View previous topic :: View next topic 
Author 
Message 
keith
Joined: 19 Sep 2005 Posts: 3287 Location: near Detroit, Michigan, USA

Posted: Sun Mar 17, 2013 1:09 am Post subject: Impossible Menneske No. 09 


Can't believe the last No. 08 could be solved by a skyscraper!
Code:  M6911894 Impossible (4637)
++++
 2 5 .  6 . .  . . 9 
 . . .  . . .  . 5 . 
 . . .  . 3 .  . 6 . 
++++
 6 . .  . 4 3  . . . 
 . . 7  . . .  . . 5 
 . . .  2 1 .  . 4 . 
++++
 . 7 .  1 . .  4 . . 
 . . .  . . .  . 1 . 
 1 9 .  . . 6  . . 7 
++++

Play this puzzle online at the Daily Sudoku site
Keith 

Back to top 


JC Van Hay
Joined: 13 Jun 2010 Posts: 490 Location: Charleroi, Belgium

Posted: Sun Mar 17, 2013 8:08 am Post subject: 


#1. Locked Subsets and Nishio; UP27
r5c5=6
NP(89)r5c46
Skyscraper(4R19):=> 4r23c4.r8c6; HP(34)r89c4
LC(5r789c5)
XChain(7R86C8) :=> 7r1c56; LC(7r1c78);r1c5=8=r5c4,r5c6=9
LC(9r78c5) Code:  ++++
 2 5 134  6 8 14  137 37 9 
 34789 13468 134689  79 27 1247  1238 5 12348 
 4789 148 1489  579 3 12457  128 6 1248 
++++
 6 128 12589  57 4 3  12789 2789 128 
 34 1234 7  8 6 9  123 23 5 
 3589 38 3589  2 1 5(7)  389(67) 4 368 
++++
 358 7 23568  1 259 28  4 2389 2368 
 3458 23468 234568  34 25(79) 28(7)  2358(69) 1 2368 
 1 9 23458  34 25 6  2358 238 7 
++++  #2. Loop[4] : 7r6c6=(76)r6c7=(69)r8c7=(97)r8c5=7r8c6 @ :=> 7r23c6,r6c7=67,r8c7=69,r8c5=79; UP59; LC(1r12c3); UP81 

Back to top 


DonM
Joined: 15 Sep 2009 Posts: 51

Posted: Mon Mar 18, 2013 1:39 am Post subject: 


JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 

Back to top 


keith
Joined: 19 Sep 2005 Posts: 3287 Location: near Detroit, Michigan, USA

Posted: Mon Mar 18, 2013 2:46 am Post subject: 


DonM wrote:  JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 
Sudoku Susser does both Bowman's Bingo and Nishio.
Keith 

Back to top 


JC Van Hay
Joined: 13 Jun 2010 Posts: 490 Location: Charleroi, Belgium

Posted: Mon Mar 18, 2013 8:26 am Post subject: 


DonM wrote:  JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 
I borrowed the term Nishio from Ruud's SudoCue Solver to mean the elimination of all the candidates that do not occur in any of the solutions of a single digit. In some sense, it is the easiest and fastest first thing to do while filling the puzzle with pencilmarkings as it requires no technique in the usual sense, but only a global description of all the solutions of a single digit. For reasons of "esthetical presentation", to find a "simple" interpretation of each elimination is another, but not always as easy and as fast, interesting task to do afterwards!
With the Disjoint Subsets, it produces a "clean" initial Possibility Matrix. 

Back to top 


DonM
Joined: 15 Sep 2009 Posts: 51

Posted: Wed Mar 20, 2013 12:54 am Post subject: 


JC Van Hay wrote:  DonM wrote:  JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 
I borrowed the term Nishio from Ruud's SudoCue Solver to mean the elimination of all the candidates that do not occur in any of the solutions of a single digit. In some sense, it is the easiest and fastest first thing to do while filling the puzzle with pencilmarkings as it requires no technique in the usual sense, but only a global description of all the solutions of a single digit. For reasons of "esthetical presentation", to find a "simple" interpretation of each elimination is another, but not always as easy and as fast, interesting task to do afterwards!
With the Disjoint Subsets, it produces a "clean" initial Possibility Matrix. 
I'm still totally confused. Nishio was always used to refer to a method that involved plugging in a number in a cell and seeing where it went and fell out of use around 2006 after more elegant methods were 'discovered'. It has never been used (afaik) as a method to prepare a puzzle for more advanced solving.
You are such a clever solver and I'd like to be able to follow your solutions, but I find it hard to do so (except for the part(s) following a grid) because of a lack of a common standard starting point. 

