The New Sudoku Players' Forum

by **JC Van Hay** » Fri Feb 20, 2015 6:13 pm

Have a look at Steve's blog on the au site, in particular this page and the related ones.

by **bat999** » Fri Feb 20, 2015 7:40 pm

JC Van Hay wrote:Have a look at Steve's blog on the au site, in particular this page and the related ones.

That post is "Page 2 of Proof of Diabolic Puzzle of 02/20/07".
Why should I look at it?

by **JC Van Hay** » Fri Feb 20, 2015 8:15 pm

by **eleven** » Fri Feb 20, 2015 9:40 pm

bat999 wrote:What would be your thought process to find a move from my stuck position?

Normally when stuck, i try deadly patterns, x-chains, and ALS's as starting points, maybe almost xy- or w-wings, but i found nothing here.
If i really wanted to solve it nevertheless, i probably would try to find contradictions for bivalue/bilocation candidates (added: starting with those, where i already know, that i can come far in different directions), if not another idea comes, when sitting over it. But i don't want to invest so much time for a single puzzle anymore.

If you want to see, how to solve such hard puzzles manually without using the sledgehammer approach, you should read Steve's blog, who did it best in my eyes.

by **bat999** » Thu Feb 26, 2015 6:59 am

Hi
I found a move of my own.
(It's the same conclusion as Leren's but different method, I think)

Code: Select all: .------------------.------------------------.-----------------. | 1 *8-6 7 | 2689 4689 269 | 346 34 5 | | 4568 3 2 | 568 4568 7 | 1 9 *46 | | 9 *56 *46 | 3 *14 156 | 7 8 2 | :------------------+------------------------+-----------------: | 3 89 5 | 1489 2 189 | 49 6 7 | | 246 269 1 | 4569 7 569 | 8 45 3 | | 48 7 46 | 45689 3 5689 | 2 15 19 | :------------------+------------------------+-----------------: | *256 1 9 | 7 *56 4 | *36 23 8 | | *26 4 8 | 1269 *169 3 | 5 7 *169 | | 7 *256 3 | 125689 15689 125689 | *69 124 *146 | '------------------'------------------------'-----------------'

There's a problem if r1c2 is 6.

First...
(This puts a 1 in r8c5)
r1c2=6->r2c9=6->r89c9<>6
r1c2=6->r3c2=5->r9c2=2->r8c1=6->r7c1=5->r7c5=6->r7c7<>6
r89c9<>6,r7c7<>6->r9c7=6->r8c9=9
r8c1=6,r8c9=9->r8c5=1

Second...
(This puts a 1 in r3c5)
r1c2=6->r3c3=4->r3c5=1

That would put two 1's in column 5. :lol:

=> - 6 r1c2

---------------

After that there was some more work to do, but not too difficult. 8-)

: HoDoKu program._Extreme 3092).png (216.15 KiB) Viewed 977 times

by **daj95376** » Thu Feb 26, 2015 8:53 am

There is a reason why HoDoKu classified this puzzle as Extreme. It needs convoluted/network logic to crack it.

Code: Select all: +--------------------------------------------------------------------------------+ | 1 68 7 | 2689 4689 269 | 346 34 5 | | 4568 3 2 | 568 4568 7 | 1 9 46 | | 9 56 46 | 3 14 156 | 7 8 2 | |--------------------------+--------------------------+--------------------------| | 3 89 5 | 1489 2 189 | 49 6 7 | | 246 269 1 | 4569 7 569 | 8 45 3 | | 48 7 46 | 45689 3 5689 | 2 15 19 | |--------------------------+--------------------------+--------------------------| | 256 1 9 | 7 56 4 | 36 23 8 | | 26 4 8 | 1269 169 3 | 5 7 169 | | 7 256 3 | 125689 15689 125689 | 69 124 146 | +--------------------------------------------------------------------------------+ # 88 eliminations remain =5r7c5 =5r2c1 =8r6c1 =4r5c1 =4r6c4 =5r5c4 -> [r5c8]=EMPTY => -5 r7c5

Code: Select all: after Singles +-----------------------------------------------------------------------+ | 1 68 7 | 2689 489 269 | b46 3 5 | | 68-4 3 2 | 568 458 7 | 1 9 a46 | | 9 5 46 | 3 14 16 | 7 8 2 | |-----------------------+-----------------------+-----------------------| | 3 d89 5 | 1489 2 189 | c49 6 7 | | 246 269 1 | 4569 7 569 | 8 45 3 | | e48 7 46 | 45689 3 5689 | 2 15 19 | |-----------------------+-----------------------+-----------------------| | 5 1 9 | 7 6 4 | 3 2 8 | | 26 4 8 | 129 19 3 | 5 7 169 | | 7 26 3 | 12589 1589 12589 | 69 14 146 | +-----------------------------------------------------------------------+ # 69 eliminations remain XY-Chain: (4=6)r2c9 - (6=4)r1c7 - (4=9)r4c7 - (9=8)r4c2 - (8=4)r6c1 => -4 r2c1

_

by **JC Van Hay** » Thu Feb 26, 2015 11:52 am

daj95376 wrote:There is a reason why HoDoKu classified this puzzle as Extreme. It needs convoluted/network logic to crack it.

Code: Select all
+--------------------------------------------------------------------------------+ | 1 68 7 | 2689 4689 269 | 346 34 5 | | 4568 3 2 | 568 4568 7 | 1 9 46 | | 9 56 46 | 3 14 156 | 7 8 2 | |--------------------------+--------------------------+--------------------------| | 3 89 5 | 1489 2 189 | 49 6 7 | | 246 269 1 | 4569 7 569 | 8 45 3 | | 48 7 46 | 45689 3 5689 | 2 15 19 | |--------------------------+--------------------------+--------------------------| | 256 1 9 | 7 56 4 | 36 23 8 | | 26 4 8 | 1269 169 3 | 5 7 169 | | 7 256 3 | 125689 15689 125689 | 69 124 146 | +--------------------------------------------------------------------------------+ # 88 eliminations remain =5r7c5 =5r2c1 =8r6c1 =4r5c1 =4r6c4 =5r5c4 -> [r5c8]=EMPTY => -5 r7c5

This proof of the exclusion of 5r7c5 can be written as the following chain ...

Code: Select all: +--------------------+-------------------------+----------------+ | 1 68 7 | 2689 4689 269 | 346 34 5 | | 6(458) 3 2 | 68 4568 7 | 1 9 46 | | 9 56 46 | 3 14 156 | 7 8 2 | +--------------------+-------------------------+----------------+ | 3 89 5 | 1489 2 189 | 49 6 7 | | 26(4) 269 1 | 469(5) 7 569 | 8 (45) 3 | | (48) 7 6(4) | 689(45) 3 5689 | 2 15 19 | +--------------------+-------------------------+----------------+ | 26(5) 1 9 | 7 6-5 4 | 36 23 8 | | 26 4 8 | 1269 169 3 | 5 7 169 | | 7 26-5 3 | 12689(5) 15689 125689 | 69 124 146 | +--------------------+-------------------------+----------------+

[Skyscraper(5r72c1,5r29c4)=*M-Wing[(4=5)r3c8 - 5r5c4=*(5-4)r6c4=4r6c13] - 4r5c1=HP(48-5)r26c1=5r7c1]-5r9c2,5r7c5

The New Sudoku Players' Forum

[Solved]Suggest a move please.

Re: Suggest a move please.

Re: Suggest a move please.

Re: Suggest a move please.

Re: Suggest a move please.

Re: Suggest a move please.

Re: [Solved]Suggest a move please.

Re: [Solved]Suggest a move please.