| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Fri Aug 27, 2010 4:53 am Post subject: Puzzle 10/08/27: C |
 |
|
| Code: | +-----------------------+
| 8 . . | . . . | . . . |
| . 9 5 | . 3 1 | . . 2 |
| . 4 3 | . 5 . | . 1 . |
|-------+-------+-------|
| . . . | 7 1 9 | 4 2 . |
| . 2 1 | 4 . . | . 7 . |
| . 8 . | 6 . 5 | 1 . . |
|-------+-------+-------|
| . . . | 5 . 8 | 2 . . |
| . . 8 | 3 9 . | . 5 . |
| . 5 . | . . . | . . . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Fri Aug 27, 2010 7:49 am Post subject: |
|
|
Been doing a lot of these recently! This time grouped..
| Quote: | | grouped m-wing(67) (7=6)r2c1 - r3c1=(6-7)r3c6=r1c56 ; r1c3<>7 |
|
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Fri Aug 27, 2010 8:50 pm Post subject: |
|
|
I keep trying these ADPs hoping to get them right..............
1. BUG-Lite+3 (39)r169c789; SIS: r1c8=6, r9c1=3, r9c3=9;
(6)r1c8-r7c8=LS(39)r7c8|r9c7; r9c9<>39
||
(3)r9c1*-(3=9)r9c7; r9c9*<>39
||
(9)r9c3*-(9=3)r9c7; r9c9*<>39
This deletion was not very useful, so try, try again!
Use new SIS: r9c1=3, r1c4=9, r9c3=9
(3)r9c1-(3=9)r9c7; r1c7<>9
||
(9)r1c4; r1c7<>9
||
(9-2)r9c3=r1c3-(2=9)r1c4; r1c7<>9
This option is more effective except it does not delete 3 from r9c9; thus we impose both.
2. Type 6 UR (14)r78c19 with x-wing 1 overlay; r7c9,r8c1<>1
3. Type 6 UR (47)r79c59; SIS: r7c9=3 or r9c5=6
(4=7)r9c9-(7=6)r8c7-(6=3)r7c8-UR(47)r79c59[(3)r7c9=(6-4)r9c5]; r9c5<>4
4. AUR (67)r19c56; SIS: r1c8=6, r3c6=7, r9c9=7
(6)r1c8-(6=7)r2c7; r3c9<>7
||
(7)r3c6; r3c9<>7
||
(7)r9c9; r3c9<>7
Definitely not the most efficient solution but great fun.....
Ted |
|
| Back to top |
|
|
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Sat Aug 28, 2010 7:05 am Post subject: |
|
|
Equivalent to Peter's proof
| Quote: | | ALS XZ-rule : (7=6)R2C7-(6=3927)R1C3478 : =>r2c1<>7 |
|
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Aug 28, 2010 2:40 pm Post subject: |
|
|
| JC Van Hay wrote: | Equivalent to Peter's proof
| Quote: | | ALS XZ-rule : (7=6)R2C7-(6=3927)R1C3478 : =>r2c1<>7 |
|
JC, don't follow your logic. Please provide some more detail.
Ted |
|
| Back to top |
|
|
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Sat Aug 28, 2010 3:32 pm Post subject: |
|
|
ALS XZ-rule : (7=6)R2C7-(6=3927)R1C3478 : =>r2c1<>7
The cell R2C7 contains an Almost Single {67}, while the cells R1C3478 contain an Almost Quad {23679} [see (27)R1C3, (29)R1C4, (39)R1C7 and (369)R1C8].
x=6 is a Restricted Common candidate : it is either in R1C8 or in R2C7, as these 2 cells are in the same box. Therefore, either R1C7 contains the single, {7}, or the cells R1C3478 contains the quad {2379}.
The Common candidate z=7 of those 2 Locked Subsets are located in R2C8 and in R1C3. They form a derived strong set, as (7)r1c7=(7)r1c3 : => r2c1<>7.
