Puzzle Description:

(This puzzle was recovered from the 2016 database crash. Description not available.)

#1: Kristen Vognild (kristen) on Feb 7, 2019 [SPOILER]       classify

Tricky to get started, but it looks like the horns are a bit out of tune. ;)

For me, finding the spaces where the 2 in R1 and the 3 in C9 COUDLN'T be did the trick. After that, line logic.

Found to have a unique solution by valerie.

Found to be solvable with deep lookahead by valerie.

I solved it with the time-honored but not-officially-approved method of doing line logic as far as I could (and some edge logic in the top row), and then doing "save" and coloring in one square along row 9 and clicking "helper." If it got red dots, I clicked "revert" and then put a dot where I had colored in the square, and clicked on "Save." And repeat, until I could go back to line logic. I like Cupcakey's approach much better!

I'm not sure where to find the official definitions of the differences between moderate and deep lookahead. If anybody has a pointer, I'll go read about it.

Deep lookahead is more a consensus than a definition.

Moderate lookahead is one of our usual advanced logics or an unusual one that has no more than 2-3 logical steps before you can mark a pixel.

Deep lookahead is a logical move with the steps laid out that is longer than just a few steps before you can mark a pixel. It is kind of important (imo) that the method is laid out since this is by definition not using a usual solving method.

Thanks Joe!

After LL:

- whether C2R3 or C5R3 are black, the resulting dot in R6 will make C6,7,9,10R6 black.

LL to finish.

I think this qualifies as moderate lookahead, but Joe's a much more accurate judge of lookaheadedness than I am, so I'll defer to him on this one.

David, I do think that sounds like moderate lookahead.

Also, I should update what I wrote in #5, because I now understand that that is considered guessing, so I wouldn't set the solvability based on that approach anymore.

This two-way is a bit less obvious than the normal.
If the 3 in r3 does not go left into c2 (triggering the 2), then it *must* go right into c5 (triggering the 3).

> If it triggers the 3 in c5, then the rest of that column from r6 down must be white

> If it triggers the 2 in c2, then r6c2 must be white. That forces the 2,2,2 in that row into one possible layout, in which r6c5 is white.

Either way r6c5 must be white.
LL to finish

Found to be solvable with moderate lookahead by infrapinklizzard.

Goto next topic

You must register and log in to be able to participate in this discussion.