The New Sudoku Players' Forum

[champagne]

2 more
42 clues skfr 11.1
same solution grid
unsolved by 'yzfwsf" program


.2.45..894...8923..89.3...42..34..98.9..1842..48.9.3...3.8..9.28.29.3.4.97..2.8.3 2889161802 QEUMCpopAAjLJ5 42 111
.2.45..894...8923..89.3...42..34..98.9..1842..48.9.3...3.9..8.287..2.9.39.28.3.4. 2889161803 QEUMCpopAATgj2 42 111

PM after basic for the first

Code: Select all

1367 2   1367 |4    5  167  |167  8    9   
4    156 1567 |167  8  9    |2    3    1567
1567 8   9    |1267 3  1267 |1567 1567 4   
-------------------------------------------
2    156 1567 |3    4  567  |1567 9    8   
3567 9   3567 |567  1  8    |4    2    567 
1567 4   8    |2567 9  2567 |3    1567 1567
-------------------------------------------
156  3   1456 |8    67 145  |9    1567 2   
8    156 2    |9    67 3    |1567 4    1567
9    7   1456 |15   2  145  |8    156  3   

could look also for a partial exocet...

[champagne]

deleted after an erroneous update

[ghfick]

Here is a solution path for the first puzzle from PhilsFolly. Again, sets make a difference

6s at r78c5 only ones in row/column => -6 r7c6, r9c46.
7s at r78c5 only ones in row/column => -7 r7c6.
Whether 6 at r7c5 or r8c5 is true, the cells at r2c2349 are reduced to 56, 1567, 167 and 1567
Whether 2 at r3c4 or r3c6 is true, the cells at r9c3468 are reduced to 1456, 15, 145 and 16
Whether 6 at r9c3 or r9c8 is true, the cells at r2c2349 are reduced to 56, 1567, 17 and 156
Cells at r1c6, r2c4, r3c46 are reduced to 16, 17, 1267 and 1267
Whether 1 at r4c2 or r8c2 is true, the cells at r1c6, r2c4, r3c46 are reduced to 16, 17, 1267 and 126
7s at r46c6 only ones in row/column => -7 r56c4.
Cells at r1c1367 are reduced to 137, 137, 16 and 167
Whether 7 at r2c3 or r2c4 is true, the cells at r9c3468 are reduced to 456, 15, 145 and 16
Whether 6 at r1c6 or r1c7 is true, the cells at r7c13, r8c2, r9c3 are reduced to 156, 1456, 16 and 456
5s at r8c79 only ones in row/column => -5 r7c8.
Whether 5 at r8c7 or r8c9 is true, the cells at r7c6, r9c46 are reduced to 14, 15 and 145
5s at r7c13 only ones in row/column => -5 r9c3.
Whether 5 at r7c1 or r7c3 is true, the cells at r248c2 are reduced to 5, 16 and 16
Sashimi X-wing of 6s (r29\c39), fin at r2c9, r9c8, eliminating 6 from r3c8, r8c9
Finned franken swordfish of 6s (r29b6\c389), fin at r4c7, eliminating 6 from r4c3
Whether 6 at r1c6 or r1c7 is true, the cells at r4c2 and r8c2 are reduced to 6 and 1
1s at r79c8 only ones in box => -1 r36c8.
Sashimi X-wing of 1s (c91\r26), fin at r13c1, eliminating 1 from r2c3
Franken swordfish of 6s (r29b6\c389), eliminating 6 from r7c38
Finned franken swordfish (type 2) of 1s (r6r2b1\c1c9b2), fin r1c3, eliminating 1 from r1c6
Simple chain: (7=1)r1c7 - (1=6)r2c9 - (6=7)r2c3 => -7 r1c13
1s at r1c13 only ones in row/column => -1 r3c1.
Simple chain: (5)r4c7 = (5-6)r8c7 = (6)r9c8 - (6=4)r9c3 - (4=5)r7c3 => -5 r4c3
Simple chain: (5)r4c6 = (5)r4c7 - (5=6)r8c7 - (6=1)r9c8 - (1=5)r9c4 => -5 r56c4, r9c6
Naked pairs of 57 at r58c9 => -5 r6c9, -7 r6c9
Simple chain: (5=1)r4c7 - (1)r4c3 = (1-5)r6c1 = (5)r6c6 => -5 r4c6
stte

[eleven]

Whether 6 at r7c5 or r8c5 is true, the cells at r2c2349 are reduced to 56, 1567, 167 and 1567

I don't understand, can you elaborate please ?

