The New Sudoku Players' Forum

[Leren]

Code: Select all

*-----------*
|..2|.3.|6..|
|5..|..9|..8|
|..9|..1|...|
|---+---+---|
|..4|.6.|2..|
|8.7|.5.|..1|
|...|...|.7.|
|---+---+---|
|72.|..8|..5|
|..6|.2.|4..|
|...|...|...|
*-----------*
..2.3.6..5....9..8..9..1.....4.6.2..8.7.5...1.......7.72...8..5..6.2.4...........

[denis_berthier]

.

Code: Select all

Resolution state after Singles and whips[1]:
   +-------------------+-------------------+-------------------+
   ! 14    8     2     ! 457   3     45    ! 6     19    479   !
   ! 5     1347  13    ! 6     47    9     ! 17    2     8     !
   ! 6     47    9     ! 2     8     1     ! 5     3     47    !
   +-------------------+-------------------+-------------------+
   ! 139   139   4     ! 8     6     7     ! 2     5     39    !
   ! 8     6     7     ! 39    5     2     ! 39    4     1     !
   ! 2     39    5     ! 1349  149   34    ! 8     7     6     !
   +-------------------+-------------------+-------------------+
   ! 7     2     13    ! 1349  149   8     ! 139   6     5     !
   ! 139   1359  6     ! 13579 2     35    ! 4     8     379   !
   ! 1349  13459 8     ! 13579 179   6     ! 1379  19    2     !
   +-------------------+-------------------+-------------------+

1) Simplest-first solution:

Code: Select all

x-wing-in-columns: n7{c5 c7}{r2 r9} ==> r9c4 ≠ 7, r2c2 ≠ 7
singles ==> r3c2 = 7, r3c9 = 4
finned-x-wing-in-rows: n9{r5 r7}{c7 c4} ==> r9c4 ≠ 9, r8c4 ≠ 9
finned-x-wing-in-columns: n3{c9 c1}{r4 r8} ==> r8c2 ≠ 3
biv-chain[2]: r5n3{c4 c7} - c9n3{r4 r8} ==> r8c4 ≠ 3
biv-chain[3]: r5c7{n9 n3} - c9n3{r4 r8} - b9n7{r8c9 r9c7} ==> r9c7 ≠ 9
biv-chain[3]: r7c3{n1 n3} - b1n3{r2c3 r2c2} - c2n4{r2 r9} ==> r9c2 ≠ 1
biv-chain[3]: r9c8{n9 n1} - r1n1{c8 c1} - c1n4{r1 r9} ==> r9c1 ≠ 9
biv-chain[3]: c1n9{r8 r4} - r6c2{n9 n3} - c6n3{r6 r8} ==> r8c1 ≠ 3
finned-x-wing-in-columns: n3{c1 c9}{r4 r9} ==> r9c7 ≠ 3
naked-pairs-in-a-column: c7{r2 r9}{n1 n7} ==> r7c7 ≠ 1
whip[1]: b9n1{r9c8 .} ==> r9c1 ≠ 1, r9c4 ≠ 1, r9c5 ≠ 1
naked-pairs-in-a-block: b8{r8c6 r9c4}{n3 n5} ==> r8c4 ≠ 5, r7c4 ≠ 3
naked-triplets-in-a-row: r9{c5 c7 c8}{n9 n7 n1} ==> r9c2 ≠ 9
whip[1]: b7n9{r8c2 .} ==> r8c9 ≠ 9
biv-chain[3]: r8c1{n9 n1} - r7c3{n1 n3} - c1n3{r9 r4} ==> r4c1 ≠ 9
hidden-single-in-a-column ==> r8c1 = 9
biv-chain[3]: r2c3{n3 n1} - c1n1{r1 r4} - c1n3{r4 r9} ==> r7c3 ≠ 3
stte

2) Single-step solutions:
There are 25 anti-backdoors: n1r1c1 n5r1c4 n4r1c6 n9r1c8 n7r1c9 n3r2c2 n4r2c2 n1r2c3 n7r2c5 n1r2c7 n3r4c1 n1r4c2 n9r4c9 n9r5c4 n3r5c7 n9r6c2 n3r6c6 n3r7c3 n7r8c4 n5r8c6 n3r8c9 n4r9c1 n5r9c2 n7r9c7 n1r9c8
24 of which give rise to a 1-step solution with whips[≤6].

