dailysudoku.com Forum Index dailysudoku.com
Discussion of Daily Sudoku puzzles
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Can this simple forcing chain be "formalised"?

 
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Solving techniques, and terminology
View previous topic :: View next topic  
Author Message
George Woods



Joined: 28 Mar 2006
Posts: 304
Location: Dorset UK

PostPosted: Sat Jan 10, 2009 10:49 am    Post subject: Can this simple forcing chain be "formalised"? Reply with quote

Given that today's forcing chains often become tomorrow's standard technique, I have a simple focing chain here that I cannot formalise into a conventional logical technique. Can anyone help?

Code:

+------------+--------------------+---------------+
| 7   28  1  | 2689   24689 24689 | 3  2459 2459  |
| 5   238 4  | 123789 1289  23789 | 6  129  129   |
| 9   23  6  | 123    5     234   | 7  8    124   |
+------------+--------------------+---------------+
| 34  9   57 | 2578   28    1     | 58 6    23457 |
| 8   147 2  | 567    3     567   | 9  1457 1457  |
| 13  6   57 | 4      289   25789 | 58 1257 12357 |
+------------+--------------------+---------------+
| 146 5   9  | 16     7     46    | 2  3    8     |
| 126 127 8  | 123569 1269  23569 | 4  579  579   |
| 24  247 3  | 2589   2489  24589 | 1  579  6     |
+------------+--------------------+---------------+

Play this puzzle online at the Daily Sudoku site

The chain is one that shows that 4 cannot be at r1c5 and so r9c5 must be 4

if r1c5 =4 then r3c4=1, r7c4=6 But also r8c5 =6 so fails

I Know the grid above can be simplified eg 56 in box 5 - but this does not affect the argument!
Back to top
View user's profile Send private message
daj95376



Joined: 23 Aug 2008
Posts: 3854

PostPosted: Sat Jan 10, 2009 12:01 pm    Post subject: Reply with quote

Code:
(4)r1c5 - (4)r3c6 = (4-1)r3c9 = (1)r3c4 - (1=6)r7c4 - (6)r8c5 = (6)r1c5 => [r1c5]<>4

Your chain appears to be very generic in nature. I doubt if it has a name.
Back to top
View user's profile Send private message
Asellus



Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA

PostPosted: Sun Jan 11, 2009 11:41 am    Post subject: Reply with quote

I suspect that George probably saw the chain more in this manner, with the {23} locked set explicit:

(4)r1c5 - ALS[(4)r3c6={23}r3c26] - ({23}=1)r3c4 - (1=6)r7c4 - (6)r8c5=(6-4)r1c5; r1c5<>4

Viewed as such, it still has no name. However, there's another way to see this that can be named. First, note the {1234} ALS in r3c246. This contains the strong inference ALS[(1)r3c4=(4)r3c6], which is a 14 "pseudocell". This can be used to form an "XY-Wing" with <6> pincers:

(6=1)r7c4 - ALSr3c246[(1)r3c4=(4)r3c6] - (4=6)r7c6; r8c5<6> r1c5=6 (i.e. r1c5<>4)

What's more, this "XY-Wing with pseudocell" could be formed with <1> pincers or <4> pincers as well. That's because it's a continuous loop. So we could call it an "XY-Wing with pseudocell loop"! Here it is with a fully reduced basic grid:
Code:
+---------------+-------------------------+----------------+
| 7     28   1  | 2689     24689  #2689-4 | 3   2459  2459 |
| 5     238  4  |#23789-1  1289    23789  | 6   129   129  |
| 9     23   6  |b123      5      b234    | 7   8     124  |
+---------------+-------------------------+----------------+
| 34    9    57 | 278      28      1      | 58  6     34   |
| 8     14   2  | 56       3       56     | 9   147   147  |
| 13    6    57 | 4        89      789    | 58  12    123  |
+---------------+-------------------------+----------------+
|#14-6  5    9  |a16       7      c46     | 2   3     8    |
| 126   17   8  |#2359-16 #129-6  #2359-6 | 4   579   579  |
| 24    47   3  | 2589     2489   #2589-4 | 1   579   6    |
+---------------+-------------------------+----------------+

(6=1)r7c4 - ALSr3c246[(1)r3c4=(4)r3c6] - (4=6)r7c6 - Loop; r2c3<>1; r19c6<>4; r7c1|r8c456<>6
I have marked all those victims with "#".

This grid is actually quite interesting. There is also a Sue de Coq, marked "@" below:
Code:
+---------------+-------------------------+----------------+
| 7     28   1  | 2689    #46-289  24689  | 3   2459  2459 |
| 5     238  4  | 123789  #1-289   23789  | 6   129   129  |
| 9     23   6  | 123      5       234    | 7   8     124  |
+---------------+-------------------------+----------------+
| 34    9    57 | 278     @28      1      | 58  6     34   |
| 8     14   2  | 56       3       56     | 9   147   147  |
| 13    6    57 | 4       @89      789    | 58  12    123  |
+---------------+-------------------------+----------------+
| 146   5    9  |@16       7      @46     | 2   3     8    |
| 126   17   8  |#2359-16 @1269   #2359-6 | 4   579   579  |
| 24    47   3  | 2589    @2489   #2589-4 | 1   579   6    |
+---------------+-------------------------+----------------+

While not what George saw, it has the same overall effect.
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Solving techniques, and terminology All times are GMT
Page 1 of 1

 
Jump to:  
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