#84498 in mith's 158276 T&E(3) min-expands

Post puzzles for others to solve here.

#84498 in mith's 158276 T&E(3) min-expands

Postby denis_berthier » Fri May 12, 2023 5:13 am

.
This puzzle can be solved as usual (here, with tridagon + 2 impossible patterns + chains of length ≤ 7).
But a different game can be to find as many impossible patterns as possible.

Code: Select all
+-------+-------+-------+
! 1 2 . ! . . . ! . 8 9 !
! . 5 . ! 1 . 9 ! 2 . 6 !
! . . . ! 2 . . ! 5 1 . !
+-------+-------+-------+
! . . . ! 6 . 8 ! . . 5 !
! . . . ! 9 1 . ! . 2 . !
! . . 8 ! . . . ! . 6 . !
+-------+-------+-------+
! 7 3 . ! . . . ! . 5 1 !
! . . 6 ! 5 9 . ! . . . !
! . 4 . ! . . . ! . . . !
+-------+-------+-------+
12.....89.5.1.92.6...2..51....6.8..5...91..2...8....6.73.....51..659.....4.......;21225;440648
SER =

Code: Select all
Resolution state after Singles and whips[1]:
   +----------------------+----------------------+----------------------+
   ! 1      2      347    ! 347    34567  34567  ! 347    8      9      !
   ! 348    5      347    ! 1      3478   9      ! 2      347    6      !
   ! 34689  6789   3479   ! 2      3478   347    ! 5      1      347    !
   +----------------------+----------------------+----------------------+
   ! 2349   179    123479 ! 6      2347   8      ! 13479  3479   5      !
   ! 3456   67     3457   ! 9      1      3457   ! 3478   2      3478   !
   ! 23459  179    8      ! 347    23457  23457  ! 13479  6      347    !
   +----------------------+----------------------+----------------------+
   ! 7      3      29     ! 48     246    246    ! 4689   5      1      !
   ! 28     18     6      ! 5      9      12347  ! 3478   347    23478  !
   ! 2589   4      1259   ! 378    2367   12367  ! 36789  379    2378   !
   +----------------------+----------------------+----------------------+
199 candidates.
denis_berthier
2010 Supporter
 
Posts: 4213
Joined: 19 June 2007
Location: Paris

Re: #84498 in mith's 158276 T&E(3) min-expands

Postby Cenoman » Fri May 12, 2023 5:01 pm

Solves with TH + two AIC's. No need of another impossible pattern.
Code: Select all
 +--------------------------+------------------------+-------------------------+
 |  1       2      347      |  347*  56      56      |  347*    8      9       |
 |  348     5      347      |  1     3478*   9       |  2       347*   6       |
 |  34689   6789   3479     |  2     3478    347*    |  5       1      347*    |
 +--------------------------+------------------------+-------------------------+
 |  2349    179    123479   |  6     2347*   8       |  13479   3479*  5       |
 |  3456    67     3457     |  9     1       3457*   |  3478*   2      3478    |
 |  23459   179    8        |  347*  23457   23457   |  13479   6      347*    |
 +--------------------------+------------------------+-------------------------+
 |  7       3      29       |  48    246     246     |  469-8   5      1       |
 |  28      18     6        |  5     9       12347   |  3478    347    23478   |
 |  259-8   4      1259     |  378   2367    12367   |  3679-8  379    2378    |
 +--------------------------+------------------------+-------------------------+

1. TH(347)b2356 (*) having five guardians: 8r2c5, 2r4c5, 5r5c6, 9r4c8, 8r5c7
(8)r2c5 - (8=3479)r3c3569 - r7c3 = (96)r79c7
(2)r4c5 - r6c56 = r6c1 - (2=8)r8c1 - (8=3479)b1p3469 - r7c3 = (96)r79c7
(5)r5c6 - (5=3479)r1235c3 - r7c3 = (96)r79c7
(9)r4c8 - r9c8 = (96)r79c7
(8)r5c7
=> -8 r79c7

2. (8=2)r8c1 - r7c3 = (21-5)r49c3 = (5)r9c1 => -8 r9c1; lcls, 25 placements

Code: Select all
 +-------------------+----------------------+--------------------+
 |  1     2    347   |  347*  56     56     |  347   8     9     |
 |  34    5    347   |  1     8      9      |  2     347   6     |
 |  6     8    9     |  2     347*   347*   |  5     1     347*  |
 +-------------------+----------------------+--------------------+
 |  2     7    1     |  6     34     8      |  34    9     5     |
 |  345*  6    34    |  9     1      3457   |  8     2     37-4  |
 |  345*  9    8     |  347*  3457   2      |  1     6     37-4  |
 +-------------------+----------------------+--------------------+
 |  7     3    2     |  8     46     46     |  9     5     1     |
 |  8     1    6     |  5     9      347    |  347   347   2     |
 |  9     4    5     |  37    2      1      |  6     37    8     |
 +-------------------+----------------------+--------------------+

3. (4)r3c9 = r3c56 - r1c4 = r6c4^ - r6c1 = r5c13 => -4 r5c9, r6c9^; ste
Cenoman
Cenoman
 
Posts: 2974
Joined: 21 November 2016
Location: France

Re: #84498 in mith's 158276 T&E(3) min-expands

Postby denis_berthier » Sat May 13, 2023 2:50 am

Cenoman wrote:Solves with TH + two AIC's. No need of another impossible pattern.

1. TH(347)b2356 (*) having five guardians: 8r2c5, 2r4c5, 5r5c6, 9r4c8, 8r5c7
(8)r2c5 - (8=3479)r3c3569 - r7c3 = (96)r79c7
(2)r4c5 - r6c56 = r6c1 - (2=8)r8c1 - (8=3479)b1p3469 - r7c3 = (96)r79c7
(5)r5c6 - (5=3479)r1235c3 - r7c3 = (96)r79c7
(9)r4c8 - r9c8 = (96)r79c7
(8)r5c7
=> -8 r79c7

Total length of this forcing chain = 6 + 8 + 5 + 2 + 0 + 1 = 21

This puzzle can indeed be solved with no other impossible pattern than tridagon, with max length = 7. This way, it has nothing noticeable.

But the fun here is to find many impossible patterns.
.
denis_berthier
2010 Supporter
 
Posts: 4213
Joined: 19 June 2007
Location: Paris

Re: #84498 in mith's 158276 T&E(3) min-expands

Postby denis_berthier » Wed May 24, 2023 5:30 am

.
This is the puzzle with the largest number of ORk-relations I've found (not that I have done a systematic search for this).

There are 57 different ORk-relations with no more than 8 guardians (all for the same 3 digits). Notice that this doesn't mean 57 different patterns: the same pattern can appear in more than one ORk-relation (on different cells or with different sets of guardians).
What's not possible is having an ORk-relation with a set of guardians that would be a superset of another one.
.
denis_berthier
2010 Supporter
 
Posts: 4213
Joined: 19 June 2007
Location: Paris


Return to Puzzles