| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Wed Oct 20, 2010 2:14 am Post subject: Puzzle 10/10/20: B XY |
 |
|
| Code: | +-----------------------+
| . 8 . | 5 . 9 | 3 2 7 |
| 7 5 . | 2 . . | . . . |
| . . 9 | 3 . . | . . 4 |
|-------+-------+-------|
| 9 3 6 | 7 . 2 | 5 . . |
| . . . | . 3 . | . . 6 |
| 5 . . | 6 . 1 | . 7 3 |
|-------+-------+-------|
| 2 . . | 1 . . | . . . |
| 8 . . | . . 3 | . 6 . |
| 3 . 5 | . 2 6 | . . . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Wed Oct 20, 2010 6:32 am Post subject: |
|
|
| Code: | +-----------+--------------+----------------+
| 4 8 1 | 5 6 9 | 3 2 7 |
| 7 5 3 | 2 18 4 | 6 189 189 |
| 6 2 9 | 3 178 78 | 18 5 4 |
+-----------+--------------+----------------+
| 9 3 6 | 7 4 2 | 5 18 18 |
| 1 7 28 | 89 3 5 | 249 49 6 |
| 5 4 28 | 6 89 1 | 29 7 3 |
+-----------+--------------+----------------+
| 2 6 4 | 1 789 78 | 89 3 5 |
| 8 19 7 | 49 5 3 | 149 6 2 |
| 3 19 5 | 489 2 6 | 7 1489 189 |
+-----------+--------------+----------------+ |
(9=1)r9c2 - (1)r8c2 = (1)r8c7 - (1=8)r3c7 - (8=9)r7c7; r9c89 <> 9 |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Wed Oct 20, 2010 3:34 pm Post subject: |
|
|
This puzzle is an AUR heaven, but I could not find a quick solution using them. I may try again later.......
| Quote: | | w-wing (49)r5c8|r8c4 SL(4)r58c7; r5c4<>9 |
Ted |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Thu Oct 21, 2010 4:40 pm Post subject: |
|
|
| I eliminated the same two 9s as Norm, but that was because they were invalid as killers of the potential 18 DP. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Thu Oct 21, 2010 6:45 pm Post subject: |
|
|
Ted: I like the JC-style notation!
====== ===== ===== ===== ===== ===== ===== ===== ===== =====
After Ted mentioned all of the UR possibilities, I found a 6-cell DP in r249c89.
| Code: | +---------------------------------------------------------------+
| 4 8 1 | 5 6 9 | 3 2 7 |
| 7 5 3 | 2 18 4 | 6 *189 *189 |
| 6 2 9 | 3 178 78 | 18 5 4 |
|--------------------+--------------------+---------------------|
| 9 3 6 | 7 4 2 | 5 *18 *18 |
| 1 7 28 | 89 3 5 | 249 49 6 |
| 5 4 28 | 6 89 1 | 29 7 3 |
|--------------------+--------------------+---------------------|
| 2 6 4 | 1 789 78 | 89 3 5 |
| 8 19 7 | 49 5 3 | 149 6 2 |
| 3 19 5 | 489 2 6 | 7 *189+4 *189 |
+---------------------------------------------------------------+
# 35 eliminations remain
|
It was difficult convincing myself that this was a valid DP because I kept running into contradictions and multiple choices to resolve. Then, I realized that a DP should produce contradictions, like those in [b9] and [c7] for <8> below, and that I could use r9c2=19 to eliminate the multiple choices.
| Code: | r9c8<>4 and r2c8=9 r24c9=18 r9c9=9 (r9c2=1) r9c8=8 r4c8=1 r4c9=8 r2c9=1
=> r249c8=918 and r249c9=189
*--------------------------------------------------*
| 4 8 1 | 5 6 9 | 3 2 7 |
| 7 5 3 | 2 8 4 | 6 @9 1 |
| 6 2 9 | 3 178 78 | 8 5 4 |
|----------------+----------------+----------------|
| 9 3 6 | 7 4 2 | 5 1 8 |
| 1 7 28 | 89 3 5 | 249 4 6 |
| 5 4 28 | 6 89 1 | 29 7 3 |
|----------------+----------------+----------------|
| 2 6 4 | 1 789 78 | 8 3 5 |
| 8 9 7 | 49 5 3 | 14 6 2 |
| 3 1 5 | 48 2 6 | 7 8 9 |
*--------------------------------------------------*
|
| Code: | r9c8<>4 and r2c9=9 r24c8=18 r9c8=9 (r9c2=1) r9c9=8 r4c9=1 r4c8=8 r2c8=1
=> r249c8=189 and r249c9=918
*--------------------------------------------------*
| 4 8 1 | 5 6 9 | 3 2 7 |
| 7 5 3 | 2 8 4 | 6 1 @9 |
| 6 2 9 | 3 178 78 | 8 5 4 |
|----------------+----------------+----------------|
| 9 3 6 | 7 4 2 | 5 8 1 |
| 1 7 28 | 89 3 5 | 249 4 6 |
| 5 4 28 | 6 89 1 | 29 7 3 |
|----------------+----------------+----------------|
| 2 6 4 | 1 789 78 | 8 3 5 |
| 8 9 7 | 49 5 3 | 14 6 2 |
| 3 1 5 | 48 2 6 | 7 9 8 |
*--------------------------------------------------*
|
If r9c8<>4, then identical grids result except for the DP cells.
