The New Sudoku Players' Forum

by **wapati** » Wed Nov 15, 2006 8:33 am

Ditto, no fancy tricks, enough steps.

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

by **Carcul** » Wed Nov 15, 2006 10:30 am

Wapati wrote:This is not easy. Perhaps by chains, not easy by patterns.

Code: Select all: *---------------------------------------------------* | 4 17 6 | 178 178 2 | 5 3 9 | | 17 9 3 | 6 4 5 | 17 8 2 | | 8 2 5 | 17 3 9 | 6 17 4 | |----------------+------------------+---------------| | 1257 157 9 | 1278 6 1478 | 1247 17 3 | | 127 4 8 | 5 17 3 | 127 9 6 | | 3 6 12 | 127 9 147 | 1247 5 8 | |----------------+------------------+---------------| | 25 58 7 | 9 28 6 | 3 4 1 | | 6 18 4 | 3 5 18 | 9 2 7 | | 9 3 12 | 4 127 17 | 8 6 5 | *---------------------------------------------------*

r5c1=2 or r1c5=8, and so r7c1<>2 (because of r7c5) solving the puzzle.

Carcul

by **Carcul** » Wed Nov 15, 2006 10:55 am

Wapati wrote:I like this one. Not very technical, hard enough!

Code: Select all: *------------------------------------------------------* | 45 6 459 | 2 89 3 | 1 7 58 | | 8 1 37 | 67 4 5 | 36 9 2 | | 37 29 259 | 679 789 1 | 3456 46 358 | |------------------+-----------------+-----------------| | 9 24 24 | 3 1 6 | 8 5 7 | | 37 5 6 | 8 279 4 | 239 1 39 | | 1 8 37 | 79 5 29 | 2349 24 6 | |------------------+-----------------+-----------------| | 46 49 1489 | 5 3 289 | 7 26 19 | | 2 7 589 | 1 6 89 | 59 3 4 | | 56 3 159 | 4 29 7 | 26 8 159 | *------------------------------------------------------*

1. [r7c2]-4-[r7c18](-2-[r7c6])-6-[r9c1]-5-[r9c39]=5|4=[r7c3]-4-[r7c2], => r7c2<>4 (where the "=5|4=" link is due to a TIUR argument).

2. Now we have a BUG+3 where r3c7=6 or r5c5=9 or r6c7=9, and so r5c9<>9 solving the puzzle.

Carcul

by **Carcul** » Wed Nov 15, 2006 11:16 am

Wapati wrote:Ditto, no fancy tricks, enough steps.

Code: Select all: *-----------------------------------------------------* | 1359 6 139 | 7 8 135 | 2 359 4 | | 7 358 4 | 35 2 9 | 1 6 38 | | 1359 1358 2 | 6 13 4 | 359 7 389 | |-----------------+-----------------+-----------------| | 2 7 5 | 4 9 13 | 6 8 13 | | 13 9 13 | 8 6 7 | 4 2 5 | | 6 4 8 | 13 5 2 | 7 39 139 | |-----------------+-----------------+-----------------| | 4 2 139 | 1359 13 6 | 8 1359 7 | | 1359 135 7 | 2 4 8 | 359 1359 6 | | 8 135 6 | 1359 7 135 | 359 4 2 | *-----------------------------------------------------*

1. [r8c1]=9=[r7c3]-9-[r7c4]=9=[r9c4]=5=[r1c1|r8c8]-5-[r8c1], => r8c1<>5.

2. [r9c4]=9|8=[r3c2]=1=[r9c2]-1-[r9c4], => r9c4<>1 solving the puzzle.

Carcul

by **daj95376** » Wed Nov 15, 2006 6:56 pm

ronk wrote:
daj95376 wrote:
Code: Select all
# sdIC on [r15c7]=2 => [r2c2]=[r7c9]=[r8c1]=[r9c4]=2 and ... # 13 eliminations on 2 # 3 " " 1 # 2 " " 4 # 4 " " 6 # 2 " " 7 # 1 " " 8 # 25 " total *-------------------------------------------------------------* | 9 4 6-2 | 567 1 5678 | 268 2568 3 | | 3 2-168 5 | 9 4 68 | 7 168-2 16-2 | | 16 168 7 | 56 3 2 | 4 1568 9 | |--------------------+-------------------+--------------------| | 7 5 126 | 3 8 146 | 9 1246 146-2 | | 46-2 9 1268 | 14567 67 14567 | 268 3 1467-2 | | 46 3 168 | 1467 2 9 | 68 14678 5 | |--------------------+-------------------+--------------------| | 5 67-2 3 | 8 67 467 | 1 9 2-467 | | 2-16 167-2 4 | 167-2 9 3 | 5 67-2 8 | | 8 167-2 9 | 2-1467 5 1467 | 3 467-2 467-2 | *-------------------------------------------------------------*

That's a mighty tall claim for one grouped turbot fish.

