Almost X-Wing Example in Loop

Advanced methods and approaches for solving Sudoku puzzles

Almost X-Wing Example in Loop

Postby doduff » Fri Jun 02, 2006 7:16 am

Here is a puzzle I made. The solution is probably the most ordered solution that can exist('all traveling triples'), but What I find interesting is the neat loop following. After Simple Sudoku could give no more hints I found the folowing elimination and looked at it through two 'different' chains.

I just entered the solution and deleted numbers until Simple Sudoku said the puzzle had multiple solutions if I deleted another number.

Code: Select all
Original Puzzle:
 *-----------*
 |1..|..6|.89|
 |...|.2.|4.6|
 |...|.8.|...|
 |---+---+---|
 |9..|.45|6.8|
 |6..|...|...|
 |.4.|.7.|..2|
 |---+---+---|
 |..1|...|...|
 |567|...|23.|
 |.3.|5.7|...|
 *-----------*

After basic logic and coloring:

*--------------------------------------------------------------------*
| 1      2      34     | 47     5      6      | 37     8      9      |
| 378    589    389    | 179    2      139    | 4      157    6      |
| 347    59     6      | 1479   8      1349   | 1357   2      357    |
|----------------------+----------------------+----------------------|
| 9      17     23     | 23     4      5      | 6      17     8      |
| 6      17     258    | 189    139    1289   | 13579  4      357    |
| 38     4      358    | 6      7      189    | 1359   159    2      |
|----------------------+----------------------+----------------------|
| 48     89     1      | 23     36     24     | 5789   5679   57     |
| 5      6      7      | 1489   19     1489   | 2      3      14     |
| 2      3      489    | 5      16     7      | 89     69     14     |
*--------------------------------------------------------------------*


I found the following loop with an almost x wing. For a good explanation check this forum(Almost X-Wing) and this forum(Multiple implication and grouped nodes).

Here are some pictures:
Image

In the picture above:
[r4c3]-3-[r6c1]=3=[Almost X-Wing:r23c16]-3-[r3c79]=3=[r1c7]-3-[r56c7]=3=[r5c9]-3-[r5c5]=3=[r4c4]-3-[r4c3] => [r4c3]<>3

Where the extra -3's around the x-wing are those exclude by the x-wing and its locked candidates.

Image

In the picture above:
[r4c3]-3-[r4c4]=3=[r5c5]-3-[r5c9]=3=[r56c7]-3-[r13c7]=3=[r3c9](-3-[r3c1])-3-[r3c6]=3=[r2c6]-3-[r2c13|(r3c1)]=3=[r1c3]-3-[r4c3] => [r4c3]<>3

Where you really have to follow the chain in a particular direction to see the conclusion easily.
doduff
 
Posts: 32
Joined: 29 May 2006

Re: Almost X-Wing Example in Loop

Postby Myth Jellies » Fri Jun 02, 2006 8:42 am

Code: Select all
*--------------------------------------------------------------------*
| 1      2     *34     | 47     5      6      |*37     8      9      |
| 378    589    389    | 179    2      139    | 4      157    6      |
| 347    59     6      | 1479   8      1349   | 1357   2      357    |
|----------------------+----------------------+----------------------|
| 9      17    -23     | 23     4      5      | 6      17     8      |
| 6      17     258    | 189    139    1289   | 13579  4      357    |
|#38     4     *358    | 6      7      189    |*1359   159    2      |
|----------------------+----------------------+----------------------|
| 48     89     1      | 23     36     24     | 5789   5679   57     |
| 5      6      7      | 1489   19     1489   | 2      3      14     |
| 2      3      489    | 5      16     7      | 89     69     14     |
*--------------------------------------------------------------------*

...or you could just use a finned r16c37 x-wing (fin = r6c1). Either the fin is 3 or the x-wing is true, and either one will eliminate the 3 in r4c3.
Myth Jellies
 
Posts: 593
Joined: 19 September 2005

Postby Carcul » Fri Jun 02, 2006 9:44 am

The coloring is not "needed":

Code: Select all
 *--------------------------------------------------------------------*
 | 1      2      34     | 347    5      6      | 37     8      9      |
 | 378    589    389    | 1379   2      139    | 4      157    6      |
 | 347    59     6      | 13479  8      1349   | 1357   2      357    |
 |----------------------+----------------------+----------------------|
 | 9      17     23     | 23     4      5      | 6      17     8      |
 | 6      17     2358   | 12389  139    12389  | 13579  4      357    |
 | 38     4      358    | 6      7      1389   | 1359   159    2      |
 |----------------------+----------------------+----------------------|
 | 48     89     1      | 234    36     234    | 5789   5679   57     |
 | 5      6      7      | 1489   19     1489   | 2      3      14     |
 | 2      3      489    | 5      16     7      | 89     69     14     |
 *--------------------------------------------------------------------*

[r4c3](-3-[r4c4]-2-[r7c4])-3-[r6c1]-8-[r7c1]-4-[r7c4](-3-[r5c4])-3-
[r123c4]=3=[r23c6]-3-[r56c6]=3=[r5c5]-3-[r5c79]=3=[r6c7]-3-[r1c7]
=3=[r1c3]-3-[r4c3], => r4c3<>3 and the puzzle is solved.

Carcul
Carcul
 
Posts: 724
Joined: 04 November 2005

Postby doduff » Fri Jun 02, 2006 5:21 pm

Man... Everytime I find something that I think is cool, someone shows me an easier way to reach the same conclusion!

Thanks for pointing that stuff out to me.
doduff
 
Posts: 32
Joined: 29 May 2006


Return to Advanced solving techniques

cron