Here are the simplest ones:

2a) With whips[4]

Code: Select all

whip[4]: r2c7{n7 n1} - c3n1{r2 r7} - r8n1{c2 c4} - r8n7{c4 .} ==> r1c9 ≠ 7
stte

OR:

Code: Select all

whip[4]: c3n1{r2 r7} - r8n1{c2 c4} - r8n7{c4 c9} - b3n7{r1c9 .} ==> r2c7 ≠ 1
stte

OR:

Code: Select all

whip[4]: r9n7{c5 c7} - r2c7{n7 n1} - c3n1{r2 r7} - r8n1{c1 .} ==> r8c4 ≠ 7
stte

OR:

Code: Select all

whip[4]: r2c7{n7 n1} - c3n1{r2 r7} - r8n1{c2 c4} - r8n7{c4 .} ==> r9c7 ≠ 7
stte

2b) With z-chains[5]:

Code: Select all

z-chain[5]: r4n1{c1 c2} - r8n1{c2 c4} - r8n7{c4 c9} - c7n7{r9 r2} - r2n1{c7 .} ==> r1c1 ≠ 1
stte

OR:

Code: Select all

z-chain[5]: r2n4{c2 c5} - b2n7{r2c5 r1c4} - r8n7{c4 c9} - c9n3{r8 r4} - b4n3{r4c1 .} ==> r2c2 ≠ 3
stte

OR:

Code: Select all

z-chain[5]: r2c3{n3 n1} - r2c7{n1 n7} - b9n7{r9c7 r8c9} - c9n3{r8 r4} - c1n3{r4 .} ==> r7c3 ≠ 3
stte

[pjb]

Code: Select all

a14      8       2      | 457    3      45     | 6     b19    c479   
 5       1347    3-1    | 6      47     9      | 17     2      8     
 6       47      9      | 2      8      1      | 5      3      47     
------------------------+----------------------+---------------------
e139     139     4      | 8      6      7      | 2      5     d39     
 8       6       7      | 39     5      2      | 39     4      1     
 2       39      5      | 1349   149    34     | 8      7      6     
------------------------+----------------------+---------------------
 7       2      g13     | 1349   149    8      | 139    6      5     
f39-1    1359    6      | 13579  2      35     | 4      8      379   
f349-1   13459   8      | 13579  179    6      | 1379   19     2     


(1)r1c1 = (1-9)r1c8 = (9)r1c9 - (9=3)r4c9 - (3)r4c1 = (3)r89c1 - (3=1)r7c3 => -1 r2c3, r89c1; stte

Phil

[eleven]

So we have another game: Find one not part of Denis' solution flood. (No one else seems to have the idea to post 8 solutions, one's solver spits out)

Code: Select all

 *-----------------------------------------------------------------*
 | d14     8       2    |  457     3    d45   |  6      19   479   |
 |  5      1347   c13   |  6       47    9    |  17     2    8     |
 |  6      47      9    |  2       8     1    |  5      3    47    |
 |----------------------+---------------------+--------------------|
 |  139    139     4    |  8       6     7    |  2      5    39    |
 |  8      6       7    |  39      5     2    |  39     4    1     |
 |  2      39      5    |  1349    149   34   |  8      7    6     |
 |----------------------+---------------------+--------------------|
 |  7      2      b13   |  1349    149   8    |  139    6    5     |
 | a139   a1359    6    |  1579-3  2   ea35   |  4      8    79-3  |
 |  1349   13459   8    |  13579   179   6    |  1379   19   2     |
 *-----------------------------------------------------------------*

(3=591)r8c126 - r7c3 = r2c3 - (1=45)r1c16 = (5=3)r8c6 => -3r8c49, stte

[denis_berthier]

So we have another game: Find one not part of Denis' solution flood. (No one else seems to have the idea to post 8 solutions, one's solver spits out)

Sorry for you that your solver is not able to find simple solutions

[eleven]

Different to you i don't need a solver to find simple solutions.

[denis_berthier]

Do you claim you never use a solver?

[jco]

Code: Select all

.---------------------------------------------------------.
|  14     8       2  |  457    3    45 |  6      19   479 |
|  5      1347   a13 |  6      47   9  | g7-1    2    8   |
|  6      47      9  |  2      8    1  |  5      3    47  |
|--------------------+-----------------+------------------|
|  139    139     4  |  8      6    7  |  2      5    39  |
|  8      6       7  |  39     5    2  |  39     4    1   |
|  2      39      5  |  1349   149  34 |  8      7    6   |
|--------------------+-----------------+------------------|
|  7      2      b13 |  1349   149  8  |  139    6    5   |
| c139   c1359    6  | d13579  2    35 |  4      8   e379 |
|  1349   13459   8  |  13579  179  6  | f1379   19   2   |
'---------------------------------------------------------'

(1)r2c3=r7c3-(1)r8c12=(1-7)r8c4=(7)r8c9-r9c7=(7)r2c7 => -1 r2c7; ste

[eleven]

Do you claim you never use a solver?

For about 50% of the solutions i post, i neither have nor know a solver, which does it that way.

[denis_berthier]

Do you claim you never use a solver?

For about 50% of the solutions i post, i neither have nor know a solver, which does it that way.

50%
I could say something similar: for 80% of what I post, no other solver does it the same way. The remaining 20% are for Subsets, Finned Fish, bivalue-chains (AIC), oddagons, UR, BUG, sk-loops, J-Exocets...)

