| View previous topic :: View next topic |
| Author |
Message |
ronk
Joined: 07 May 2006
Posts: 398
|
 Posted: Thu Mar 17, 2011 3:48 am Post subject: |
 |
|
| keith wrote: | | I was writing that in response to the idea that the 4s in R7C12 imply that R7C7 has not been solved. |
 Reminds me of a guy whose debating style went ..
On the one hand ...
On the other hand ...
On the other other hand ... |
|
| Back to top |
|
 |
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Thu Mar 17, 2011 6:04 am Post subject: |
|
|
| ronk wrote: | | keith wrote: | | I was writing that in response to the idea that the 4s in R7C12 imply that R7C7 has not been solved. |
Reminds me of a guy whose debating style went ..
On the one hand ...
On the other hand ...
On the other other hand ... |
Reminds me of a guy who couldn't get the issues straight.
On the one hand, | Code: | +--------------+--------------+--------------+
| 46 1246 9 | 7 14 5 | 8 24 3 |
| 47 147 3 | 8 149 2 | 59 45 6 |
| 5 24 8 | 49 3 6 | 7 1 249 |
+--------------+--------------+--------------+
| 278 378 4 | 1259 59 19 | 56 3567 57 |
| 1 9 5 | 6 7 34 | 2 34 8 |
| 27 37 6 | 25 8 34 | 1 9 457 |
+--------------+--------------+--------------+
| 4689 4568 17 | 3 2 19 | 4 5678 579 |
| 4689 4568 12 | 1459 4569 7 | 3 2568 259 |
| 3 456 27 | 459 4569 8 | 569 2567 1 |
+--------------+--------------+--------------+ |
Do the 4s in R7C12 invalidate the 4 in R7C7? (My answer is no. It follows from the uniqueness assumption.)
On the other hand, given a partially solved puzzle where there is no distinction between initial cells and solved cells, how do you apply uniqueness techniques that involve a DP (or an ADP, depending on your semantics) that includes a single-candidate cell? (My answer is you cannot.)
http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=4429
Those two hands seem very different to me.
On the third hand, play high.
Keith |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Thu Mar 17, 2011 12:10 pm Post subject: |
|
|
| keith wrote: | On the one hand, | Code: | +--------------+--------------+--------------+
| 46 1246 9 | 7 14 5 | 8 24 3 |
| 47 147 3 | 8 149 2 | 59 45 6 |
| 5 24 8 | 49 3 6 | 7 1 249 |
+--------------+--------------+--------------+
| 278 378 4 | 1259 59 19 | 56 3567 57 |
| 1 9 5 | 6 7 34 | 2 34 8 |
| 27 37 6 | 25 8 34 | 1 9 457 |
+--------------+--------------+--------------+
| 4689 4568 17 | 3 2 19 | 4 5678 579 |
| 4689 4568 12 | 1459 4569 7 | 3 2568 259 |
| 3 456 27 | 459 4569 8 | 569 2567 1 |
+--------------+--------------+--------------+ |
Do the 4s in R7C12 invalidate the 4 in R7C7? (My answer is no. ...) |
Agreed, the correct answer is "no". R7c7 must be an initial clue because, if not, the puzzle has multiple solutions, as daj95376 and you noted earlier. To reach this conclusion, the candidates in unsolved cells are irrelevant. Likewise, whether or not the original puzzle is partially solved is irrelevant.
If required, assessing the correctness of posted candidates should chronologically follow.
Since the puzzle has a unique solution, uniqueness techniques may be validly used. Whether or not the original puzzle is partially solved is irrelevant.
Hence, IMO the "other hands" are moot ... and mute. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Fri Mar 18, 2011 1:24 am Post subject: |
|
|
Here is a partially solved puzzle (A):
| Code: | +-------------+-------------+-------------+
| 4 6 7 | 2 5 1 | 9 8 3 |
| 5 3 8 | 9 467 467 | 267 17 12 |
| 9 2 1 | 3 67 8 | 67 4 5 |
+-------------+-------------+-------------+
| 3 5 6 | 8 19 47 | 27 179 124 |
| 2 17 49 | 6 3 5 | 8 179 14 |
| 8 17 49 | 47 19 2 | 3 5 6 |
+-------------+-------------+-------------+
| 1 9 3 | 47 467 467 | 5 2 8 |
| 7 8 5 | 1 2 3 | 4 6 9 |
| 6 4 2 | 5 8 9 | 1 3 7 |
+-------------+-------------+-------------+ |
Note the 78 potential deadly pattern in R68C12. R6C2 must be 1, which solves the puzzle!
Keith
Last edited by keith on Fri Mar 18, 2011 2:09 am; edited 1 time in total |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Fri Mar 18, 2011 2:02 am Post subject: |
|
|
Here is a partially solved puzzle (B): | Code: | +----------------+----------------+----------------+
| 7 34 1 | 258 6 9 | 258 38 234 |
| 6 9 35 | 2578 25 478 | 258 1 234 |
| 45 8 2 | 45 3 1 | 7 6 9 |
+----------------+----------------+----------------+
| 9 6 7 | 1 4 5 | 3 2 8 |
| 58 2 58 | 9 7 3 | 6 4 1 |
| 3 1 4 | 6 8 2 | 9 7 5 |
+----------------+----------------+----------------+
| 28 347 6 | 378 9 478 | 1 5 237 |
| 1248 5 38 | 347 12 6 | 28 9 237 |
| 12 37 9 | 25 125 78 | 4 38 6 |
+----------------+----------------+----------------+ |
Note the potential deadly pattern 14 in R16C23. Which says that R1C2 must be 3, which is an invalid solution.