Note : 6=3927 suggests the succesive assignments NP(39), 2 and 7 ... JC |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Sat Aug 28, 2010 4:56 pm Post subject: |
|
|
tlanglet, you're definitely making good progress with the uniqueness techniques. I've only two minor points to add. First, the 'N' of BUG-Lite+N and BUG+N refers to the number of cells with extra candidates, not the number of candidates. Second, the AUR chain ...
| you wrote: | 3. Type 6 UR (47)r79c59; SIS: r7c9=3 or r9c5=6
(4=7)r9c9-(7=6)r8c7-(6=3)r7c8-UR(47)r79c59[(3)r7c9=(6-4)r9c5]; r9c5<>4 |
... can be shortened to ... (4)r9c9 = aur(47)r79c59:[(4-3)r7c9 = (6)r9c5] |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Aug 28, 2010 10:22 pm Post subject: |
|
|
| JC Van Hay wrote: | ALS XZ-rule : (7=6)R2C7-(6=3927)R1C3478 : =>r2c1<7> r2c1<>7.
Note : 6=3927 suggests the succesive assignments NP(39), 2 and 7 ...[/list]JC |
JC,
Here is my code after basics:
| Code: |
*--------------------------------------------------------------------*
| 8 1 27 | 29 467 467 | 39 369 5 |
| 67 9 5 | 8 3 1 | 67 4 2 |
| 267 4 3 | 29 5 67 | 8 1 79 |
|----------------------+----------------------+----------------------|
| 5 3 6 | 7 1 9 | 4 2 8 |
| 9 2 1 | 4 8 3 | 5 7 6 |
| 47 8 47 | 6 2 5 | 1 39 39 |
|----------------------+----------------------+----------------------|
| 134 67 49 | 5 47 8 | 2 369 13479 |
| 124 67 8 | 3 9 24 | 67 5 14 |
| 234 5 249 | 1 467 2467 | 39 8 3479 |
*--------------------------------------------------------------------*
|
We seem to have different code after basics.
I have checked my results twice.
Ted |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Aug 28, 2010 10:41 pm Post subject: |
|
|
Ted, JC's ALS broken down into smaller components:
| Code: | (7=6)r2c7 - ANP(6=39)r1c78 - (9=2)r1c4 - (2=7)r1c3 => r2c1<>7
|
When the logic in the last four cells are merged (in my notation and ordering):
| Code: | (7=6)r2c7 - ALS(6=392=7)r1c7843 => r2c1<>7
|
|
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Aug 28, 2010 10:55 pm Post subject: |
|
|
| ronk wrote: | tlanglet, you're definitely making good progress with the uniqueness techniques. I've only two minor points to add. First, the 'N' of BUG-Lite+N and BUG+N refers to the number of cells with extra candidates, not the number of candidates. Second, the AUR chain ...
| you wrote: | 3. Type 6 UR (47)r79c59; SIS: r7c9=3 or r9c5=6
(4=7)r9c9-(7=6)r8c7-(6=3)r7c8-UR(47)r79c59[(3)r7c9=(6-4)r9c5]; r9c5<>4 |
... can be shortened to ... (4)r9c9 = aur(47)r79c59:[(4-3)r7c9 = (6)r9c5] |
Ron, Thank you sir!
Point 1: Yes, I counted the number of SIS. I understand the proper notation for BUGs and should have understood the same meaning applies for BUG-Lite patterns. I get to focused and loose perspective.
Point 2: I am usually just happy to find a path. With experience I would hope to seek a cleaner path.
Ted |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Aug 28, 2010 11:02 pm Post subject: |
|
|
Personally, I like this alternative using a hidden pair to force a grouped strong link:
| Code: | (7=6)r2c7 - r1c8 = HP(46-7)r1c56 = (7)r1c3 => r2c1<>7
|
JC: Thanks for providing an ALS that I could have so much fun! |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sun Aug 29, 2010 12:56 am Post subject: |
|
|
The hit with a 2x4 worked! 
Ted |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Aug 29, 2010 5:17 am Post subject: |
|
|
| W-Wing on 67 in boxes 12, with strong link 7 in r1; r3c1<>6. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