One year ago, when I first made SudoRules public, it was not designed to search for one-step solutions. Now, after I spent a few days on it, using exactly the same generic resolution rules, it can. And it seems to me that what it finds brings some new knowledge about 1-step (and also 2-step) solutions:
- their complexity (say in terms of the length of the chain used) is unrelated to the difficulty (say W or SER) of a puzzle: the simplest 1-step solution (if any) for relatively easy puzzles may be very complex (see other recent puzzles in this forum);
- if there is a 1-step solution, there are generally many (as my multi-solution posts demonstrate);
- whips (and simpler chains such as z-chains) are as efficient for 1-step solutions as they are for normal ones.
- the central role of the notion of an anti-backdoor, which remains hidden when such solutions are presented only one at a time, has been made explicit.

[eleven]

I don't have Sudo Rules, and i am not interested in it, because the technique it uses is not manual solver like/friendly (at least i don't know about a good manual player, who would use it).

There are some public solvers around, which are very mighty and present most eliminations, which can be done by Sudo Rules, in a much better/easier/more intuitive way - of course this is my personal opinion, but i know, that most players here share this.
Again i don't want to start a discussion, it is obsolete like one about musical flavor.

However i want to ask you also to follow the unwritten rule in this thread, that everybody only presents one solution, not to spoil the fun for others, who found their own.
Whoever wants to make just a puzzle analysis is welcome, but only 1 or 2 days after the puzzle was posted and everyone had the chance to post a solution.

[Sudtyro2]

However i want to ask you also to follow the unwritten rule in this thread, that everybody only presents one solution, not to spoil the fun for others, who found their own.
Whoever wants to make just a puzzle analysis is welcome, but only 1 or 2 days after the puzzle was posted and everyone had the chance to post a solution.

Hear, hear to eleven!

SteveC

[SteveG48]

Code: Select all

 *--------------------------------------------------------------------*
 | 14     8      2      | 457    3      45     | 6      19    c479    |
 | 5      1347  d13     | 6      47     9      |d17     2      8      |
 | 6      47     9      | 2      8      1      | 5      3     c47     |
 *----------------------+----------------------+----------------------|
 |b139    139    4      | 8      6      7      | 2      5     c39     |
 | 8      6      7      | 39     5      2      | 39     4      1      |
 | 2      39     5      | 1349   149    34     | 8      7      6      |
 *----------------------+----------------------+----------------------|
 | 7      2      1-3    | 1349   149    8      | 139    6      5      |
 |a139    1359   6      | 13579  2      35     | 4      8      379    |
 |a1349   13459  8      | 13579  179    6      | 1379   19     2      |
 *--------------------------------------------------------------------*


3r89c1 = r4c1 - (3=947)r134c9 - (7=13)r2c37 => -3 r7c3 ; stte

[denis_berthier]

I don't have Sudo Rules, and i am not interested in it, because the technique it uses is not manual solver like/friendly (at least i don't know about a good manual player, who would use it).
There are some public solvers around, which are very mighty and present most eliminations, which can be done by Sudo Rules, in a much better/easier/more intuitive way - of course this is my personal opinion, but i know, that most players here share this.
Again i don't want to start a discussion, it is obsolete like one about musical flavor.

If you don't want to start a discussion, then don't start it by making false claims.
Let's keep this real: "most players here" means a handful and a half of people (and this includes those that use disguised forms of whips). This is to be compared to the 35,933 members this forum claims, as of today.

If, on the other hand, I listen to the readers of my books and users of SudoRules (which is orders of magnitude larger), whips are very user friendly: a single chain pattern allows to replace a lot of the abstruse variations of chain rules that are both less powerful and harder to find - and are generally presented without any name, in illegible personal variations of the originally inconsistent AIC notation.
If you speak of MSLS and other set covering techniques, they are computationally much costly than chains (which implies they are also harder to find manually) and I know for sure that most of those proposed here are obtained with solvers.

However i want to ask you also to follow the unwritten rule in this thread, that everybody only presents one solution, not to spoil the fun for others, who found their own.
Whoever wants to make just a puzzle analysis is welcome, but only 1 or 2 days after the puzzle was posted and everyone had the chance to post a solution.

This is quite contradictory with your previous claims. If SudoRules solutions were so much worse than those of "most players here", I don't see how they could spoil the fun for them. The fact is, if I merely wrote my 1-step solutions without saying they were found by my solver (as most of the other participants do), nobody could know they were not found manually.

However, I understand SudoRules solutions are indeed not so far from the manual ones and, in order to avoid spoiling their fun, I agree to postpone part of my 1-or-2-step answers.

[Cenoman]

However i want to ask you also to follow the unwritten rule in this thread, that everybody only presents one solution, not to spoil the fun for others, who found their own.
Whoever wants to make just a puzzle analysis is welcome, but only 1 or 2 days after the puzzle was posted and everyone had the chance to post a solution.

Thank you, eleven, for crossing the T's and dotting the I's !