The New Sudoku Players' Forum

[AnotherLife]

Can anyone offer a good solution to this puzzle without help from software (one step is not obligatory)?

Code: Select all

2.4.371.6.7..6....6..4.5.78...57.68.5362489177.869....96.7.48..3.7.8..6...13.67.9

Code: Select all

.-------------------.----------------.----------------------.
| 2     5689   4    | 89   3    7    | 1      359     356   |
| 138   7      359  | 189  6    1239 | 23459  23459   2345  |
| 136   169    369  | 4    123  5    | 239    2379    8     |
:-------------------+----------------+----------------------:
| 14    1249   29   | 15   7    13   | 6      234589  2345  |
| 5     3      69   | 2    4    8    | 49     1       7     |
| 1467  1246   8    | 156  9    136  | 2345   2345    2345  |
:-------------------+----------------+----------------------:
| 9     2568   2356 | 7    125  4    | 2358   2358    1235  |
| 347   245    2357 | 159  8    129  | 2345   6       12345 |
| 468   24568  1    | 3    25   26   | 7      2458    9     |
'-------------------'----------------'----------------------'


[Leren]

Code: Select all

*-----------------------------------------------*
| 2     58a  4  | 89  3    7  | 1    59    6    |
|A18a   7    35 | 8-1 6    29 | 2345 23459 2345 |
| 6     19   39 | 4   12   5  | 23   7     8    |
|---------------+-------------+-----------------|
| 14    1249 29 | 5   7    13 | 6    8     234  |
| 5     3    6  | 2   4    8  | 9    1     7    |
| 7     124  8  | 6   9    13 | 2345 2345  2345 |
|---------------+-------------+-----------------|
| 9     6    25 | 7   125  4  | 8    235   1235 |
| 3   Bb245b 7  |a19  8   a29 | 245  6     1245 |
|A48    2458 1  | 3   25   6  | 7    245   9    |
*-----------------------------------------------*

3 Petal Death Blossom: Stem Cell r8c2 {245} :

(1=2) r8c46      - (2) r8c2;
(1=4) r29c1      - (4) r8c2;
(1=5) r1c2, r2c1 - (5) r8c2; => - 1 r2c4; stte

Leren

[AnotherLife]

3 Petal Death Blossom: Stem Cell r8c2 {245} :

(1=2) r8c46 - (2) r8c2;
(1=4) r29c1 - (4) r8c2;
(1=5) r1c2, r2c1 - (5) r8c2; => - 1 r2c4; stte

Hello, Leren,
It is great that there is a pattern leading to one-step solution, and it is more convincing than the corresponding forcing chain verity (kraken). Have you found it yourself?

[denis_berthier]

.
There is an elementary solution, using only very classical Subsets and bivalue-chains.

Code: Select all

   +-------------------+-------------------+-------------------+
   ! 2     589   4     ! 89    3     7     ! 1     59    6     !
   ! 18    7     359   ! 189   6     129   ! 2345  23459 2345  !
   ! 6     19    39    ! 4     12    5     ! 23    7     8     !
   +-------------------+-------------------+-------------------+
   ! 14    1249  29    ! 5     7     13    ! 6     8     234   !
   ! 5     3     6     ! 2     4     8     ! 9     1     7     !
   ! 7     124   8     ! 6     9     13    ! 2345  2345  2345  !
   +-------------------+-------------------+-------------------+
   ! 9     6     25    ! 7     125   4     ! 8     235   1235  !
   ! 3     245   7     ! 19    8     129   ! 245   6     1245  !
   ! 48    2458  1     ! 3     25    6     ! 7     245   9     !
   +-------------------+-------------------+-------------------+
107 candidates, 461 csp-links and 461 links. Density = 8.13%

whip[1]: r3n9{c3 .} ==> r2c3 ≠ 9, r1c2 ≠ 9
whip[1]: b5n1{r6c6 .} ==> r8c6 ≠ 1, r2c6 ≠ 1
hidden-pairs-in-a-row: r2{n1 n8}{c1 c4} ==> r2c4 ≠ 9
finned-x-wing-in-rows: n5{r1 r8}{c2 c8} ==> r9c8 ≠ 5, r7c8 ≠ 5
biv-chain[2]: r3n2{c7 c5} - c6n2{r2 r8} ==> r8c7 ≠ 2
biv-chain[3]: r2c3{n3 n5} - r7c3{n5 n2} - r7c8{n2 n3} ==> r2c8 ≠ 3
biv-chain[3]: r4c1{n4 n1} - b1n1{r2c1 r3c2} - c2n9{r3 r4} ==> r4c2 ≠ 4
biv-chain[3]: r7c3{n2 n5} - b1n5{r2c3 r1c2} - c2n8{r1 r9} ==> r9c2 ≠ 2
finned-x-wing-in-rows: n2{r9 r3}{c5 c8} ==> r2c8 ≠ 2
biv-chain[3]: r9n2{c8 c5} - r3c5{n2 n1} - r7n1{c5 c9} ==> r7c9 ≠ 2
biv-chain[3]: r9n2{c8 c5} - c5n5{r9 r7} - r7c3{n5 n2} ==> r7c8 ≠ 2
naked-single ==> r7c8 = 3
biv-chain[4]: r1c8{n5 n9} - c4n9{r1 r8} - b8n1{r8c4 r7c5} - r7c9{n1 n5} ==> r2c9 ≠ 5
biv-chain[4]: r2c3{n5 n3} - r3n3{c3 c7} - r3n2{c7 c5} - r7n2{c5 c3} ==> r7c3 ≠ 5
stte