??
Keith |
|
| Back to top |
|
|
Pat
Joined: 23 Feb 2010
Posts: 207
|
| Posted: Tue Mar 22, 2011 11:18 am Post subject: |
|
|
| keith wrote: | partially solved puzzle (A)---
Note the 78 potential deadly pattern in R68C12.
R6C2 must be 1, which solves the puzzle!
|
in this puzzle,
it just so happens that r6c2 = 1
however,
your "proof" assumes that all 3 known cells
are solved cells ( none given )
what basis do you have for this assumption?
| keith wrote: | partially solved puzzle (B)---
Note the potential deadly pattern 14 in R16C23.
Which says that R1C2 must be 3, which is an invalid solution.
??
|
- if all 3 known cells
are solved cells ( none given )
then this puzzle has an even number of answers
assuming the puzzle has 1 answer
we conclude that r1c2 = 3
and thus discard all the answers,
creating the impression of a puzzle with 0 answers
- but if any of those 3 known cells was given,
then the puzzle has 1 answer ( r1c2 = 4 )
sorry about jumping in like this,
i may be off-topic
~ Pat |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Mar 22, 2011 3:05 pm Post subject: |
|
|
Pat,
I believe that Keith was posting examples in response to:
| ronk wrote: | Since the puzzle has a unique solution, uniqueness techniques may be validly used. Whether or not the original puzzle is partially solved is irrelevant.
|
If Ron had written: ... may be validly used in cells known not to be clues/givens.
... then I doubt if Keith would have posted his examples. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Wed Mar 23, 2011 1:12 am Post subject: |
|
|
Pat,
Thank you for the reply. I agree, your reaction is "What??"
I was surprised in this thread that I ran afoul of both Danny and Ron. I was simply trying to make two points:
1. If you post a partially solved puzzle, please distinguish the the initial given cells from the single-digit solved cells.
1a. The candidates you specify in the unsolved multi-candidate cells may indicate some of your deductions, but they have nothing to do with defining the puzzle.
2. In any puzzle, a single-digit cell eliminates that candidate in any of its peers.
2a. Any candidate 123456789 is a starting possibility in any unsolved cell, until we start to use the initial cells, and then logic, to eliminate candidates.
WOW! was I surprised that these points are controversial, especially with cognoscenti like Ron and Danny.
After Ron posted his message that he was mute (synonym = dumb), I posted my message.
Danny's last post is correct as to my motives.
Puzzle A is in this thread:
http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=4429
It was posted by Wapati, and the partially solved state is his. I will agree that this kind of state is unusual, but it is not an unrealistic situation manufactured by me.
The uniqueness deduction for Puzzle A is valid, for it does not involve initial values in given cells.
Puzzle B is the Free Press puzzle of Feb 25 after basics. I did not have to look far to find this example. The uniqueness deduction for Puzzle B involves an initial value, and is invalid.
http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=5548
Again, my simple plea is: If you must post a partially solved puzzle, please include the original puzzle in your post. It makes a difference in the subsequent solution possibilities.
It will also allow helpers to point out if there are errors in your posted partial solution.
Keith |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Wed Mar 23, 2011 1:22 am Post subject: |
|
|
| Quote: | | ...especially with cognoscenti like Ron and Danny. |
Don't be modest; anyone who uses the word is one of them.  |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Wed Mar 23, 2011 1:28 am Post subject: |
|
|
| Marty R. wrote: | | Quote: | | ...especially with cognoscenti like Ron and Danny. |
Don't be modest; anyone who uses the word is one of them. |
The Godfather's people will be visiting ...  |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Wed Mar 23, 2011 3:29 am Post subject: |
|
|
| keith wrote: | I was surprised in this thread that I ran afoul of both Danny and Ron. I was simply trying to make two points:
1. If you post a partially solved puzzle, please distinguish the the initial given cells from the single-digit solved cells.
1a. The candidates you specify in the unsolved multi-candidate cells may indicate some of your deductions, but they have nothing to do with defining the puzzle.
2. In any puzzle, a single-digit cell eliminates that candidate in any of its peers.
2a. Any candidate 123456789 is a starting possibility in any unsolved cell, until we start to use the initial cells, and then logic, to eliminate candidates.
WOW! was I surprised that these points are controversial, especially with cognoscenti like Ron and Danny. |
Please quote where I took issue with any of these points.
| ronk wrote: | | IMO the "other hands" are moot ... and mute. |
| keith wrote: | | After Ron posted his message that he was mute (synonym = dumb), I posted my message. |
The intended meaning was that "the other hands" were without voice. Your intentional twisting of my meaning can't possibly put you in a better light.
| keith wrote: | | Danny's last post is correct as to my motives. |
Sigh, I didn't think it was necessary to say ... When presented with a partially solved puzzle, [ed: but not the initial clues], both clues and solved cells must be treated as clues. |
|
| 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
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
