The New Sudoku Players' Forum

by **ArkieTech** » Sun Dec 08, 2019 11:51 am

Code: Select all: *-----------* |...|...|17.| |.31|.4.|5..| |9..|1..|..6| |---+---+---| |...|71.|6..| |...|658|...| |..3|.94|...| |---+---+---| |3..|..1|..2| |..9|.3.|41.| |.48|...|...| *-----------* ......17..31.4.5..9..1....6...71.6.....658.....3.94...3....1..2..9.3.41..48......

Play/Print this puzzle online

by **SpAce** » Sun Dec 08, 2019 1:14 pm

faulty original: Show

Code: Select all: .--------------------.---------------------.---------------------. | 24568 2568 2456 | 3 268 259 | 1 7 49-8 | | g67 3 1 | c89 4 g67 | 5 2 d(8)9 | | 9 2578 2457 | 1 278 25 | e38 348 6 | :--------------------+---------------------+---------------------: | 258 2589 25 | 7 1 3 | 6 4589 45-8 | | 47 19 47 | 6 5 8 | 2 39 13 | | f568 1568 3 | 2 9 4 | f78 f58 157-8 | :--------------------+---------------------+---------------------: | 3 567 567 | 4 a6[8]-7 1 | 9 568 2 | | 2567 2567 9 | b58 3 h6(7) | 4 1 a5[8]-7 | | 1 4 8 | 59 f267 29 | 37 356 357 | '--------------------'---------------------'---------------------'

(8)r7c5&r8c9 =X= (856)r6c781 - (6,7)r2c16 = (7)r8c6 => -7 r7c5,r8c9, (-8 r146c9); btte

full chain: Show

--
Edit. Funny mistake (hopefully) corrected, thanks to eleven.

by **eleven** » Sun Dec 08, 2019 4:03 pm

SpAce wrote:(8)r7c5&r8c9 =X= (87)r69c7 - r9c5 = (7)r8c6 => -7 r7c5,r8c9, (-8 r146c9); btte

[7]r9c5 = (7)r8c6 ? Do you remember (8)r7c5 ?

Code: Select all: *--------------------------------------------------------------------* | c24568 2568 c2456 | 3 268 259 | 1 7 b489 | | Hd67 3 1 | 89 4 G67 | 5 2 89 | | 9 2578 2457 | 1 278 25 | 38 348 6 | |-------------------------+-------------------+----------------------| | 258 2589 25 | 7 1 3 | 6 4589 a458 | | d47 19 47 | 6 5 8 | 2 39 13 | | i568 1568 3 | 2 9 4 | 7-8 ia58 157-8 | |-------------------------+-------------------+----------------------| | 3 567 D567 | 4 678 1 | 9 568 2 | | E2567 E2567 9 | 58 3 F67 | 4 1 578 | | 1 4 8 | 59 267 29 | 37 356 357 | *--------------------------------------------------------------------*

Both 4r1c13 imply 6r2c1:
(58=4)b6p38 - r1c9 = 476r152 | [(4-6)r1c3 = r7c3 - r8c12 = r8c6 - r2c7 = r2c1] - (6=58)r6c18 => -8r6c79, stte

by **Cenoman** » Sun Dec 08, 2019 5:22 pm

Code: Select all: +------------------------+-------------------+---------------------+ | 24568 2568 a2456 | 3 268 259 | 1 7 489 | |xf67 3 1 | 89 4 e67 | 5 2 89 | | 9 2578 A2457 | 1 278 25 | 38 B348 6 | +------------------------+-------------------+---------------------+ | 258 2589 25 | 7 1 3 | 6 C4589 458 | | x47 D19 w47 | 6 5 8 | 2 C39 C13 | |yg568 zh568-1 3 | 2 9 4 | 78 58 1578 | +------------------------+-------------------+---------------------+ | 3 567 b567 | 4 678 1 | 9 568 2 | | c2567 c2567 9 | 58 3 d67 | 4 1 578 | | 1 4 8 | 59 267 29 | 37 356 357 | +------------------------+-------------------+---------------------+

Kraken column (4)r135c3
(4-6)r1c3 = r7c3 - r8c12 = r8c6 - r2c6 = r2c1 - r6c1 = (6)r6c2
(4)r3c3 - r3c8 = (4-931)b6p256 = (1)r6c9
(4)r5c3 - (4=76)r25c1 - r6c1 = (6)r6c2
=>-1r6c2; ste

by **SpAce** » Sun Dec 08, 2019 7:08 pm

eleven wrote:
SpAce wrote:(8)r7c5&r8c9 =X= (87)r69c7 - r9c5 = (7)r8c6 => -7 r7c5,r8c9, (-8 r146c9); btte

[7]r9c5 = (7)r8c6 ? Do you remember (8)r7c5 ?