Back to top 


DonM
Joined: 15 Sep 2009 Posts: 51

Posted: Wed Mar 20, 2013 1:17 am Post subject: 


Menneske Impossible 9 (ER=7.2) at SSTS (post Simple Sudoku Technique Set ie. 'No hint possible'):
Code: 
**
 2 5 1348  6 78 148  1378 378 9 
 34789 13468 134689  789 2789 124789  12378 5 12348 
 4789 148 1489  5789 3 1245789  1278 6 1248 
++
 6 128 12589  57 4 3  12789 2789 128 
 34 1234 7  89 6 89  123 23 5 
 3589 38 3589  2 1 57  36789 4 368 
++
 358 7 23568  1 2589 289  4 2389 2368 
 3458 23468 234568  34 25789 2789  235689 1 2368 
 1 9 23458  34 258 6  2358 238 7 
** 
1. grp(7)r23c4=r4c4r4c8=(7)r1c8 => 3 singles > LC(9):9r2c5, LC(7):7r23c6
2. (7)r4c4=grp(7)r23c4r2c5=(79)r8c5=r7c5r7c8=(9)r4c8 => 7r4c8 > r1c8=7 > nq(1238)r1235c7 => r9c7=5
stte 

Back to top 


JC Van Hay
Joined: 13 Jun 2010 Posts: 490 Location: Charleroi, Belgium

Posted: Wed Mar 20, 2013 7:16 pm Post subject: 


DonM wrote:  JC Van Hay wrote:  DonM wrote:  JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 
I borrowed the term Nishio from Ruud's SudoCue Solver to mean the elimination of all the candidates that do not occur in any of the solutions of a single digit. In some sense, it is the easiest and fastest first thing to do while filling the puzzle with pencilmarkings as it requires no technique in the usual sense, but only a global description of all the solutions of a single digit. For reasons of "esthetical presentation", to find a "simple" interpretation of each elimination is another, but not always as easy and as fast, interesting task to do afterwards!
With the Disjoint Subsets, it produces a "clean" initial Possibility Matrix. 
I'm still totally confused. Nishio was always used to refer to a method that involved plugging in a number in a cell and seeing where it went and fell out of use around 2006 after more elegant methods were 'discovered'. It has never been used (afaik) as a method to prepare a puzzle for more advanced solving.
 To avoid any confusion, I ought to use the acronym USDE for Unit and single Digit Eliminations instead of Locked Subsets and Nishio ! The basic idea being, before using more "advanced solving", to get a "clean initial Possibility Matrix" where each candidate belongs to a solution of each unit in which it is sitting and of the digit it represents as a direct application of the Sudoku puzzle rules. The only player I know that would do so is my intellectual guide, Stephen Kurzhals, as I can infer from his blog on the au site.
Concerning the actual puzzle and as I never used neither Simple Sudoku Solver nor SSTS, I am confused by the fact that your XWing in step 1 is not included in SSTS as it gives 7r1c56, while SSTS would only give 7r1c6 !? 

Back to top 


DonM
Joined: 15 Sep 2009 Posts: 51

Posted: Wed Mar 20, 2013 9:02 pm Post subject: 


JC Van Hay wrote:  DonM wrote:  JC Van Hay wrote:  DonM wrote:  JC, I'm wondering where the Nishio came from. I haven't seen it mentioned as part of a solution for a long time. 
I borrowed the term Nishio from Ruud's SudoCue Solver to mean the elimination of all the candidates that do not occur in any of the solutions of a single digit. In some sense, it is the easiest and fastest first thing to do while filling the puzzle with pencilmarkings as it requires no technique in the usual sense, but only a global description of all the solutions of a single digit. For reasons of "esthetical presentation", to find a "simple" interpretation of each elimination is another, but not always as easy and as fast, interesting task to do afterwards!
With the Disjoint Subsets, it produces a "clean" initial Possibility Matrix. 
I'm still totally confused. Nishio was always used to refer to a method that involved plugging in a number in a cell and seeing where it went and fell out of use around 2006 after more elegant methods were 'discovered'. It has never been used (afaik) as a method to prepare a puzzle for more advanced solving.
 To avoid any confusion, I ought to use the acronym USDE for Unit and single Digit Eliminations instead of Locked Subsets and Nishio ! The basic idea being, before using more "advanced solving", to get a "clean initial Possibility Matrix" where each candidate belongs to a solution of each unit in which it is sitting and of the digit it represents as a direct application of the Sudoku puzzle rules. The only player I know that would do so is my intellectual guide, Stephen Kurzhals, as I can infer from his blog on the au site.

