The Ultimate FISH Guide

Advanced methods and approaches for solving Sudoku puzzles

Postby ronk » Tue Feb 27, 2007 1:02 pm

StrmCkr wrote:i was wondering if it is some kind of fish pattern or something else

Code: Select all
 .  .  . |  .  .  . |  .  .  .
 .  .  . |  .  .  x |  x  .  .
 .  .  . |  *  .  # |  .  x  .
---------+----------+----------
 .  .  * |  .  .  . |  #  x  .
 .  .  . |  .  .  . |  .  .  .
 .  x  # |  .  .  . |  *  .  .
---------+----------+----------
 .  x  . |  #  .  * |  .  .  .
 .  .  x |  x  .  . |  .  .  .
 .  .  . |  .  .  . |  .  .  .

There are no valid exclusions based on this single-digit pattern.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby StrmCkr » Tue Feb 27, 2007 6:50 pm

removed
Last edited by StrmCkr on Wed Jun 26, 2013 1:31 am, edited 1 time in total.
Some do, some teach, the rest look it up.
stormdoku
User avatar
StrmCkr
 
Posts: 1430
Joined: 05 September 2006

Postby ronk » Tue Feb 27, 2007 7:05 pm

StrmCkr wrote:i used the only possible views of both x's chains to identify a pattern that leads to 2 solutions/ or same candidate for mutiple rows/columns in same box to find out the stars are never slected.

The following pattern is also a valid possibility -- with all the '*'s and none of the '#'s.
Code: Select all
 .  .  . |  .  .  . |  .  .  .
 .  .  . |  .  .  x |  x  .  .
 .  .  . |  *  .  . |  .  x  .
---------+----------+----------
 .  .  * |  .  .  . |  .  x  .
 .  .  . |  .  .  . |  .  .  .
 .  x  . |  .  .  . |  *  .  .
---------+----------+----------
 .  x  . |  .  .  * |  .  .  .
 .  .  x |  x  .  . |  .  .  .
 .  .  . |  .  .  . |  .  .  .
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby StrmCkr » Wed Feb 28, 2007 1:00 am

removed
Last edited by StrmCkr on Wed Jun 26, 2013 1:32 am, edited 1 time in total.
Some do, some teach, the rest look it up.
stormdoku
User avatar
StrmCkr
 
Posts: 1430
Joined: 05 September 2006

Postby ronk » Wed Feb 28, 2007 1:41 am

StrmCkr wrote:combine the 2 views and you get [...]

I no longer see a claim that valid exclusions exist in this pattern. That's good ... because there aren't any.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby StrmCkr » Wed Feb 28, 2007 1:49 am

removed
Last edited by StrmCkr on Wed Jun 26, 2013 1:32 am, edited 1 time in total.
Some do, some teach, the rest look it up.
stormdoku
User avatar
StrmCkr
 
Posts: 1430
Joined: 05 September 2006

Postby ronk » Wed Feb 28, 2007 1:59 am

StrmCkr wrote:the *'s are never valid.

All the '*'s can never be simultaneously valid, but some of them can be.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby StrmCkr » Wed Feb 28, 2007 2:11 am

removed
Last edited by StrmCkr on Wed Jun 26, 2013 1:32 am, edited 1 time in total.
Some do, some teach, the rest look it up.
stormdoku
User avatar
StrmCkr
 
Posts: 1430
Joined: 05 September 2006

Postby ronk » Wed Feb 28, 2007 3:11 am

StrmCkr wrote:if the *'s where valid at all pls explain to me this then.

why are none of them represented in any subgrid chain using the x's chains?
and only #'s squares actually contain a "yes" in the 2 views i found as valid