[Leren]

Hi AnotherLife,

No, I used my solver to find the solution because, like yzfwsf, I freely admit to being a poor manual solver.

However, someone with better manual solving skills than me (that's probably just about everyone) could have found it, and I think that's what really counts.

Leren

[Cenoman]

Code: Select all

 +-------------------+------------------+------------------------+
 |  2   c58     4    |  89   3     7    |  1      59      6      |
 |  18   7     A35   |  18   6    C29   |  2345   23459   2345   |
 |  6   B19    B39   |  4   B12    5    |  23     7       8      |
 +-------------------+------------------+------------------------+
 |  14   1249   29   |  5    7     13   |  6      8       234    |
 |  5    3      6    |  2    4     8    |  9      1       7      |
 |  7    124    8    |  6    9     13   |  2345   2345    2345   |
 +-------------------+------------------+------------------------+
 |  9    6    Aa25   |  7    125   4    |  8      235     1235   |
 |  3   b45-2   7    |  19   8    D29   |  245    6       1245   |
 |  48  c458-2  1    |  3    25    6    |  7      245     9      |
 +-------------------+------------------+------------------------+

1. (2=5)r7c3 - r8c2 = (58)r19c2 => -2 r9c2
2. (2=53)r27c3 - (3=192)r3c235 - r2c6 = (2)r8c6; ste

[Cenoman]

Hi AnotherLife,

No, I used my solver to find the solution because, like yzfwsf, I freely admit to being a poor manual solver.

However, someone with better manual solving skills than me (that's probably just about everyone) could have found it, and I think that's what really counts.

Leren

Hi Leren,
Do you consider the following solution:
Kraken cell (245)r8c2
(2)r8c2 - (2=91)r8c46
(4)r8c2 - (4=81)r29c1
(5)r8c2 - (5=81)b1p24
=> -1 r2c4; ste
as different of yours (except the pattern name...) ?

[AnotherLife]

Leren, I am not content with my solving skills, either. However, all that I have is due to your help at the initial stage of my learning process. Merci bien! Do you remember that then I had had problems even with the locked candidates rule? Frankly speaking, I tried to solve this puzzle manually and came to the elimination r2c1<>8 similar to yours but only in three steps. Actually, the most obvious moves are ineffective but both HoDoKu and YZF_ Sudoku perform these four useless eliminations, so their solutions take 7-8 main steps, which is inappropriate.

[Leren]

Hi Cenoman,

I think it's the same. AFAIK a Death Blossom is just a subset of the Kraken cell idea where all the legs of the construction happen to be ALS's.

Also a 2 petal Death Blossom is just an ALS XY Wing with the central ALS being a bi-value cell, so I've turned that off in my solver, because it adds no value.

I never coded for a 4 petal Death Blossom because I didn't have an example to work with and I thought they would be very rare, but ghfick found one here. If they found that manually, wow !

Finally, you can read Andrew Stuart's take on Death Blossoms here.

Leren

<edit> I just read through Andrew Stuart's Death Blossom article, and he now has an example with 2 ALS's and a 5 digit stem cell. So it seems that Death Blossom science has moved on. I wish all those wanting to pursue this all the best, but I'm passing, and just sticking with what I have. Leren

[SteveG48]

Code: Select all

 *--------------------------------------------------------------------*
 | 2     a58     4      | 89     3      7      | 1      59     6      |
 | 1-8    7     a35     |e18     6      29     | 2345   23459  2345   |
 | 6     a19    a39     | 4      12     5      | 23     7      8      |
 *----------------------+----------------------+----------------------|
 | 14     1249  b29     | 5      7      13     | 6      8      234    |
 | 5      3      6      | 2      4      8      | 9      1      7      |
 | 7     b124    8      | 6      9      13     | 2345   2345   2345   |
 *----------------------+----------------------+----------------------|
 | 9      6     c25     | 7      125    4      | 8      235    1235   |
 | 3     c245    7      |d19     8     d29     | 245    6      1245   |
 | 48     2458   1      | 3      25     6      | 7      245    9      |
 *--------------------------------------------------------------------*


(8=3519)b1p2689 - 9r4c3|1r6c2 = (24)b4p38 - 2r7c3|4r8c2 = 5r7c3&2r8c2 - (2=91)r8c46 - (1=8)r2c4 => -8 r2c1 ; stte

Pretty similar to Leren's.