Now, I fully expect someone will burst my balloon on my 6-cell DP. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Thu Oct 21, 2010 10:56 pm Post subject: |
|
|
Danny,
I do not have the knowledge (or desire) to burst your balloon, but my understanding is that all of the SIS must result is a common condition before that conclusion is true. This puzzle only has internal SIS, but your analysis does not appear to satisfy that condition. What about the combination r9c8<>4 and r9c9=9? Plus, I do not appreciate the technique of using SIS in combination versus using them independently.
So what happens to a AUR that does not have external SIS? In this case, col8 & col9 do not contain either a digit 1 or a digit 8, but other houses do contain these digits.
Ted |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Thu Oct 21, 2010 11:46 pm Post subject: |
|
|
Danny,
So I just performed an external SIS analysis using row/box houses rather than the empty column houses, and each of these inferences forces r9c8=4.
However, I still do not have the bases to say if your analysis is sufficient and appropriate.
Ted |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Thu Oct 21, 2010 11:51 pm Post subject: |
|
|
Hi, guys.
Danny's pattern is a valid DP, to wit, a MUG.
In this thread RW calls this same pattern a "Layered Bug-Lite," but Myth Jellies suggests that MUG is more appropriate. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Fri Oct 22, 2010 12:28 am Post subject: |
|
|
| tlanglet wrote: | Danny,
I do not have the knowledge (or desire) to burst your balloon, but my understanding is that all of the SIS must result is a common condition before that conclusion is true. This puzzle only has internal SIS, but your analysis does not appear to satisfy that condition. What about the combination r9c8<>4 and r9c9=9? Plus, I do not appreciate the technique of using SIS in combination versus using them independently.
So what happens to a AUR that does not have external SIS? In this case, col8 & col9 do not contain either a digit 1 or a digit 8, but other houses do contain these digits.
|
Ted, I marginally understand your first paragraph, but I'm lost as to the second paragraph.
I should have specifically stated that the DP was for <189> in my six cells. That would have implied that r9c8=4 would have been forced true if the DP assumption was valid. Since there was a strong link on <9> in [r2], I jumped straight to demonstrating that the DP existed by forcing r9c8<>4 and deriving a grid for each possible <9> in [r2]. The results supported my assumption that a <189> DP existed in those six cells.
This isn't an elegant approach, but it always seems to work when I'm scratching my head and need to verify that a DP exists.
Regards, Danny |
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Fri Oct 22, 2010 11:43 am Post subject: |
|
|
I don't understand. Is the meaning with Sudoko to find as complicated solutions as possible? When I read the comments I got that impression.
Or is my simple bivalue-chain method something not accepted in the Sudoko world? Almost ever I need only one step. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Fri Oct 22, 2010 2:53 pm Post subject: |
|
|
| daj95376 wrote: | | I should have specifically stated that the DP was for <189> in my six cells. |
I was looking at a 6-cell AUR(18 ), which also results in r9c8=4. My concern about the external SIS being empty is explained by the fact that the pattern is a AMUG, not a AUR. 
Sorry for any chaos..........
Ted |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Sat Oct 23, 2010 5:53 am Post subject: |
|
|
| oaxen wrote: | | I don't understand. Is the meaning with Sudoko to find as complicated solutions as possible? When I read the comments I got that impression. |
Really. Exactly what is so complicated about a Type 1 DP?
| oaxen wrote: | | Or is my simple bivalue-chain method something not accepted in the Sudoko world? |
The MUG presented above solves the puzzle at a glance. Can you say that about your method?
Oaxen, everyone's happy that you get what you need from this game solving puzzles with brute force or whatever it should be called. There is absolutely nothing wrong with doing so.
However, after all the times you've broached this topic, it's curious you would still say "I don't understand" why many in the "Sudoku world" do not embrace your particular approach.
Folks understand and accept your approach. Sooner or later you should understand and accept that some do go out of their way to not solve puzzles the way you do. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Oct 23, 2010 6:52 am Post subject: |
|
|
| oaxen wrote: | I don't understand. Is the meaning with Sudoko to find as complicated solutions as possible? When I read the comments I got that impression.
Or is my simple bivalue-chain method something not accepted in the Sudoko world? Almost ever I need only one step. |
Your approach is probably used by a great many people. My aunt uses it when she runs out of basic techniques. She's happy and satisfied. However, she doesn't expect anyone else to care about how she reached the solution.
Your approach may find the solution, but it's useless in a discussion on logical processes/techniques that solve a puzzle. Therefore, no one in this forum is likely to be interested in your approach.
Regards, Danny A. Jones |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