Because your logic is faulty? Here are two possible outcomes showing that none of the '*' candidates may be eliminated.
Code: Select all
 X  .  . |  .  .  . |  .  .  .        X  .  . |  .  .  . |  .  .  .
 .  .  . |  .  .  . |  x  .  .        .  .  . |  .  .  x |  .  .  .
 .  .  . |  *  .  . |  .  .  .        .  .  . |  .  .  . |  .  x  .
---------+----------+----------      ---------+----------+----------
 .  .  . |  .  .  . |  .  x  .        .  .  * |  .  .  . |  .  .  .
 .  .  . |  .  X  . |  .  .  .        .  .  . |  .  X  . |  .  .  .
 .  x  . |  .  .  . |  .  .  .        .  .  . |  .  .  . |  *  .  .
---------+----------+----------      ---------+----------+----------
 .  .  . |  .  .  * |  .  .  .        .  x  . |  .  .  . |  .  .  .
 .  .  x |  .  .  . |  .  .  .        .  .  . |  x  .  . |  .  .  .
 .  .  . |  .  .  . |  .  .  X        .  .  . |  .  .  . |  .  .  X

If you can't see this, then I don't think I can be of any further help.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby StrmCkr » Wed Feb 28, 2007 7:52 am

removed
Last edited by StrmCkr on Wed Jun 26, 2013 1:33 am, edited 1 time in total.
Some do, some teach, the rest look it up.
stormdoku
User avatar
StrmCkr
 
Posts: 1430
Joined: 05 September 2006

Postby ronk » Sat Mar 03, 2007 3:54 pm

Here is a puzzle by daj95376 which has three sashimi franken swordfish patterns for the digits 3, 6 and 9 that overlap "exactly." They do not overlap in the strict sense of the word "exactly ", because some candidates of the finless fish and the fin cells may be missing. They are the same pattern however.

Code: Select all
 
.721.8.54.4.....81............8.4.1.85..2..47.2.5.7............41.....3.59.4.217. #UF001

After SSTS (Simple Sudoku Technique Set)
X369    /7      /2       |/1      X369    /8       |X369    /5      /4
 369     4       3569    | 2       35679   3569    | 3679    8       1
 1369    368     135689  | 3679    4       3569    | 3679    2       369
-------------------------+-------------------------+------------------------
X7      #36     #369     |/8      X369    /4       |/25     /1      /25
 8       5       1369    | 369     2       1369    |X369    /4      /7
 1**369  2       4       | 5       1369    7       |X8      #69     #369
-------------------------+-------------------------+------------------------
 2       368     3678    | 3679    1356789 13569   | 4       69      5689
 4       1       678     | 679     56789   569     | 25      3       25689
 5       9       368     | 4       368     2       | 1       7       68

 sashimi franken swordfish r24b6\c157 plus fins r4c23 and r6c89, implies r6c1<>369

 
 X  /  /  |  /  X  /  |  X  /  /
 *  .  .  |  .  *  .  |  *  .  .
 *  .  .  |  .  *  .  |  *  .  .
----------+-----------+----------
 X  #  #  |  /  X  /  |  /  /  /
 *  .  .  |  .  *  .  |  X  /  /
**  .  .  |  .  *  .  |  X  #  #
----------+-----------+----------
 *  .  .  |  .  *  .  |  *  .  .
 *  .  .  |  .  *  .  |  *  .  .
 *  .  .  |  .  *  .  |  *  .  .

 sashimi franken swordfish r24b6\c157 plus fins r4c23 and r6c89, implies r6c1<>X

 '/'  <=> empty cell (candidate must be missing)
 'X'  <=> fish cell (candidate may be missing)
 '#'  <=> fin cell (candidate may be missing)
 '*'  <=> elimination if all fin cells are false (empty, missing)
 '**' <=> elimination whether or not a fin cell is true
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby Obi-Wahn » Mon Mar 19, 2007 11:08 pm

Hi,
I guess this thread deserves a little push.:D

I have a question for the purists about nomenclature. The following puzzle was #6 in the German P.M. magazine Sudoku Trainer Dec. 2006:

Code: Select all
..9..7.3...1......23.9...5.......4.8...31..2.16..8.....4........8.5.9...3......45

Some naked and hidden singles together with locked candidates and/or disjoint subsets lead to:

