Page 1 of 1

Solving techniques that can't be used

Posted: Tue Feb 06, 2007 5:49 am
by Princess Amy
I've found you can't trust unique rectangles in clueless specials or samurai puzzles.

Are there other techniques you can't use?

Mostly, I've been able to solve all the clueless puzzles with naked or hidden groups, and not many other techniques.

But I'm stuck on this one!

Thanks,

Amy Grace

Posted: Tue Feb 06, 2007 6:24 pm
by Ruud
Unique Rectangles, BUG and BUG-Lite are the techniques that need to be used with care in overlapping puzzles, but you do not have to throw them away completely.

When all the cells of the UR are exclusive to a single constituent puzzle, so none of the cells lie within overlapping regions, the UR move is safe to perform. A BUG or BUG-Lite move can even include cells from multiple constituent puzzles, as long as each cell (except the surplus) has 2 candidates and each row, column and box to which those cells belong also have 2 candidates for those 2 digits. This is a little harder to check in overlapping puzzles.

Ruud

Posted: Sun May 27, 2007 7:04 am
by zoltag
Ruud wrote:Unique Rectangles, BUG and BUG-Lite are the techniques that need to be used with care in overlapping puzzles, but you do not have to throw them away completely.


Ruud
In any clueless, where the grid is all blue,
I think that you can ignore UR. BUG?

I have solved one where the blue suggested multiple solutions.

Posted: Sun May 27, 2007 3:00 pm
by Ruud
Unique Rectangles in the blue grid can be part of a BUG-Lite in the complete puzzle.

Consider the following situation:

Image

The blue grid contains a deadly pattern (UR) for a+b in boxes 6 & 9.
Grid 9 has a deadly pattern (UR) for a+b in boxes 5+8.

Grid 6 has a UR type 1 in boxes 2+5. When candidate c is eliminated from r2c4 in this grid, the entire structure would form a deadly pattern, suggesting 2 or more solutions. Therefore we can place digit c in r2c4 in grid 6.

Posted: Tue May 29, 2007 4:21 am
by zoltag
Very nice graphics!

In ClueLess Specials is it fair to say that URs and BUGS are ok as long as the elimination is NOT IN THE BLUE ?

Posted: Tue May 29, 2007 6:10 pm
by Ruud
When a UR or BUG is present in the white cells of a constituent puzzle, you can apply it as usual. When it includes blue cells, you must track the situation in the other puzzle(s), as I did in the example. A deadly pattern that spans multiple grids can never be a UR, but it could be a BUG or BUG-Lite.

Eliminations in blue cells are possible. If you look at the example, the extra candidate c could also be in one of the blue cells, allowing you to eliminate a&b from that cell.

Ruud