| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Tue May 18, 2010 4:49 pm Post subject: Puzzle 10/05/18: (B) XY |
 |
|
After several steps, my final grid resembles a BUG+2 scenario -- but it isn't. The np() folks should have fun skirting the use of an XY-Chain.
| Code: | +-----------------------+
| 5 7 . | . . 3 | . 4 . |
| 1 . 9 | . 4 8 | 7 . . |
| . 4 . | . 9 . | 3 . . |
|-------+-------+-------|
| . . . | 6 2 . | . . 4 |
| . 5 4 | 3 . . | . 6 9 |
| 6 9 . | . . 7 | . . . |
|-------+-------+-------|
| . 8 7 | . . . | . . . |
| 3 . . | . 7 . | . 9 . |
| . . . | 8 3 . | . . . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Wed May 19, 2010 1:03 am Post subject: |
|
|
Two steps did it.
| Quote: | XYZ-Wing (235)
XY-Wing (125), flightless with pincer transport |
|
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed May 19, 2010 6:11 am Post subject: |
|
|
General Note: This puzzle is like many in my current file of puzzles. It ends with a limited grid that my solver couldn't solve w/o an XY-Chain, M-Wing, and/or W-Wing. So expect to see more puzzles with the "XY" designation in the title.
Addendum: Many of the puzzles in this set are BBDB with difficult steps. Let me know if you find them excessive. |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Wed May 19, 2010 8:19 am Post subject: |
|
|
I like xy-chains .. but saw this m-wing one stepper instead
[withdrawn!]
Last edited by peterj on Wed May 19, 2010 6:24 pm; edited 2 times in total |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Wed May 19, 2010 11:41 am Post subject: |
|
|
Oh alright Danny, I'll play. 
This ANP 125 in r29c8 sets off 2 streams depending on:
1.ANP (25=1)r9c8-(1=5)r9c3-r8c3=r8c4-(5=2)r2c4-r2c8 so r2c8 <> 2
2.ANP (1=25) - [XWingon1@r36c89]-(1=8)r6c7-r8c7=r8c9-(8=2)r1c9-r2c8 so r2c8 <> 2
The second assumes the NP is true and so an xwing on 1s is formed. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Wed May 19, 2010 1:24 pm Post subject: |
|
|
Taking them in order that I found them, I used four steps. I am not sure how to classify the last step since it starts as a "Almost naked triple" (ANT), but ends more like a conflicting chain. Comments would be appreciated.
ANP(25=3)r27c8 - (3=5)r7c9 ; r9c8<>5,
xy-wing 1-25 with vertex 15 in r3c9; r3c8<>1,
xy-wing 12-6 with vertex 12 in r9c8; r9c7<>6,
ANT(146=8)r789c7 - (8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7; r8c7<>8
Ted |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Wed May 19, 2010 4:43 pm Post subject: |
|
|
Speaking as someone else who likes to pop the hood and have a look, It looks fine to me Ted.
What you have is a classic contradiction - similar to some of the others I had a little while back.
Your hypothesis shows that where 8 is true in the ANT it is also true in the only other 8 in column 7 which is outside the ANT so therefore ANT=8 is FALSE and can be removed. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed May 19, 2010 5:27 pm Post subject: |
|
|
[Withdrawn: followed peterj's example.]
Last edited by daj95376 on Wed May 19, 2010 7:44 pm; edited 1 time in total |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Wed May 19, 2010 6:23 pm Post subject: |
|
|
When I say "I saw an m-wing" I obviously meant "I dreamt it with the help of a bottle of Pouilly-Fuisse"... 
Sometime highlighting bivalues in SS distracts you from what's in the other cells..
Thanks for your vigilance! |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Wed May 19, 2010 6:39 pm Post subject: |
|
|
| tlanglet wrote: | Taking them in order that I found them, I used four steps. I am not sure how to classify the last step since it starts as a "Almost naked triple" (ANT), but ends more like a conflicting chain. Comments would be appreciated.
ANT(146=8)r789c7 - (8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7; r8c7<>8
Ted
|
'Sup, Ted?
One very wee point. I think your AIC proves r6c7<>1, and that the (8)r8c7 is just collateral damage.
What do you think of this minor rewriting? Only the last node is different.
| Code: | *--------------------------------------------------*
| 5 7 68 | 12 16 3 | 9 4 28 |
| 1 3 9 | 25 4 8 | 7 25 6 |
| 2 4 68 | 7 9 56 | 3 58 1 |
|----------------+----------------+----------------|
| 8 1 3 | 6 2 9 | 5 7 4 |
| 7 5 4 | 3 8 1 | 2 6 9 |
| 6 9 2 | 4 5 7 | 18 18 3 |
|----------------+----------------+----------------|
| 4 8 7 | 9 16 2 | 16 3 5 |
| 3 26 15 | 15 7 46 | 468 9 28 |
| 9 26 15 | 8 3 456 | 14 12 7 |
*--------------------------------------------------*
|
(146=8)als:r789c7 - (8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - (8=1)r6c8 =>r6c7<>1.