[ghfick]

From PhilsFolly: Sets

A naked set is a group of 2, 3, 4, or 5 cells that occur in the same unit (row, column or box), which contain the same number of candidates. A naked triple, for example, has three cells having 3 candidates between them. If these contained, say, 123, 123, 123, then in the solved puzzle they could be one of 6 combinations: 123, 132, 213, 231, 312, 321, whereas if they contained 12, 13, 23 then there would be only two possibilities: 132, 213. When each of the possible combinations is used to initiate a forcing net, and those combination(s) causing a contradiction removed, then the remaining combinations can be re-combined. The resulting cells may have a reduced number of candidates.

In order to improve the power of the method, an extension is implemented which involves identifying all instances of two numbers in a row , column or box. Each number of the pair in turn is added to the naked set combination and as above, the forcing net generated. If a contradiction occurs in each case, and since one of the pair of numbers must be true, then the combination can be eliminated as above. The Easter Monster puzzle can be solved using this method following the SK loop eliminations.

[champagne]

Having in priority to release the next status of the player PH view with a TH, I can't follow this point

But

This set of invalid 4 clues patterns pushes me to think of a vicinity search on the 2 patterns shown in this thread.
I'll likely start it in the next days

[eleven]

@ghfick So it's a program solving method, which i cannot follow manually without a lot of output, which is not provided.
I don't see, why i should prefer it to backtracking, which can be done by a program a lot faster.

This puzzle is surprisingly difficult with it's 4 digit patterns. Impossible patterns are not easy to prove and often just allow weak eliminations. E.g. i needed these patterns to get a number (2), which did not help much then.

Code: Select all

 *-----------------------------------------------------*
 | .    .    .     | .    .  167   | 1567  .     .     |
 | .    156  1567  | 167  .  .     | .     .     1567  |
 | .    .    .     | .    .  167   | 1567  157   .     |
 |-----------------+---------------+-------------------|
 | .    156  1567  | .    .  567   | 1567  .     .     |
 | .    .    .     | .    .  .     | .     .     .     |
 | .    .    .     | .    .  .     | .     .     .     |
 |-----------------+---------------+-------------------|
 | 156  .    156   | .    .  .     | .     157   .     |
 | .    156  .     | .    .  .     | 157   .     157   |
 | .    .    .     | .    .  15    | .     .     .     |
 *-----------------------------------------------------*
 *-----------------------------------------------------*
 | .    .    .     | .    .  167   | 1567  .     .     |
 | .    156  1567  | 167  .  .     | .     .     1567  |
 | .    .    .     | .    .  167   | 1567  1567  .     |
 |-----------------+---------------+-------------------|
 | .    156  1567  | .    .  567   | 1567  .     .     |
 | .    .    .     | .    .  .     | .     .     .     |
 | .    .    .     | .    .  .     | .     .     .     |
 |-----------------+---------------+-------------------|
 | 156  .    .     | .    .  15    | .     1567  .     |
 | .    156  .     | .    .  .     | 1567  .     1567  |
 | .    .    156   | 15   .  .     | .     156   .     |
 *-----------------------------------------------------*

So this approach also does not seem to be effective for manual solving.

[champagne]

note : I made a wrong action and a false update of the second post, now deleted
here is the right place for the post

Hi "'eleven'"

I would agree that these eliminations are more analysis tools than manual solving steps.

what I find amazing with this puzzle is that if you push the partially solved version given earlier

.2345..894...8923..89.32..42..34..98.9..1842..48.9.3...348..9.28.29.3.4.97..2.8.3
to 48 given (missing 234)
.2345..894...8923..89.32..42..34..9839..1842..4829.3...348..9.28.29.3.4.97..248.3 ED=9.5/9.5/2.8

you get a 48 given pearl skfr 9.5 needing still nested chains

Code: Select all

167  2   3    |4   5  167 |167  8    9   
4    156 1567 |167 8  9   |2    3    1567
1567 8   9    |167 3  2   |1567 1567 4   
-----------------------------------------
2    156 1567 |3   4  567 |1567 9    8   
3    9   567  |567 1  8   |4    2    567
1567 4   8    |2   9  567 |3    1567 1567
-----------------------------------------
156  3   4    |8   67 15  |9    1567 2   
8    156 2    |9   67 3   |1567 4    1567
9    7   156  |15  2  4   |8    156  3   


solution grids 2889161802 and 2889161803 have surely something special. All pairs of the digits 1567 must have a 18 cells unique unavoidable set. This is why I expect interesting output of a vicinity on these 2 seeds.

On a solver view, this valid 4 digits patterrn seems very hard to solve. I see no uniqueness impossible pattern to help, but I am not a skill solver.

For sure, here, a 4 digits analysis gives you all eliminations, but with no interest for the player