| View previous topic :: View next topic   | 
	
	
	
		| Author | 
		Message | 
	
	
		dongrave
 
 
  Joined: 06 Mar 2014 Posts: 572
 
  | 
		
			
				 Posted: Thu Nov 26, 2015 2:38 pm    Post subject: Another 1sudolu.net Evil Puzzle | 
				     | 
			 
			
				
  | 
			 
			
				The only single step solution that I found for this is awful ugly!  Could someone out there show me a 'nice' single-stepper?
 
 	  | Code: | 	 		  
 
 1sudoku.net #528902 Evil
 
 +-------+-------+-------+ 
 
 | 6 . . | . 9 . | . . 4 | 
 
 | 8 . . | 2 . 4 | . . 7 | 
 
 | . . 4 | . 6 . | 8 . . | 
 
 +-------+-------+-------+ 
 
 | . 2 . | 1 . 8 | . 5 . | 
 
 | . . . | . . . | . . . | 
 
 | . 7 . | 3 . 5 | . 4 . | 
 
 +-------+-------+-------+ 
 
 | . . 2 | . 5 . | 4 . . | 
 
 | 5 . . | 7 . 9 | . . 2 | 
 
 | 7 . . | . 1 . | . . 5 | 
 
 +-------+-------+-------+ 
 
 | 	 
  | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		JC Van Hay
 
 
  Joined: 13 Jun 2010 Posts: 494 Location: Charleroi, Belgium
  | 
		
			
				 Posted: Thu Nov 26, 2015 6:53 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				| To optimally solve the puzzle, I need Singles, 1 Locked Candidates, 1 Naked Pair and 2 "2-Fishes" or Singles, 2 Locked Candidates, 1 Naked Pair and 1 Kraken 9R9. Which do you prefer ? | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		dongrave
 
 
  Joined: 06 Mar 2014 Posts: 572
 
  | 
		
			
				 Posted: Thu Nov 26, 2015 7:53 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | JC Van Hay wrote: | 	 		  | To optimally solve the puzzle, I need Singles, 1 Locked Candidates, 1 Naked Pair and 2 "2-Fishes" or Singles, 2 Locked Candidates, 1 Naked Pair and 1 Kraken 9R9. Which do you prefer ? | 	  
 
Holy crow!  No WONDER my solution is SO ugly!  ugh. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 3:37 am    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Code: | 	 		  
 
+------------+--------+---------------+
 
| 6  35  7   | 8 9  1 | 235  23   4   |
 
| 8  159 159 | 2 3  4 | 1569 169  7   |
 
| 2  139 4   | 5 6  7 | 8    139  139 |
 
+------------+--------+---------------+
 
| 34 2   69  | 1 47 8 | 379  5    369 |
 
| 34 158 158 | 9 47 6 | 1237 1238 138 |
 
| 19 7   68  | 3 2  5 | 19   4    68  |
 
+------------+--------+---------------+
 
| 19 189 2   | 6 5  3 | 4    7    189 |
 
| 5  4   13  | 7 8  9 | 136  136  2   |
 
| 7  6   389 | 4 1  2 | 39   389  5   |
 
+------------+--------+---------------+
 
 | 	  
 
Play this puzzle online at the Daily Sudoku site
 
 
Just wanted to post basics. I can't make JC's Kraken work. There's a Skyscraper on 8. There are some UR and DP possibilities, including a 34-47-37 in r5 but so far I haven't been able to do anything with them. I also haven't found JC's fish yet, | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		dongrave
 
 
  Joined: 06 Mar 2014 Posts: 572
 
  | 
		
			
				 Posted: Fri Nov 27, 2015 6:00 am    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Marty R. wrote: | 	 		  
 
Just wanted to post basics. I can't make JC's Kraken work. There's a Skyscraper on 8. There are some UR and DP possibilities, including a 34-47-37 in r5 but so far I haven't been able to do anything with them. I also haven't found JC's fish yet, | 	  
 
 
Thanks Marty!  That looks familiar alright!  You know what?  If I ever publish my book, I'm going to use the pen name 'JC Kraken'!  Here's my one-stepper that I told you about (pretty darn ugly, isn't it?)
 
 
(8=6)r6c3-(6=9)r4c3-(9=1)r6c1[-(1=9)r7c1-r7c9;-(1=9)r6c7-(9=3)r9c7-(3=16)r8c78]-(19=8)r7c9-(8=6)r6c9 => ugh. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 12:51 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Quote: | 	 		  | (8=6)r6c3-(6=9)r4c3-(9=1)r6c1[-(1=9)r7c1-r7c9;-(1=9)r6c7-(9=3)r9c7-(3=16)r8c78]-(19=8)r7c9-(8=6)r6c9 => ugh.. | 	  
 
 
Don, you're getting the hang of it. As much as I hate to rain on your parade, I'm gonna do it anyways.
 
 
1)   In the red area not all the inferences alternate. There are two consecutive weak inferences; the 2nd dash needs to be an =. 
 
 
2) The end is in cyan. If the intent here was a pincer attack, the notation should end with =8, so as to eliminate something. I've taken the liberty of rewriting.
 
