The New Sudoku Players' Forum

Website · by **denis_berthier** » Thu Sep 24, 2020 4:59 am

Code: Select all: +-------+-------+-------+ ! . . 2 ! . . . ! . . . ! ! . . . ! . . . ! . 9 . ! ! 9 . . ! . 6 1 ! . 7 5 ! +-------+-------+-------+ ! . . 3 ! 4 . . ! . . . ! ! . 5 . ! . . 9 ! 1 2 . ! ! 7 . . ! . 2 . ! . . . ! +-------+-------+-------+ ! . 1 . ! 5 . 7 ! . . 8 ! ! 6 . . ! . 9 . ! . . 7 ! ! . . . ! . . . ! . 3 4 ! +-------+-------+-------+ ..2.............9.9...61.75..34......5...912.7...2.....1.5.7..86...9...7.......34

by **Mathimagics** » Thu Sep 24, 2020 7:05 am

Ok, I'll take the bait ... what myth is being busted here?

Website · by **denis_berthier** » Thu Sep 24, 2020 7:31 am

Mathimagics wrote: what myth is being busted here?

Finding out is part of the fun.

by **Mathimagics** » Thu Sep 24, 2020 8:18 am

Ok, I guess I'll have to leave it to our resident puzzle-analyst freaks to figure out

by **SpAce** » Thu Sep 24, 2020 8:29 am

I have no idea what the myth is, but perhaps we'll find out.

Code: Select all: .------------------------.-------------------.-----------------. | 3458 34678 2 | 9 34578 3458 | 3468 48 1 | | 13458 34678 1567-48 | 378 34578 3458 | 3468 9 2 | | 9 348 a48 | 2 6 1 | 348 7 5 | :------------------------+-------------------+-----------------: | 128 2689 3 | 4 158 568 | 7 58 69 | | 48 5 6-48 | 3678 378 9 | 1 2 36 | | 7 689 169-8 | 1368 2 3568 | 48 458 369 | :------------------------+-------------------+-----------------: | 234 1 a4'9 | 5 34 7 | b2'9 6 8 | | 6 2348 a48'5 | 38 9 2348 | b2'5 1 7 | | 258 2789 579-8 | 168 18 268 | 59-2 3 4 | '------------------------'-------------------'-----------------'

Step 1. (9=48'5)r738c3 - (5=2'9) - loop => -48 r2569c3, -2 r9c7

Code: Select all: .-------------------.--------------------.------------. | 35-8 367 2 | 9 3578 c58 | 368 4 1 | | c135-8 367 b157 | 378 34578 458 | 368 9 2 | | 9 34 a48 | 2 6 1 | 38 7 5 | :-------------------+--------------------+------------: | 12 29 3 | 4 b15 c56 | 7 8 69 | | 4 5 6 | 78 78 9 | 1 2 3 | | 7 8 a19 | b16 2 3 | 4 5 69 | :-------------------+--------------------+------------: | 23 1 a49 | 5 34 7 | 29 6 8 | | 6 234 458 | 38 9 248 | 25 1 7 | | 258 279 579 | 16 18 268 | 59 3 4 | '-------------------'--------------------'------------'

Step 2. (8=491)r376c3 - (1)r2c3|(15)b5p72 = (1,58)r2c1,r41c6 => -8 r12c1; stte

by **Mauriès Robert** » Thu Sep 24, 2020 8:40 am

Perhaps a clue would put us on the right track!
I note that the puzzle has 28 backdoors.
Robert

Website · by **denis_berthier** » Thu Sep 24, 2020 10:09 am

Mauriès Robert wrote:Perhaps a clue would put us on the right track!
I note that the puzzle has 28 backdoors.
Robert

I expect everyone to use his favorite methods and see where that leads, that's all.

by **pjb** » Thu Sep 24, 2020 1:06 pm

1) Double ALS at r78c7 and r378c3, with X-Z values 5 and 9 => -2 r9c7, -48 r2c3, -48 r5c3, -8 r69c3 =>

Code: Select all: 358 367 2 | 9 3578 58 | 368 4 1 1358 367 157 | 378 34578 458 | 368 9 2 9 34 48 | 2 6 1 | 38 7 5 ------------------------+----------------------+--------------------- 12 29 3 | 4 15 56 | 7 8 69 4 5 6 | 78 78 9 | 1 2 3 7 8 19 | 16 2 3 | 4 5 69 ------------------------+----------------------+--------------------- 23 1 49 | 5 34 7 | 29 6 8 6 234 458 | 38 9 248 | 25 1 7 258 279 579 | 16 18 268 | 59 3 4

2) Kraken row:
(8)r1c1 - (8=5)r1c6 - (5=6)r4c6
(8-1)r2c1 = r4c1 - (1=5)r4c5 - (5=6)r4c6
(8)r9c1 - (8=1)r9c5 - (1=5)r4c5 - (5=6)r4c6 => -5 r4c6; stte

