Nonogram / Tsunami / Picross help needed

[notnow]

Is it actually possible this puzzle properly? (For those interested, it is from Picross DS, 11-A of free mode).

I'm stuck near then end and I can't see a logical solution without having to "guess".
Can anyone help?

I won't post the whole thing since i have completed most of it (correctly) and there is only a small section in the top right left, so here is the reduced part:

Code: Select all

__________1
____1 2 1 2 2 2 2 2
1 2 . . . . . . . .
2 2 . . . . . . . .
1 2 . . . . . . . .
2 2 . . . O . . . .
__1 # . . . . # # .

Where
# = Blank
O = Filled in
. = Unknown

If this is hard to read then let me know a better way of posting them

Thanks!

[kiwijam]

I noticed some nice logic that solves it for you (in a spoiler so that others can think about it first):

PRIME_BBCODE_SPOILER_SHOW PRIME_BBCODE_SPOILER:

The 4th row contains two pairs.
If the marked "O" is in the right pair of the two then the rightmost 3 cells are all empty, and so the 2nd row must end with three full cells. Not OK.
So the marked "O" is in the left pair, and then the 7th cell must be "O" also. The rest will solve quickly.

[notnow]

Ah thanks that makes sense!

It's tricky when they start requiring multiple levels of recursion in order to find contradictions.
I think this is the first time i've come across one like this.

Hopefully i'll get better at these types eventually hehe.

Thanks!

[notnow]

Code: Select all

        |    1             2
        |  3 3 1 2 4   1 2 2
        |5 2 2 1 2 2 2 2 2 3
----------------------------
       2|# # . O . # # O O #
   1 1 2|. . . # . . . # O O
   2 3 1|. . . . . O . . # O
     1 3|. . . . . O . . # #
     2 1|. . . . . O . . . .
     1 2|. . . . . . . # O O
     1 2|. . . . . . . . . .
     1 2|. . . . . . . . # #
   2 2 1|. . . . . . . . # O
 2 1 1 1|. . . . . . . . # O
   2 1 1|. . . . . . . . # O

Okay, so i've solved a few with contradictions but this one seems a lot harder!

Here is the full puzzle, if anyone is interested:

http://pastebin.com/gx856cFz

You can save it to a text file and open it with this tool if you want:

http://jsimlo.sk/griddlers/

[PuzzleScot]

PRIME_BBCODE_SPOILER_SHOW PRIME_BBCODE_SPOILER:

Look at column 1. If the 5 started in the topmost free space, rows 3 and 5 have a forced single in column 2...

Edit: There is a mistake in your diagram - you have 3 cells filled in the top row, but there's only a single 2 clue.
Even if it was "2 2", the puzzle still does not solve.

[notnow]

Ah thanks!

I see what you are saying - if 5 is in the top 5 columns then it forces 2 singles in column 2 which can't be as it has a 3 and a 2 in.
So the 5 must start from at LEAST the 5 row down (inc the already blank sq) in order to avoid those 1s, so you can fill in from the 7th row to the 9th in the first column and work from there.

At least, i think that is what you are saying.

And yes, i did make a mistake in my diagram sorry. It's correct in the griddler link one.

Here is a picture of how i meant to post it (including the rest of the puzzle):

[sknight]

My immediate reaction is that you can do a lot to nail down where that "5" in the left column is, based on what has to happen in the second column.

PRIME_BBCODE_SPOILER_SHOW PRIME_BBCODE_SPOILER:

For example, if the 5 started in that topmost blank space, the fact that the first entries across go "1, 2, 1, 2, 1" in each of those rows would leave
two single isolated blocks in column 2, which is incompatible with the given clues for column 2. In fact, I think you can narrow it down to
either one or two blank spaces at the bottom before the 5 kicks in, which lets you fill in four black squares and white out a lot of others.