:

Code: Select all: *--------------------------------------------------------------------* | 679 1 8 | 4 3679 2 | 39 379 5 | | 5 2 367 | 67 36789 38 | 1 379 4 | | 4 379 37 | 1 379 5 | 6 2 8 | |----------------------+----------------------+----------------------| | 1 8 2 | 3 4 6 | 7 5 9 | | 67 67 4 | 2 5 9 | 8 1 3 | | 3 5 9 | 8 1 7 | 2 4 6 | |----------------------+----------------------+----------------------| | 678 4 1 | 9 3678 38 | 5 36 2 | | 2679 3679 5 | 67 2367 4 | 39 8 1 | | 2689 369 36 | 5 2368 1 | 4 369 7 | *--------------------------------------------------------------------* *--------------------------------------------------------* | 69 1 8 | 4 3679 2 | 39 379 5 | | 5 2 367 | 67 36789 38 | 1 379 4 | | 4 39 37 | 1 379 5 | 6 2 8 | |-------------------+------------------+-----------------| | 1 8 2 | 3 4 6 | 7 5 9 | | 67 67 4 | 2 5 9 | 8 1 3 | | 3 5 9 | 8 1 7 | 2 4 6 | |-------------------+------------------+-----------------| | 678 4 1 | 9 3678 38 | 5 36 2 | | 2679 3679 5 | 67 2367 4 | 39 8 1 | | 28 369 36 | 5 28 1 | 4 369 7 | *--------------------------------------------------------*

which is different than the candidates you start with (example the 7 in [r1c1].
What am I missing? Are there candidate elimination steps prior to your solution?

by **Carcul** » Fri Dec 01, 2006 8:36 pm

ArkieTech wrote:What am I missing?

Locked candidates (in column 3, "7" occurs only in box 1) and naked triple in row 9.

Carcul

by **ArkieTech** » Fri Dec 01, 2006 8:53 pm

Carcul wrote:

Locked candidates (in column 3, "7" occurs only in box 1) and naked triple in row 9.

Thanks,

The New Sudoku Players' Forum

CHAINS, not required. Good puzzles.