Puzzle Description:

Not the prettiest pattern in the world but never mind

#1: Josh Greifer (joshgreifer) on Mar 19, 2010 [HINT]

Work from the bottom row up.

not sure if guessing is required, but i did anyway :) edge logic gave me a choice of placements for the bottom and i got lucky the first time and picked the right one

i solved it w/o guessing .... but no reward afterwards

Hmmm...

Yes, edge logic on the bottom row narrows it down to two possibilities, but it is a fancier edge logic than usual with one more level of look ahead needed. Both possibilities for the bottom row require the 4 in the second row from the bottom to be in the same place, so you can fill in that entire row. After that there is a column you can fill in, which allows you to scatter some dots along the edges. If I remember right, I then noticed that there were only two possibilities for the 4 in the top row, and either-or logic on that got me a bit further. In the end I made it all the way through, but I felt like I was doing 3-level look-ahead at several points. That kind of works out OK because this puzzle is so small and the pattern of clues so simple that you can do more in your head than usual, but I admit I had to make some temporary marks on the screen and undo them to really convince myself that what I thought I was seeing was really there.

Jan, I now think I get the idea of what is acceptable level of analysis for my puzzles not to get a question-mark: For me, it's when I have to make significant use of the "undo" button, i.e. I have to do more than a minimal amount of backtracking.

Can't tell you how much I enjoy this site!

Yes. Obviously any puzzle with a unique solution is logically solvable if you are willing to allow sufficiently long and complex chains of reasoning. But clearly there is some point where that ceases to be fun. This point is of course different for different people.

One of the nice things about having a community component to this web site is that it has allowed us to evolve a local consensus as to what is and is not "unreasonably hard". We've standardized a few techniques that go beyond standard line solving, like edge logic, smile logic, and two-way logic. (Or did I call that either-or logic? I can never remember.) Hints in the comments help people unfamiliar with those ideas get up to speed on them, so that puzzle requiring that level of complexity are considered "normal" here, though some places won't publish a puzzle that isn't line-solvable.

So there is no automatic right answer for what is and is not acceptable, only a local consensus that holds within a puzzle solving community. I always want this site to allow people to push the envelope, because sometimes they can take us along with them into whole new realms. That's happening a lot over the years. If you look at the comments on the early puzzles on this site, you'll see people calling puzzles "hard" that by current standards aren't really very hard. We've collectively become much better solvers, and that impacts the design of puzzles, since it encourages the posting of harder puzzles.

once i got all of C4, all of R9 and 3 of the 4 in R2...and dots in C1,9,10 of the odd numbered rows, here is what i did:

Can all three 1's in C4 fit below the dot in R2? If those are blacks then you get dots in R3,5,7 of C8 which leaves nowhere for the middle 2. That means they can't all pack in under R2, making C4R1 a black. Line logic to end.

Looks kind of like the roll of paper coming off a seismograph.

Found to be logically solvable by gator.

David - good find. I agree that this is not looking too far ahead (and I love logical constraints).

Ditto #8, exactly

Jan, you really are missed!

Goto next topic

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