View previous topic :: View next topic |
Author |
Message |
garytorborg
Joined: 19 Jan 2011 Posts: 28
|
Posted: Wed Jan 19, 2011 9:35 pm Post subject: stuck on fiendish puzzle |
|
|
I'm stuck in a fiendish puzzle on an android version of the game. I've solved other fiendish puzzles before without help but this one has got me stumped, and the only thing I can think of is that I am missing a candidate somewhere. The software will not give me hints (I don't want any anyway) and it tells me that I am "correct so far".
After basics, an xy-wing, an xyz-wing, and looking at the 2 w-wings (neither of which result in any eliminations), I am at this point:
Code: |
-------------------------------------------------------------------------
| 1369 5 2 | 36 69 8 | 7 4 136 |
| 4 679 38 | 3567 2 1 | 89 359 568 |
| 1369 679 138 | 4 5679 3569 | 2 359 13568 |
------------------------------------------------------------------------
| 5 1 9 | 8 3 7 | 4 6 2 |
| 7 2 4 | 56 1 56 | 3 8 9 |
| 8 3 6 | 9 4 2 | 5 1 7 |
------------------------------------------------------------------------
| 2 4 5 | 367 6789 369 | 1 379 38 |
| 1369 69 13 | 2 5789 359 | 89 3579 4 |
| 39 8 7 | 1 59 4 | 6 2 35 |
------------------------------------------------------------------------
|
Can anyone help? Or at least point out any errors in the pencilmarks?
Last edited by garytorborg on Wed Jan 19, 2011 10:09 pm; edited 1 time in total |
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Wed Jan 19, 2011 10:00 pm Post subject: |
|
|
As far as I can see you have "solved" it
The grid you posted has a solved 1 in block 9 but still a candidate in r7c8?
If you look at 7's in block 9 you should see a next move also... |
|
Back to top |
|
|
garytorborg
Joined: 19 Jan 2011 Posts: 28
|
Posted: Wed Jan 19, 2011 10:07 pm Post subject: |
|
|
Oops! The cell in question was mis-typed by me, and I didn't proof-read properly. The candidates in r7c8 should be: 379. Here's the corrected code:
Code: |
-------------------------------------------------------------------------
| 1369 5 2 | 36 69 8 | 7 4 136 |
| 4 679 38 | 3567 2 1 | 89 359 568 |
| 1369 679 138 | 4 5679 3569 | 2 359 13568 |
------------------------------------------------------------------------
| 5 1 9 | 8 3 7 | 4 6 2 |
| 7 2 4 | 56 1 56 | 3 8 9 |
| 8 3 6 | 9 4 2 | 5 1 7 |
------------------------------------------------------------------------
| 2 4 5 | 367 6789 369 | 1 379 38 |
| 1369 69 13 | 2 5789 359 | 89 3579 4 |
| 39 8 7 | 1 59 4 | 6 2 35 |
------------------------------------------------------------------------ |
Sorry for the typing. Hope I didn't miss any others. |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Thu Jan 20, 2011 12:16 am Post subject: |
|
|
Next, an X-wing on 9 in R19.
Keith |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Thu Jan 20, 2011 1:38 am Post subject: |
|
|
After the X-Wing, I don't see anything "pleasant". |
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Thu Jan 20, 2011 8:46 am Post subject: |
|
|
I am not sure this counts as "pleasant"! "Effective" perhaps is the word...
(Before the x-wing...)
Code: | Almost xy-wing(69-5) fin (3)r7c6 ; r8c6<>5
xy-wing(69-5)[(5=6)r5c6 - (6=9)r7c6 - (9=5)r9c5]
||
(3)r7c6 - (3=8)r7c9 - (8=9)r8c7 - r7c8=r7c56 - (9=5)r9c5 |
|
|
Back to top |
|
|
garytorborg
Joined: 19 Jan 2011 Posts: 28
|
Posted: Thu Jan 20, 2011 4:41 pm Post subject: |
|
|
peterj wrote: | I am not sure this counts as "pleasant"! "Effective" perhaps is the word...
(Before the x-wing...)
Code: | Almost xy-wing(69-5) fin (3)r7c6 ; r8c6<>5
xy-wing(69-5)[(5=6)r5c6 - (6=9)r7c6 - (9=5)r9c5]
||
(3)r7c6 - (3=8)r7c9 - (8=9)r8c7 - r7c8=r7c56 - (9=5)r9c5 |
|
PeterJ: I took care of the x-wing (God only knows how I missed that one) before seeing your response. I don't understand the notation. Could you explain further?
Thanks folks for finding the x-wing right under my nose... I have a tendency to miss them when they're in rows as opposed to columns. |
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Thu Jan 20, 2011 10:58 pm Post subject: |
|
|
OK - you asked for it!
So an "almost" pattern is a way of tackling harder puzzles but still basing it on searching for named patterns. The idea is to find something that would be a pattern e.g. a wing or fish, if it weren't for one (or more) candidates that get in the way.
Notation means..
One of the following statements must be true...
1) there is an xy-wing(69-5) pivoted on r7c6
2) there is a 3 in r7c6
(1) leads to r8c6<>5 as usual for an xy-wing
(2) results in a chain as follows:
if (3)r7c6 then (8)r7c9 and so (9)r8c7...
so not (9)r7c8 and must be a 9 in r7c5 or r7c6..
so as 9 in block 8 r9c5 must be a 5 - hence r8c6 can't be 5
Since both possibilities result in the same elimination we can remove 5 from r8c6. We don't know which initial proposition is true (in fact the xy-wing is true) but in either case the same elimination can be made. |
|
Back to top |
|
|
garytorborg
Joined: 19 Jan 2011 Posts: 28
|
Posted: Fri Jan 21, 2011 3:15 pm Post subject: |
|
|
Thank you, PeterJ!
I have never before used an "almost" pattern, as you describe there, but it worked perfectly. After the elimination, a quad appeared in row 8, eliminating no less than 2 more 9s, and then it came tumbling down. |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Fri Jan 21, 2011 5:27 pm Post subject: |
|
|
garytorborg wrote: | I have never before used an "almost" pattern ... |
Gary, an extra digit in an "almost" pattern is often called a "fin".
A finned X-wing is an almost X-wing.
Keith |
|
Back to top |
|
|
|