The New Sudoku Players' Forum

[peubear]

I am somewhat of a newbie. In trying to evaluate a computer program to work with, I had the following *hard* sudoko created by Sadman's Sudoko program. It is partially solved:

+---+---+---+
|500|010|040|
|947|508|000|
|060|000|005|
+---+---+---+
|050|070|002|
|080|156|070|
|700|040|560|
+---+---+---+
|800|000|030|
|000|780|106|
|070|020|009|
+---+---+---+

However, even though Sadman's program can find the unique solution, Pappocom's program cannot even VERIFY that this is a VALID sudoko puzzle. ????

Any comments would be appreciated...

Thanks, Stephen

[Sped]

Code: Select all

 *-----------*
 |5..|.1.|.4.|
 |947|5.8|...|
 |.6.|...|..5|
 |---+---+---|
 |.5.|.7.|..2|
 |.8.|156|.7.|
 |7..|.4.|56.|
 |---+---+---|
 |8..|...|.3.|
 |...|78.|1.6|
 |.7.|.2.|..9|
 *-----------*


After a few exclusions based on locked candidates, we have:

Code: Select all

 
 *-----------------------------------------------------------------------------*
 | 5       23      238     | 269     1       279     | 36789   4       378     |
 | 9       4       7       | 5       36      8       | 236     12^     13^     |
 | 123     6       1238    | 249     39      2479    | 3789    89      5       |
 |-------------------------+-------------------------+-------------------------|
 | 1346    5       13469   | 389     7       39      | 3(4)89  189     2       |
 | 234     8       2349    | 1       5       6       | 3(4)9   7       34*     |
 | 7       139     139     | 2389    4       239     | 5       6       138     |
 |-------------------------+-------------------------+-------------------------|
 | 8       129     124569  | 469     69      1459    | 247     3      (4)7     |
 | 234     239     23459   | 7       8       3459    | 1       25^     6       |
 | 1346    7       13456   | 346     2       1345    | 48*     58^     9       |
 *-----------------------------------------------------------------------------*


The xy chain

4-(r9c7)-8-(r9c8)-5-(r8c8)-2-(r2c8)-1-(r2c9)-3-(r5c9)-4 eliminates the 4s in r45c7 and r7c9.

Above, the ends of the chain are marked with *, other cells in the chain are marked with ^, and the candidates that get excluded have () around them.

After that it can be solved with nothing more complicated than triples.

In nice loop notation:

[r45c7,r7c9]-4-[r9c7]-8-[r9c8]-5-[r8c8]-2-[r2c8]-1-[r2c9]-3-[r5c9]-4-[r45c7,r7c9] => r45c7<>4, r7c9<>4

[tso]

This is a design issue with Pappocom software, not you. It is intentional. Puzzles that have anything more complicated than an an X-Wing (more or less) will either be called "very hard, but arguably unfair" -- the program will load these -- or "invalid" -- the program will *not* load these. Many of us have complained about it and argued strenuously for this to be changed, since the word "invalid" causes great confusion. (Many of the harder puzzles published in the LA Times and the UK Daily Telegraph are called "invalid" by Pappocom software.) It's unlikely that he's going to do anything about it. The software hasn't been updated in over a year. The majority of the puzzles you'll see posted in the advanced area of the forum cannot be dubbed into Pappocom software -- but they can be loaded into any of the other 5000 Sudoku programs available.