The New Sudoku Players' Forum

[eleven]

The external SIS is simply 9r7c3 & 9r9c9, thus
AUR(19)r79c27[9r9c9=(9-3)r7c3]=3r8c3-(3=5)r8c7 => r9c9<>5 to complete the puzzle.

Ah nice, i just saw the SIS 9r3c2=9r3c7 => r3c9=8, which did not solve the puzzle.

[Marty R.]

Code: Select all

+-------------+-------+------------+
| 25  3   569 | 8 1 7 | 4   26 59  |
| 258 78  56  | 9 3 4 | 57  26 1   |
| 1   79  4   | 2 5 6 | 379 38 89  |
+-------------+-------+------------+
| 9   2   7   | 4 6 1 | 8   5  3   |
| 6   5   1   | 3 7 8 | 2   9  4   |
| 3   4   8   | 5 9 2 | 6   1  7   |
+-------------+-------+------------+
| 4   189 39  | 7 2 5 | 139 38 6   |
| 7   6   35  | 1 8 9 | 35  4  2   |
| 58  189 2   | 6 4 3 | 159 7  589 |
+-------------+-------+------------+


Play this puzzle online at the Daily Sudoku site

(5=3)r8c3-(3=5)r8c7-(5=7)r2c7-(78=25)r2c2,r12c1=>r9c1<>5

[tlanglet]

Sorry, I posted twice; see below.

Ted

[tlanglet]

Thanks, Leren, Ronk. I wondered about the underlining in the chain. Now I see the significance.

This is a really pretty little puzzle- and a pretty solution.

The more I look at it, the neater it looks. Is that called something in particular?

The way I like to look at it is that 9 in r1c3 is a weak link to the 5,6 pair in r1c3, followed by strong links to the 6 in r2c3 and the 5 in r7c3, setting 5 and 6 in those cells and excluding them in r1c3.

Another view is that of ANP(9=56))r12c3-5r8c3=5r8c7-5r2c7=5r1c9 => ric9><9

Ted

[Leren]

SteveG48 wrote: The way I like to look at it is that 9 in r1c3 is a weak link to the 5,6 pair in r1c3, followed by strong links to the 6 in r2c3 and the 5 in r7c3, setting 5 and 6 in those cells and excluding them in r1c3.

I don't think this is quite right. This move does not (in general) depend on there being a strong link on 6 in Column 3.

SteveG48 wrote: Thanks, Leren. I see you making good use of things I haven't learned about like L2 and H2 wings. Can you point me to a reference on these?

Look at the post by Danny Jones in this thread for definitions of H, L, M, W and S wings.

http://forum.enjoysudoku.com/vanhegan-extreme-march-22-2013-t31035.html

This thread is the bible for M wings and M rings.

http://forum.enjoysudoku.com/m-wings-m-rings-exemplars-examples-t30030.html?hilit=strmckr#p227022

Leren

[Luke]

Another exploration of when DPs can be enjoined and when not. Saving grace ... only two strong inferences, big whoop.

(5)r9c79=(7)r3c7-(7=5)r2c7 ==>r8c7<>5

Code: Select all

 *--------------------------------------------------*
 | 25   3    569  | 8    1    7    | 4    26   59   |
 | 258  78   56   | 9    3    4    | 57   26   1    |
 | 1    79   4    | 2    5    6    |*39+7 *38 *89   |
 |----------------+----------------+----------------|
 | 9    2    7    | 4    6    1    | 8    5    3    |
 | 6    5    1    | 3    7    8    | 2    9    4    |
 | 3    4    8    | 5    9    2    | 6    1    7    |
 |----------------+----------------+----------------|
 | 4   *189  39   | 7    2    5    |*139 *38   6    |
 | 7    6    35   | 1    8    9    | 3-5  4    2    |
 | 58  *189  2    | 6    4    3    |*19+5 7   *89+5 |
 *--------------------------------------------------*

MUG(1389)==>r8c7<>5
||
(5)r9c79
||
(7)r3c7-(7=5)r2c7


Notes and blather follow, should you suffer from ennui.


If many of us saw this BUG-Lite in a puzzle, we'd immediately try to exploit it.

Code: Select all

 ----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  39 38 89       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  .       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  39 38 89       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  .       
----------+-----------+-----------

This puzzle almost has that pattern, but with the (89) wandering off in r9c9.

Code: Select all

 ----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  39 38 89       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  .       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  39 38 .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  89       
----------+-----------+-----------

If the (89) can be squared off again, the DP is restored.

Code: Select all

 ----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  39 38 89       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  .       
----------+-----------+-----------
 .  89 .  |  .  .  .  |  39 38 .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  89 .  |  .  .  .  |  .  .  89       
----------+-----------+-----------

Combined with (19)UR, a new pattern is easily built. Three of the four internal outs of the (19)AUR are now accounted for, which made this seem worth pursuing.

Code: Select all

 ----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  39 38 89       
----------+-----------+-----------
 .  .  .  |  .  .  .  |  .  .  .       
 .  .  .  |  .  .  .  |  .  .  .     
 .  .  .  |  .  .  .  |  .  .  .       
----------+-----------+-----------
 . 189 .  |  .  .  .  | 139 38 .       
 .  .  .  |  .  .  .  |  .  .  .     
 . 189 .  |  .  .  .  |  19 .  89       
----------+-----------+-----------

Is it a DP, specifically a MUG? Dunno for sure. It's balanced, and seems to handily pass the Julp MUG Test by proving all reductions lead to a BUG-Lite or no solution.

A bigger question to me these days is, can combined patterns find anything its components cannot? As confusing as combined pattern appear, at the very least they do seem capable of revealing SIS that were not so obvious at first glance.

As always, cross testing, debate welcome.

Edit: spelling gaffe

[eleven]

Nice find, Like.

A bigger question to me these days is, can combined patterns find anything its components cannot?

How do you mean that ?
Obviously the combined DP reduces the extra candidates in the single DP's, so it gives a smaller/better SIS. You would not get it by combining the single SIS sets with AND.
On the other hand the external SIS's are smaller for the single DP's, and thus of more worth than the combined external SIS.

[SteveG48]

[quote="Leren

SteveG48 wrote: Thanks, Leren. I see you making good use of things I haven't learned about like L2 and H2 wings. Can you point me to a reference on these?

Look at the post by Danny Jones in this thread for definitions of H, L, M, W and S wings.

http://forum.enjoysudoku.com/vanhegan-extreme-march-22-2013-t31035.html

This thread is the bible for M wings and M rings.

http://forum.enjoysudoku.com/m-wings-m-rings-exemplars-examples-t30030.html?hilit=strmckr#p227022

Leren[/quote]

Quite a bunch of material! Thanks.