Yes, I recognized Steve K's influence. To the best of my knowledge, outside of his blog, Steve did virtually all of his solving of advanced puzzles with us at Eureka starting in 2007 where we solved the UK forum Extremes and puzzles of ER=8.1 and above for 23 years. For a period after that he continued to present advanced solutions with the solver known as ttt.
What Steve presented in his blog and solving with us were two different things. In his blog, he knew that his audience varied from beginners to advanced solvers so he presented puzzles of varying difficulty and didn't start the puzzle from an SSTS position so he could show beginners the more basic beginning moves.
However, virtually all the puzzles that he, we or ttt presented on Eureka started at the SSTS position. None of us wanted to bother with intitial basic methods and we wanted to be able to compare each other's manual solutions which was only possible because we were starting at the same SSTS position (which means simply loading the puzzle into Simple Sudoku and pressing the F11 key until you get a 'No hint available' message at the bottom of the puzzle).
I have to say that I don't think I was ever able to come up with a solution that was better than Steve's. To this day, I think he had a better grasp on the application of the logic of sudoku to manual solving than virtually anyone else and I certainly learned more about advanced manual solving from him than almost anyone else (though there were other great influential sudoku minds around at that time).
edit: Actually, doing a quick check, Steve did use the SSTS position in his later, and final, blogs:
http://sudoku.com.au/sudokutips.aspx?Go=T1311991
http://sudoku.com.au/sudokutips.aspx?Go=P1311991
http://sudoku.com.au/sudokutips.aspx?Go=C921991
Quote: 
Concerning the actual puzzle and as I never used neither Simple Sudoku Solver nor SSTS, I am confused by the fact that your XWing in step 1 is not included in SSTS as it gives 7r1c56, while SSTS would only give 7r1c6 !? 
I don't see an XWing, but my step 1 is essentially a skyscraper using a group. You're right that ordinarily the skyscraper move would result in 7r1c56 (if 7r1c6 was still there), but Simple Sudoku doesn't have things like skyscrapers or kites in its technique set. Just for giggles I went back and ran the puzzle through SS and it turned out that it removed 7r1c6 using Multiple Colors. 

Back to top 


oaxen
Joined: 10 Jul 2006 Posts: 96

Posted: Thu Mar 21, 2013 12:45 pm Post subject: 


To me the Menneske puzzles are interesting as there are very few bivalues and my method is to find out which of them are possible ways to solve. In this case Block E with the 5 and 7 is interesting, all the other ones leads to dead ends. 5 in R6C6 goes direct to the target. 5 in R4C4 needs opening af a seconadary chain. I have a feeling that every puzzle can be one steppers  if you find the right bivalue. 

Back to top 


JC Van Hay
Joined: 13 Jun 2010 Posts: 490 Location: Charleroi, Belgium

Posted: Thu Mar 21, 2013 1:16 pm Post subject: 


oaxen wrote:  ... I have a feeling that every puzzle can be one steppers  if you find the right bivalue.  Hmmm... What about this puzzle? 

Back to top 


oaxen
Joined: 10 Jul 2006 Posts: 96

Posted: Thu Mar 21, 2013 2:45 pm Post subject: 


JC Van Hay wrote:  oaxen wrote:  ... I have a feeling that every puzzle can be one steppers  if you find the right bivalue.  Hmmm... What about this puzzle? 
As I said, with a 5 in R6C6 all the rest is basic. Just for fun I afterwards tested other bivalues, but none lead to a solution without opening secondary chains, like 5 in R4C4 demands. 

Back to top 


oaxen
Joined: 10 Jul 2006 Posts: 96

Posted: Thu Mar 21, 2013 2:50 pm Post subject: 


Sorry I misunderstood you. You sent me a link to something. Now already printed, we will se what happens when I have the time to start with it. 

Back to top 


ronk
Joined: 07 May 2006 Posts: 398

Posted: Thu Mar 21, 2013 10:38 pm Post subject: 


JC Van Hay wrote:  #1. Locked Subsets and Nishio; UP27
r5c5=6
NP(89)r5c46 
To add some precision to this discussion, would you please provide the pencilmarks and tag the candidates involved in the first (or the most interesting) such "Nishio" move of your solution path? 

Back to top 


JC Van Hay
Joined: 13 Jun 2010 Posts: 490 Location: Charleroi, Belgium

Posted: Fri Mar 22, 2013 6:50 am Post subject: 


ronk wrote:  JC Van Hay wrote:  #1. Locked Subsets and Nishio; UP27
r5c5=6
NP(89)r5c46 
To add some precision to this discussion, would you please provide the pencilmarks and tag the candidates involved in the first (or the most interesting) such "Nishio" move of your solution path?  I posted a less trivial example of what I called "Nishio" on the Players forum here 

Back to top 


keith
Joined: 19 Sep 2005 Posts: 3287 Location: near Detroit, Michigan, USA

Posted: Sat Mar 23, 2013 8:56 pm Post subject: 


JC Van Hay wrote:  ronk wrote:  JC Van Hay wrote:  #1. Locked Subsets and Nishio; UP27
r5c5=6
NP(89)r5c46 
To add some precision to this discussion, would you please provide the pencilmarks and tag the candidates involved in the first (or the most interesting) such "Nishio" move of your solution path?  I posted a less trivial example of what I called "Nishio" on the Players forum here 
What you did in that puzzle reminds me of Medusa Coloring.
I learned something today, and used it to solve a puzzle that otherwise had me stumped. For a pencil and paper pattern guy, that's something!
Keith 

Back to top 


keith
Joined: 19 Sep 2005 Posts: 3287 Location: near Detroit, Michigan, USA


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