In an AIC, one endpoint must be true. Armed with that, check to see if there's a contradiction. The rewritten AIC starts with a locked set (146)r789c7 and ends with (1)r6c8. One must be true, so the elim is r6c7<>1.
If you want to end on (8)r6c7, the als is not needed at all 
(8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7; r8c7<>8 |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Wed May 19, 2010 8:32 pm Post subject: |
|
|
| Quote: | | If you want to end on (8)r6c7, the als is not needed at all |
I got the sense that Ted was looking at the ANT as the deductive point here which I guess is why it was selected to see what implications could be followed. Notational accuracy or not, there are plenty of sudoku structures eg x-cycles/Baphomet that provide a contradiction that eliminates the original hypothesis.
What is incontrovertible here (to me at least !!) from a logic standpoint is:
(1)that If 8 is in ANT then following Ted's chain there is also an 8 in the same column brought about by the strong (8s) link between r6c8 and r6c7. Contradiction proves original hypothesis incorrect.
(2)If the 8 is not in ANT then r6c7 still = 8.
Be gentle with me.  |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Wed May 19, 2010 9:03 pm Post subject: |
|
|
| Luke451 wrote: | If you want to end on (8)r6c7, the als is not needed at all
(8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7; r8c7<>8 |
That certainly cuts to the chase! But why use six strong links when nine native strong sets will do the job? And why throw away a perfectly good ANT?  |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed May 19, 2010 11:34 pm Post subject: |
|
|
I understand Ted's desire to use the ANT. I understand Luke catching that not all of the Is were dotted and Ts crossed in Ted's chain. I understand that Mogulmeister is perfectly happy w/o all the Is and Ts. As always, ronk's depth of perception exceeds mine and I'm lost as to his observation.
I'm going to make Ted's chain work with the ANT ... by turning it into a discontinuous loop.
(146=8)r789c7 - (8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7 - (8=146)r789c7
At this point, you may either conclude that r789c7=146 or r8c7<>8. As Luke mentioned, the green part of the discontinuous loop is sufficient for the conclusion r8c7<>8. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Thu May 20, 2010 12:12 am Post subject: |
|
|
I thank one and all for the informative comments and additional oration.
Luke: I understand your point about an AIC having one end point that must be true.
Mogulmeister: You are correct. I spotted the ANT and used it as the starting point for my chain, which then turned into a contradiction event.
Ronk: You are also correct. I did not want to spoil a good ANT, especially since it was the first one I ever found 
Danny: Thanks for making my ANT work
Ted |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Thu May 20, 2010 6:01 am Post subject: |
|
|
Thanks one and all. (146=8)r789c7 - (8=2)r8c9 - (2=6)r8c2 - r9c2 = r9c6 - (6=5)r3c6 - (5=8)r3c8 - r6c8 = (8)r6c7 - (8=146)r789c7
Peter - it seems we are rapidly going up the Burgundy quality ladder with your Pouilly-Fuissé! You're clearly a man of discernment and next time I make a mistake I feel it would only be proper that I have a glass of La Tāche in my hand! |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Thu May 20, 2010 3:16 pm Post subject: |
|
|
| Mogulmeister wrote: | | Peter - it seems we are rapidly going up the Burgundy quality ladder with your Pouilly-Fuissé! You're clearly a man of discernment and next time I make a mistake I feel it would only be proper that I have a glass of La Tāche in my hand! |
When I was younger and could occasionally afford to treat myself to a bottle of wine that didn't hurt my stomach, I use to have a glass of TBA Spatlese as dessert after dinner. Probably too crude for you folks. |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Thu May 20, 2010 3:38 pm Post subject: |
|
|
Not at all.
I would not profess great expertise but I had the good fortune to have some Riesling TBA while in Germany and it was exquisite but pricey. |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Thu May 20, 2010 4:30 pm Post subject: |
|
|
I met a girl in Murphy, California last summer. She was wearing a black tee shirt with one word sewn on the front, in sequins no less. I read the word, and it was love at first sight.
The word?
Gotta like a girl like that!  |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Thu May 20, 2010 4:33 pm Post subject: |
|
|
 |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Thu May 20, 2010 4:56 pm Post subject: |
|
|
Romanee Conti for me please!
Sadly way way out of my price bracket |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