Phil

by **Cenoman** » Thu Sep 24, 2020 2:23 pm

pjb wrote:1) Double ALS at r78c7 and r378c3, with X-Z values 5 and 9 => -2 r9c7, -48 r2c3, -48 r5c3, -8 r69c3 =>
2) Kraken row:
(8)r1c1 - (8=5)r1c6 - (5=6)r4c6
(8-1)r2c1 = r4c1 - (1=5)r4c5 - (5=6)r4c6
(8)r9c1 - (8=1)r9c5 - (1=5)r4c5 - (5=6)r4c6 => -5 r4c6; stte

Hi Phil,

I was close to posting quite the same: Sue de Coq + Kraken.
Why do you add the node (5=6)r4c6 appending all three chains in your kraken, if 5r4c6 is your target ?
Kraken row:
(8)r1c1 - (8=5)r1c6
(8-1)r2c1 = r4c1 - (1=5)r4c5
(8)r9c1 - (8=1)r9c5 - (1=5)r4c5 => -5 r4c6
... that can be presented as an almost ALS W-Wing: [(5=8)r1c6 - r1c1 = r9c1 - (8=15)r49c5] = (8-1)r2c1 = r4c1 - (1=5)r4c5 => -5 r4c6

by **mith** » Thu Sep 24, 2020 2:47 pm

I feel personally attacked.

by **Cenoman** » Thu Sep 24, 2020 3:47 pm

One step:

Code: Select all: +---------------------------+------------------------+---------------------+ | 3458 34678 2 | 9 34578 3458 | 3468 48 1 | | 13458 34678 14678-5 | 378 34578 3458 | 3468 9 2 | | 9 348 48 | 2 6 1 | 348 7 5 | +---------------------------+------------------------+---------------------+ | 128 2689 3 | 4 158 568 | 7 58 69 | | 48 5 468 | 3678 378 9 | 1 2 36 | | 7 689 1689 | 1368 2 3568 | 48 458 369 | +---------------------------+------------------------+---------------------+ | 234 1 49 | 5 34 7 | 29 6 8 | | 6 2348 458 | 38 9 2348 | 25 1 7 | | 258 2789 789-5 | 168 18 268 | 259 3 4 | +---------------------------+------------------------+---------------------+

Multi-kraken (as a net)

Code: Select all: (3)r6c9 - (3=6)r5c9 - r5c3 = (617)r269c3 || (3)r6c4 - (3=81)b8p48 - r9c4 = r6c4 - r6c3 = (17)r29c3 || (3)r6c6 - (3)r1c6 || (5)r1c6-r12c5=(5-1)r4c5= r6c4 - r6c3 = (17)r29c3 || (48)r1c68 - (8)r1c1 || (8-1)r2c1 = (17)r29c3 || (8)r4c1 - (8=56)r4c68 - r9c6 = (6-1)r9c4 = r6c4 - r6c3 = (17)r29c3 || (8-4)r5c1 = r5c3 - (4=85)r38c3 || (8)r9c1 - (8=1)r9c5 - r9c4 = r6c4 - r6c3 = (17)r29c3 -------------------- -5 r29c3; ste

Matrix TM 16x16

Hidden Text: Show

Website · by **denis_berthier** » Thu Sep 24, 2020 4:28 pm

mith wrote:I feel personally attacked.

Hehe
mith and myth are different!

by **SpAce** » Thu Sep 24, 2020 5:01 pm

Cenoman wrote:One step: Multi-kraken

Nice!

Matrix TM 16x16

I love your new matrix layout! Very pretty and readable!

by **SpAce** » Thu Sep 24, 2020 5:03 pm

mith wrote:I feel personally attacked.

Henceforth, you shall be known as Darth... Myth 8-)

Website · by **denis_berthier** » Fri Sep 25, 2020 4:58 am

This puzzle is #2 of the controlled-bias collection (I just tweaked the floors). Its SER is 8.1 and its W rating is 5.

The unbiased means for minimal puzzles are:
- 4.73 for SER, with standard deviation 2.49
- 2.45 for W, with standard deviation 1.39
This means that this puzzle is:
- 1.35 standard deviations above the mean for SER
- 1.83 standard deviations above the mean for W
i.e. it is a relatively hard puzzle in unbiased stats.

But, of course, it is still extremely far from the hardest known puzzles. In particular, it is in T&E(1), so that no embedded chains, nets or OR branchings are needed.

Apart from this, it has absolutely nothing noticeable: it was just the first non trivial puzzle to appear in the collection. Random puzzles have nothing noticeable.

