View previous topic :: View next topic |
Author |
Message |
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Thu Aug 04, 2011 2:26 pm Post subject: Puzzle 11/08/04: ~ Advanced |
|
|
Code: | +-----------------------+
| 8 7 . | . . 9 | . . 2 |
| 5 . . | . . . | . . . |
| . . . | . 7 4 | . . 9 |
|-------+-------+-------|
| . . . | . 9 6 | 5 . 1 |
| . . 1 | 5 3 . | . 4 6 |
| 6 . 5 | 4 . 1 | 9 . . |
|-------+-------+-------|
| . . . | 7 . 2 | . . . |
| . . . | . 1 . | . 2 7 |
| 7 . 2 | 9 4 . | . 1 5 |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
Back to top |
|
|
susume
Joined: 13 May 2011 Posts: 36 Location: Southeastern US
|
Posted: Sat Aug 06, 2011 8:08 pm Post subject: |
|
|
I had fun with this one.
Quote: | kite (8)r3c7 =r3c4 -r2c6 =r9c6; r9c7<>8
skyscraper (8) r34; r2c8<>8
XY wing (14)(48)(18) pivot r2c9; r1c4<>1 |
|
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Sat Aug 06, 2011 9:03 pm Post subject: |
|
|
susume wrote: | I had fun with this one.
|
I'm glad. I try to post puzzles with varying levels of difficulty for each rating category. I thought this one would get more attention. One just never knows! |
|
Back to top |
|
|
Marty R.
Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
|
Posted: Sun Aug 07, 2011 4:54 am Post subject: |
|
|
Same three moves as susume. |
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Sun Aug 07, 2011 4:02 pm Post subject: |
|
|
Marginally quicker with a pretty als-xz... (second today!)
Code: | skyscraper(8) r3,r4 ; r2c8<>8
als-xz (1=4)r1c7 - (4=361)r1c3|r3c32 ; r3c7<>1 |
Last edited by peterj on Sun Aug 07, 2011 8:10 pm; edited 1 time in total |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Sun Aug 07, 2011 4:56 pm Post subject: |
|
|
I'm attempting to add a restricted version of ANS(size=2/3/4) to my solver. So far, I'm still debugging all of the changes needed to implement the ANS(). However, my solver did detect an interesting W-Wing type pattern in this puzzle, and I decided to share.
Code: | after Peter's skyscraper in <8>
+-----------------------------------------------------------------------+
| 8 7 346 | 136 5 9 | 14 36 2 |
| 5 13469 3469 | 1368 2 38 | 7 36 48 |
| 2 136 36 | 1368 7 4 | 18 5 9 |
|-----------------------+-----------------------+-----------------------|
| 34 34 7 | 2 9 6 | 5 8 1 |
| 9 8 1 | 5 3 7 | 2 4 6 |
| 6 2 5 | 4 8 1 | 9 7 3 |
|-----------------------+-----------------------+-----------------------|
| 1 5 348 | 7 6 2 | 348 9 48 |
| 34 3469 34689 | 38 1 5 | 3468 2 7 |
| 7 36 2 | 9 4 38 | 368 1 5 |
+-----------------------------------------------------------------------+
# 49 eliminations remain
(36=1)r1c84 - r1c7 = r3c7 - (1=36)r3c23 => r1c3<>36
|
|
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Sun Aug 07, 2011 6:24 pm Post subject: |
|
|
Danny, nice. I like "pattern" moves that have one or more elements using an als - it seems a natural progression in solving techniques... |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Sun Aug 07, 2011 8:09 pm Post subject: |
|
|
Thanks Peter. Did you mean r3c7 for the als-xz elimination cell?
BTW: Your ALS pattern, (4=361)r1c3|r3c32, isn't one that my ANS() routine would detect. That's because I require every NS value to occur at least twice in the cells. In this case, <1> occurs once in the cells for NS values <361>.
If I get around to adding ALS patterns within cells to my solver, then my solver would create your chain as:
Code: | (1=4)r1c7 - (4=36)r13c3 - (36=1)r3c2 ; r3c7<>1
|
Last edited by daj95376 on Sun Aug 07, 2011 8:13 pm; edited 2 times in total |
|
Back to top |
|
|
peterj
Joined: 26 Mar 2010 Posts: 974 Location: London, UK
|
Posted: Sun Aug 07, 2011 8:11 pm Post subject: |
|
|
daj95376 wrote: | Thanks Peter. Did you mean r3c8 for the als-xz elimination cell? |
Sorry. I meant r3c7 - corrected above. |
|
Back to top |
|
|
|