View previous topic :: View next topic |
Author |
Message |
lkmckin
Joined: 25 May 2007 Posts: 9
|
Posted: Fri May 25, 2007 9:38 pm Post subject: LA times diabolical |
|
|
Hi, new here. I've been trying all the VH puzzles on the site, and have gotten pretty good at X-wing, XY-wing, swordfish, etc.
But todays LA Times puzzle is a real stumper. For starters, the grade given to it by this site's "play on line" function is TOO HARD, even after inputting the puzzle raw. I've worked up to the following, still rated TOO HARD. Can one of the real experts give me a clue as to how to proceed (other than trial and error, I hate that).
Thanks in advance.
Lindsay
Code: |
+-------------+-------------+----------------+
| 68 379 2 | 3456 1 349 | 4578 3458 4589 |
| 4 1379 137 | 35 8 39 | 57 2 6 |
| 68 39 5 | 2346 7 2349 | 1 348 489 |
+-------------+-------------+----------------+
| 9 47 478 | 1 3 5 | 6 48 2 |
| 35 6 134 | 9 2 8 | 45 7 145 |
| 2 15 18 | 7 4 6 | 589 1589 3 |
+-------------+-------------+----------------+
| 7 2 9 | 48 5 1 | 3 6 48 |
| 1 8 34 | 234 6 234 | 59 59 7 |
| 35 345 6 | 348 9 7 | 2 148 148 |
+-------------+-------------+----------------+
|
|
|
Back to top |
|
|
Nwohio
Joined: 12 May 2007 Posts: 9
|
Posted: Fri May 25, 2007 10:32 pm Post subject: |
|
|
I'm no expert, but take a look at r6c7c8 and r8c7c8. |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Fri May 25, 2007 11:02 pm Post subject: |
|
|
Look for a unique rectangle on <59> I'm not mad about 'em but it sure works. |
|
Back to top |
|
|
Ruud
Joined: 18 Jan 2006 Posts: 31
|
Posted: Fri May 25, 2007 11:16 pm Post subject: |
|
|
If you do not fancy unique rectangles (who doesn't?), there is a nice 2SDG (Sue de Coq) alternative:
r56c8 has candidates {4,5,8,9}
because r4c8 = {4,8}, r56c8 must contain either 5 or 9.
because r8c7 = {5,9}, r56c8 must contain either 4 or 8.
so: r56c8 contains {5 or 9} + {4 or 8}
as a result, you have a naked pair {4,8} in box 6, and a naked pair {5,9} in column 7.
Ruud |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Fri May 25, 2007 11:20 pm Post subject: |
|
|
Nice one Ruud - do you see lots of these ? It's a new one on me - looks like a multi-candidates ALS.
Quote: | r56c8 has candidates {4,5,8,9} |
I assume you mean r56c7 ?? |
|
Back to top |
|
|
lkmckin
Joined: 25 May 2007 Posts: 9
|
Posted: Tue May 29, 2007 9:09 pm Post subject: |
|
|
Very interesting. Kind of an bifurcation of naked pairs. Where does the term 2SDG(Q?) Sue de Coq come from?
Of note is that the puzzle grade goes from "too hard" to "easy" after the application of this logic.
Thanks, Ruud (and Mogulmeister without whom I never would have understood). |
|
Back to top |
|
|
Ruud
Joined: 18 Jan 2006 Posts: 31
|
Posted: Tue May 29, 2007 10:11 pm Post subject: |
|
|
Thanks Mogulmeister for the correction.
I found another one in the recent Andrew Stuart puzzle, but that does not really advance the puzzle. These patterns occur frequently in Sudoku puzzles, but because of their complexity, they are not easy to spot.
2SDG means 2-Sector Disjoint Groups/Subsets, with Disjoint Group/Subset being the early name for naked/hidden subset. The technique was found in 2005 by Rubylips, one of the early sudoku programmers. At that time, he was using the nickname Sue De Coq on the Pappocom forum and the technique was soon named after its inventor.
There is another example in Sudopedia, and there is an interesting discussion on another forum.
Ruud |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Wed May 30, 2007 4:49 am Post subject: |
|
|
Cheers Ruud - always nice to continue learning! |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|