Code: Select all
.---------------.---------------.---------------.
| 468  5    9   | 1    2    7   | 68   3    46  |
| 468  7    1   | 48   5    3   | 268  89   2469|
| 2    3    48  | 9    6    48  | 17   5    17  |
:---------------+---------------+---------------:
| 5    2    3   | 7    9    6   | 4    1    8   |
| 48   9    478 | 3    1    45  | 567  2    67  |
| 1    6    47  | 24   8    245 | 357  79   379 |
:---------------+---------------+---------------:
| 9    4    5   | 6    3    1   | 278  78   27  |
| 7    8    2   | 5    4    9   | 13   6    13  |
| 3    1    6   | 28   7    28  | 9    4    5   |
'---------------'---------------'---------------'

I found this interesting because you can use a Turbot Fish, Finned X-Wings or even a Broken Wing to eliminate the digit 4 from r5c6 and r6c3, which is quite sophisticated for a newspaper puzzle.

But consider the following classical Finned X-Wing r35c36 with fin r5c1:
Code: Select all
 4  .  . | .  .  . | .  .  4
 4  .  . | 4  .  . | .  .  4
 .  . *4 | .  . *4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
#4  . *4 | .  . *4 | .  .  .
 .  . -4 | 4  .  4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .

Sashimi or not sashimi, that's the question. At first it appears to be not sashimi because all four vertices of the X-Wing are present. But if it wasn't for the fin r5c1, there would be locked candidates in c3b4 leading to r3c3<>4 => r3c6=4 => r5c6<>4 => r5c3=4. So, without the fin the X-Wing degenerates to singles and thus could be called sashimi.

What's your opinion?
User avatar
Obi-Wahn
 
Posts: 63
Joined: 05 January 2007
Location: Darmstadt, Germany

Postby Para » Mon Mar 19, 2007 11:41 pm

Hi

I think the difference is that a sashimi would be singles without interaction with cells outside the sashimi in comparison to this one falling apart in singles because of locked candidates outside the original finned x-wing. I mean a normal x-wing could also create a chain that leads to singles. But that doesn't make it sashimi.

Para
Para
 
Posts: 46
Joined: 20 February 2007

Postby ronk » Tue Mar 20, 2007 12:06 am

Obi-Wahn wrote:But consider the following classical Finned X-Wing r35c36 with fin r5c1:
Code: Select all
 4  .  . | .  .  . | .  .  4
 4  .  . | 4  .  . | .  .  4
 .  . *4 | .  . *4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
#4  . *4 | .  . *4 | .  .  .
 .  . -4 | 4  .  4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .

Sashimi or not sashimi, that's the question.

It's not sashimi.

When all fin cells are false, the fish itself should degenerate to be sashimi. That the finless fish would no longer be the simplest move on the grid is irrelevant.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby Myth Jellies » Tue Mar 20, 2007 11:50 pm

Obi-Wahn wrote:But consider the following classical Finned X-Wing r35c36 with fin r5c1:
Code: Select all
 4  .  . | .  .  . | .  .  4
 4  .  . | 4  .  . | .  .  4
 .  . *4 | .  . *4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
#4  . *4 | .  . *4 | .  .  .
 .  . -4 | 4  .  4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .

Sashimi or not sashimi, that's the question....What's your opinion?


I agree with Ronk on this one, but this layout is interesting. If you look at the complementary finned fish in r126/c149...

Code: Select all
*4  .  . |*.  .  . | .  . *4
*4  .  . |*4  .  . | .  . *4
 .  .  4 | .  .  4 | .  .  .
---------+---------+---------
 .  .  . | .  .  . | .  .  .
#4  .  4 | .  .  4 | .  .  .
*.  . -4 |*4  .  4 | .  . *.
---------+---------+---------
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .
 .  .  . | .  .  . | .  .  .

...you get a 231 structure in rows which probably does qualify as sashimi.

You can also describe it with
r5c4 = r12/c19 - r2c4 = r6c4 => r5c6 & r6c3 <> 1
if you don't mind using fish/constraint groups in AICs
Myth Jellies
 
Posts: 593
Joined: 19 September 2005

PreviousNext

Return to Advanced solving techniques