Obviously not!

Thanks! I had to reconstruct my coloring to see how I made such a weird mistake. The coloring was correct, and the 7r8c6 was correct, but due to a different route (now fixed). I think the mistake happened because I had both 7r9c7 and 7r8c6 colored (correctly) but only needed the latter, and then I looked too quickly how to get there (7r7c5 being grayed out as an elimination made it look like there was a strong link in box 8). Yet it's obvious that I didn't think very hard to let that slip by

by **Mauriès Robert** » Sun Dec 08, 2019 9:59 pm

Hi,
Resolution with TDP :
P(8r2c4) = {8r2c4, 8r7c5, 8r8c9, ...}
P(8r2c9) = {8r2c9, 8r7c8, 8r8c4, 5r9c4, 5r8c9, 8r6c7, 5r6c8, 6r6c1, 6r2c6, 7r8c6, ...}
=> (see Th2 TDP part2) : -8r146c9 (equivalent to X-wing 8r2/8r8), -7r7c5, -7r8c9 => solution by basic techniques.
Robert

by **SpAce** » Sun Dec 08, 2019 11:41 pm

Hi eleven,

eleven wrote:(58=4)b6p38 - r1c9 = 476r152 | [(4-6)r1c3 = r7c3 - r8c12 = r8c6 - r2c7 = r2c1] - (6=58)r6c18 => -8r6c79, stte

I like the logic, and it's relatively easy to read, but I don't think it's a valid AIC. As far as I see, the nested chain fragment doesn't work as a boolean argument.

It's similar to what I once suggested as a shorthand to write simple nets (because it's definitely convenient), but I was successfully convinced that it wasn't a great idea for this very reason. It looks too much like an AIC but isn't. A memory chain at least warns the reader that it's not a valid AIC, but this doesn't.

Alternatives:

AIC-Net (my favorite; also closest to yours):

Code: Select all: (4)r1c1 - (4=7)r5c1 ------------- (7)r2c1 || \\ (58=4)b6p83 - (4)r1c9 (6)r2c1 - (6=58)r6c18 => -8 r6c7, -58 r6c9 || // (4-6)r1c3 = r7c3 - r8c12 = r8c6 - (6)r2c6

Kraken 4R1:

Code: Select all: (4)r1c1 - (4=7)r5c1 - (7=6)r2c1 - (6=58)r6c18 || (4-6)r1c3 = r7c3 - r8c12 = r8c6 - r2c6 = r2c1 - (6=58)r6c18 || (4)r1c9 - (4=58)b6p38 => -8 r6c7, -58 r6c9

8x8 TM:

Code: Select all: 58r6c18 6r6c1 6r2c1 7r2c1 7r5c1 4r5c1 6r2c1 6r2c6 6r8c6 6r8c12 6r7c3 6r1c3 4r1c1 4r1c3 4r1c9 58b6p38 4r4c9 --------------------------------------------------- -8 r6c7 -58 r6c9

Alien ObiFish 12\14 (Rank 2):

{4R1 6R28 6C3 25N1 4N9 6N11888 \ 4667c1 6c6 4c9 6b7 1n3 [5588r6 58b6]} => -8 r6c7, -58 r6c9

Memory chain:

(58=6)r6c81 - r2c1 = (74*)r25c1&(6)r2c6 - r8c6 = r8c12 - r7c3 = r1c3 - (4)r1c3*1 = r1c9 - (4=58)b6p38 => -8 r6c7, -58 r6c9

Split-node AIC (my least favorite in this case):

(58=4)b6p83 - r1c9 = r1c13 - (46)r5c1,r1c3 = (7)r5c1|(6)r7c3 - (7)r2c1&(6)r8c12 = r2c1|r8c6 - r2c6 = r2c1 - (6=58)r6c18 => -8 r6c7, -58 r6c9

The New Sudoku Players' Forum

December 8, 2019

December 8, 2019

Re: December 8, 2019

Re: December 8, 2019

Re: December 8, 2019

Re: December 8, 2019

Re: December 8, 2019

Re: December 8, 2019