 
(8=6)r6c3-(6=9)r4c3-(9=1)r6c1[-(1=9)r7c1-(9=18)r7c9=> -8r6c9,although I don't know how far the elimination take us. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		JC Van Hay
 
 
  Joined: 13 Jun 2010 Posts: 494 Location: Charleroi, Belgium
  | 
		
			
				 Posted: Fri Nov 27, 2015 1:18 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				Marty : after the Skyscraper on 8s, there is a Skyscraper on 9s.
 
 
dongrave : I have the same single step   : 8r6c9 -> 9R9={}. Here is a less ugly interpretation using only 5 constraints :
 
	r9c8=9 -> r7c9=8, r6c9=6
 
	or
 
	Skyscraper(9r6c17, 9r9c37) -> r4c3=6, r6c8=8, r6c9=6
 
 
	Or using your notation :
 
	(8=6)r6c3-(6=9)r4c3-[9r6c1=9r6c8-9r9c7; 9r9c3]=(9-8)r9c8=8r7c9 :=> -8r6c9  | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		dongrave
 
 
  Joined: 06 Mar 2014 Posts: 572
 
  | 
		
			
				 Posted: Fri Nov 27, 2015 1:29 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Marty R. wrote: | 	 		  | Don, you're getting the hang of it. As much as I hate to rain on your parade, I'm gonna do it anyways. | 	  
 
Hi Marty, Thanks!  You didn't rain on my parade, I appreciate your help!  I see that I screwed up a few things with my ugly expression.  I intended to use the square brackets (even though it's obviously not right) to have two sub-expressions that both begin after the strong link '(9=1)r6c1' (which is why they both begin with a weak link) and have the 1st sub-expression end by showing you that 'r7c9<>9' and then have the 2nd one end by showing you that 'r8c78=16' so 'r7c9<>1' so r7c9=8 so r6c9=6 and then I forgot to put 'contradiction' (because r6c3 and c9 are both 6) and then I also didn't put the conclusion '=> r6c3<>6' (at least I remembered to put the 'ugh' at the end).  How the heck do you represent what I was trying to do in Eureka?  I thought I had it (darn).  Oh!  one last thing - when I look for single step chains, I always make sure that the conclusion solves it so at least that part I got right! (I think)      Thanks again! | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 3:57 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Quote: | 	 		   	  | Code: | 	 		  +------------+--------+---------------+
 
| 6  35  7   | 8 9  1 | 235  23   4   |
 
| 8  159 159 | 2 3  4 | 1569 169  7   |
 
| 2  139 4   | 5 6  7 | 8    139  139 |
 
+------------+--------+---------------+
 
| 34 2   69  | 1 47 8 | 379  5    369 |
 
| 34 158 158 | 9 47 6 | 1237 1238 138 |
 
| 19 7   68  | 3 2  5 | 19   4    68  |
 
+------------+--------+---------------+
 
| 19 189 2   | 6 5  3 | 4    7    189 |
 
| 5  4   13  | 7 8  9 | 136  136  2   |
 
| 7  6   389 | 4 1  2 | 39   389  5   |
 
+------------+--------+---------------+  | 	 
  | 	  
 
 
Don, this is my best shot at achieving your desired result:
 
 
(8=6)r6c3-(6=9)r4c3-r4c79=r6c7-(9=3)r9c7-(3=16)r8c78-(1=98)r7c9=> -8r6c9.
 
 
The (3=16) might be clearer to the reader as (36=1), but I write ALS's the way that the experts like them.
 
 
If we want to use r7c1 we could use the following from the 3rd term through the end.
 
 
(9=1)r6c1-(1=9)r7c1-(9=18)r7c9=> -8r6c9
  Last edited by Marty R. on Fri Nov 27, 2015 4:02 pm; edited 1 time in total | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		keith
 
 
  Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 3:57 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				Not a single-stepper, but not so bad.
 
 
In Marty's after basics grid there is an extended XY-wing that take out 3 in R8C78, revealing a 16 pair.
 
 
After a few more basics there is a W-wing 89 - 68 that takes out 9 in R7C1.
 
 
Keith | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 4:09 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				| Keith, I'm afraid that I don't know what an extended XY-Wing is and I can't find anything that looks wingy with pincer 3. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		keith
 
 
  Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 4:51 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Marty R. wrote: | 	 		  | Keith, I'm afraid that I don't know what an extended XY-Wing is and I can't find anything that looks wingy with pincer 3. | 	  
 
Marty,
 
 
Maybe you call it "transport".  Look at 13 - 19 - 19 - 19 - 39 which has pincers on 3.
 
 
Keith | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 6:18 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				Keith,until recently I looked for those all the time and don't know why I stopped.. I call it an XY-Wing Chain because that's what it was called on the site where I learned it.
 
 
(I see your Lions are on a roll with four W's in the last six including some division rivals.) | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		keith
 
 
  Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 6:41 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				Marty,
 
 
Helmut and I were discussing the other day these "useless" XY-wings where you can extend the pincers  with a couple of coloring links.  They can be quite handy, but you need to remember to look for them.
 
 
I like that this puzzle succumbs to this and a classic W-wing.
 
 
Even better than the Lions is the fact that U. of Michigan once again has a football team.
 
 
Keith | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		keith
 
 
  Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
  | 
		 | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		dongrave
 
 
  Joined: 06 Mar 2014 Posts: 572
 
  | 
		
			
				 Posted: Fri Nov 27, 2015 7:16 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | keith wrote: | 	 		  ... dongrave, you may want to read it as a compendium of tips for pencil and paper solvers.
 
Keith | 	  
 
Thanks Keith!  I'll do that!  And thanks to Marty and Van Hay for your help!  Don. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		Marty R.
 
 
  Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
  | 
		
			
				 Posted: Fri Nov 27, 2015 8:51 pm    Post subject:  | 
				     | 
			 
			
				
  | 
			 
			
				 	  | Quote: | 	 		  | Helmut and I were discussing the other day these "useless" XY-wings where you can extend the pincers with a couple of coloring links. They can be quite handy, but you need to remember to look for them.  | 	  
 
 
I view pincer transport as one of the most powerful tools in the box. | 
			 
		  | 
	
	
		| Back to top | 
		 | 
	
	
		  | 
	
	
		 |