Code: Select all: sdIC overlapping assignments * * * * [r1c7]=2 => [r2c2]=2 => [r8c1]=2 => [r9c4]=2 => [r7c9]=2 => [r4c8]=2 => [r5c3]=2 [r5c7]=2 => [r4c3]=2 => [r2c2]=2 => [r8c1]=2 => [r9c4]=2 => [r7c9]=2 => [r1c8]=2

When you perform the four overlapping assignments, you get the 25 eliminations I listed. Twelve of the eliminations come from candidates eliminated in the assignment cells.

by **ronk** » Wed Nov 15, 2006 7:05 pm

daj95376 wrote:When you perform the four overlapping assignments, you get the 25 eliminations I listed. Twelve of the eliminations come from candidates eliminated in the assignment cells.

Have you seen anyone else count candidates removed by placements as 'eliminations'?

by **daj95376** » Wed Nov 15, 2006 7:20 pm

ronk wrote:
daj95376 wrote:When you perform the four overlapping assignments, you get the 25 eliminations I listed. Twelve of the eliminations come from candidates eliminated in the assignment cells.

Have you seen anyone else count candidates removed by placements as 'eliminations'?

If you prefer, then I'll only count the 13 eliminations in <2>. Since they're for a single digit, I'll be happy with that distinction.

by **wapati** » Thu Nov 16, 2006 9:44 am

Busy puzzle, swordfish and 6 more patterns is the least I can do.

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

by **wapati** » Sun Nov 19, 2006 1:23 pm

There is a swordfish here but other patterns can be used instead.

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

by **wapati** » Sun Nov 19, 2006 1:44 pm

Several finned swordfish and at least one finned x-wing.

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

by **Carcul** » Sun Nov 19, 2006 3:22 pm

Wapati wrote:There is a swordfish here but other patterns can be used instead.

Code: Select all: *---------------------------------------------------------* | 4 56 1 | 9 3568 358 | 7 38 2 | | 68 2 3 | 68 1 7 | 9 4 5 | | 58 7 9 | 358 2 4 | 38 6 1 | |------------------+-------------------+------------------| | 9 56 48 | 13568 3568 2 | 145 7 36 | | 256 1 48 | 3568 7 358 | 245 9 36 | | 256 3 7 | 4 569 19 | 125 12 8 | |------------------+-------------------+------------------| | 1 4 2 | 7 358 358 | 6 38 9 | | 3 89 6 | 2 4 19 | 18 5 7 | | 7 89 5 | 138 389 6 | 1238 12 4 | *---------------------------------------------------------*

r6c7=1 or r9c7=1, but because we already have r6c8=1 or r9c8=1, then r6c1 must be "2", which solves the puzzle.

Carcul

by **re'born** » Sun Nov 19, 2006 4:08 pm

Carcul wrote:
Wapati wrote:There is a swordfish here but other patterns can be used instead.

Code: Select all
*---------------------------------------------------------* | 4 56 1 | 9 3568 358 | 7 38 2 | | 68 2 3 | 68 1 7 | 9 4 5 | | 58 7 9 | 358 2 4 | 38 6 1 | |------------------+-------------------+------------------| | 9 56 48 | 13568 3568 2 | 145 7 36 | | 256 1 48 | 3568 7 358 | 245 9 36 | | 256 3 7 | 4 569 19 | 125 12 8 | |------------------+-------------------+------------------| | 1 4 2 | 7 358 358 | 6 38 9 | | 3 89 6 | 2 4 19 | 18 5 7 | | 7 89 5 | 138 389 6 | 1238 12 4 | *---------------------------------------------------------*

r6c7=1 or r9c7=1, but because we already have r6c8=1 or r9c8=1, then r6c1 must be "2", which solves the puzzle.

Carcul

I don't understand your solution. How do you get the first 'or' statement? Are you setting up incompatible UR's again? And how do you conclude r6c1 = 2 from your two 'or' assertions?

Another solution is to color the 1's to deduce r8c7=8. After this the puzzle is easy (one type 1 UR and singles).

by **Carcul** » Sun Nov 19, 2006 4:21 pm

Wapati wrote:Several finned swordfish and at least one finned x-wing.

Code: Select all: *---------------------------------------------------------------* | 367 3567 4 | 678 3578 1 | 678 9 2 | | 16 9 378 | 2 378 38 | 16 4 5 | | 167 2 578 | 6789 4 589 | 3 167 1678 | |--------------------+--------------------+---------------------| | 4 1 2 | 89 6 7 | 89 5 3 | | 79 57 6 | 3 589 2 | 4 17 1789 | | 8 37 3579 | 4 1 59 | 679 2 679 | |--------------------+--------------------+---------------------| | 3679 367 1 | 789 2 3489 | 5 37 479 | | 5 4 379 | 1 379 6 | 2 8 79 | | 2 8 379 | 5 379 349 | 1679 1367 14679 | *---------------------------------------------------------------*

Look at the AUR in cells r89c35, and the sets A(3,5,7,8)=[r1c5|r2c56], B(1,3,5,6,7)=[r123c1|r1c2]. So, r2c3<>7 solving the puzzle.

Carcul

by **wapati** » Sun Nov 19, 2006 6:51 pm

This one has a finned swordfish but it can be passed up and still solved using patterns. XYZs, and a UR or two.

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

by **wapati** » Sun Nov 19, 2006 7:04 pm

x, xy, xyz and finned-x work on this one. Looks like a lazy 8 :!:

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

The New Sudoku Players' Forum

CHAINS, not required. Good puzzles.