Starting after the first obvious steps:

Code: Select all: (solve-sukaku-grid +------------------------+-------------------+-----------------+ ! 3458 34678 2 ! 9 34578 3458 ! 3468 48 1 ! ! 13458 34678 156748 ! 378 34578 3458 ! 3468 9 2 ! ! 9 348 48 ! 2 6 1 ! 348 7 5 ! +------------------------+-------------------+-----------------+ ! 128 2689 3 ! 4 158 568 ! 7 58 69 ! ! 48 5 648 ! 3678 378 9 ! 1 2 36 ! ! 7 689 1698 ! 1368 2 3568 ! 48 458 369 ! +------------------------+-------------------+-----------------+ ! 234 1 49 ! 5 34 7 ! 29 6 8 ! ! 6 2348 485 ! 38 9 2348 ! 25 1 7 ! ! 258 2789 5798 ! 168 18 268 ! 592 3 4 ! +------------------------+-------------------+-----------------+ ) ,

here is one of the possible SudoRules solutions, where simplicity of each step is always preferred:

Code: Select all: *********************************************************************************************** *** SudoRules 20.1.s based on CSP-Rules 2.1.s, config = W+SFin *** Using CLIPS 6.32-r770 *********************************************************************************************** 164 candidates, 907 csp-links and 907 links. Density = 6.79% biv-chain[3]: r8c4{n3 n8} - r9c5{n8 n1} - b5n1{r4c5 r6c4} ==> r6c4 ≠ 3 biv-chain[4]: r7c7{n2 n9} - r7c3{n9 n4} - b8n4{r7c5 r8c6} - b8n2{r8c6 r9c6} ==> r9c7 ≠ 2 biv-chain[4]: r9c5{n8 n1} - b5n1{r4c5 r6c4} - c3n1{r6 r2} - c3n7{r2 r9} ==> r9c3 ≠ 8 whip-bn[4]: b8n4{r7c5 r8c6} - b8n2{r8c6 r9c6} - b7n2{r9c2 r8c2} - b7n3{r8c2 .} ==> r7c1 ≠ 4 biv-chain[5]: r4n2{c2 c1} - b4n1{r4c1 r6c3} - c4n1{r6 r9} - b8n6{r9c4 r9c6} - b8n2{r9c6 r8c6} ==> r8c2 ≠ 2 biv-chain-rn[3]: r8n2{c6 c7} - r7n2{c7 c1} - r7n3{c1 c5} ==> r8c6 ≠ 3 biv-chain[4]: r8n2{c6 c7} - r7n2{c7 c1} - r7n3{c1 c5} - b8n4{r7c5 r8c6} ==> r8c6 ≠ 8 t-whip[4]: b7n4{r8c3 r8c2} - r8n3{c2 c4} - r8n8{c4 c3} - r3c3{n8 .} ==> r5c3 ≠ 4 hidden-single-in-a-block ==> r5c1 = 4 t-whip[4]: c2n4{r3 r8} - r8n3{c2 c4} - r8n8{c4 c3} - r3c3{n8 .} ==> r2c3 ≠ 4 t-whip[4]: r3c3{n8 n4} - c2n4{r3 r8} - r8n3{c2 c4} - r8n8{c4 .} ==> r5c3 ≠ 8, r6c3 ≠ 8, r2c3 ≠ 8 singles ==> r5c3 = 6, r5c9 = 3, r6c6 = 3, r6c8 = 5, r4c8 = 8, r1c8 = 4, r6c7 = 4, r6c2 = 8, r2c2 ≠ 4 naked-pairs-in-a-column: c2{r3 r8}{n3 n4} ==> r2c2 ≠ 3, r1c2 ≠ 3 naked-pairs-in-a-block: b1{r1c2 r2c2}{n6 n7} ==> r2c3 ≠ 7 hidden-single-in-a-column ==> r9c3 = 7 hidden-pairs-in-a-block: b7{r8c3 r9c1}{n5 n8} ==> r9c1 ≠ 2, r8c3 ≠ 4 hidden-pairs-in-a-column: c4{n1 n6}{r6 r9} ==> r9c4 ≠ 8 finned-x-wing-in-columns: n3{c4 c2}{r8 r2} ==> r2c1 ≠ 3 biv-chain[3]: r4c5{n5 n1} - r6n1{c4 c3} - r2c3{n1 n5} ==> r2c5 ≠ 5 biv-chain[3]: r2n1{c1 c3} - c3n5{r2 r8} - c3n8{r8 r3} ==> r2c1 ≠ 8 naked-pairs-in-a-block: b1{r2c1 r2c3}{n1 n5} ==> r1c1 ≠ 5 whip[1]: r1n5{c6 .} ==> r2c6 ≠ 5 finned-x-wing-in-columns: n8{c1 c6}{r9 r1} ==> r1c5 ≠ 8 biv-chain[4]: r9c5{n8 n1} - r4n1{c5 c1} - c1n2{r4 r7} - r9n2{c2 c6} ==> r9c6 ≠ 8 whip[1]: c6n8{r2 .} ==> r2c4 ≠ 8, r2c5 ≠ 8 biv-chain[3]: r2c4{n7 n3} - b8n3{r8c4 r7c5} - c5n4{r7 r2} ==> r2c5 ≠ 7 naked-pairs-in-a-column: c5{r2 r7}{n3 n4} ==> r1c5 ≠ 3 whip[1]: b2n3{r2c5 .} ==> r2c7 ≠ 3 biv-chain-rn[4]: r7n4{c3 c5} - r2n4{c5 c6} - r2n8{c6 c7} - r3n8{c7 c3} ==> r3c3 ≠ 4 stte

The New Sudoku Players' Forum

myth-buster

myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster

Re: myth-buster