Forums

Number: Puzzle #1932
Genre: Masyu
Author: Bram
Appeared at: August 3, 2010
Nice puzzle. I got stuck at some point and used a sort of "intuitive" uniqueness along the lines of "why so much empty space in lower left? no way to fill this uniquely with lines so it must remain empty... why would it? a-ha..." and then it took me a couple seconds to finish the diagram.

Was that your idea or did you have some other way in mind?

Marcin

Interesting, given that a topic about using uniqueness as a clue just resurfaced yesterday on the general forum.

This puzzle has another 'uniqueness clue' somewhere that's very blatant and hard to ignore while solving. It feels kind of stupid to look for a more 'fair' leap of logic when a uniqueness clue is staring you in the face.

Consider me interested in Mucha's closing question as well.

It wasn't exactly the plan to give a uniqueness move directly, but I like Masyus that have unused space, which at first doesn't exactly seem logical that you can't go there.

It was more a way for me to close the loop in a nice way though. The two white balls in row 6 are the last ones I placed. If you remove those 2 balls, the path through the remaining balls is still unique. Not the loop. But all info you can get from the remaining balls will get one path that ends in an open area on the left bottom. I closed the loop with those 2 balls.

I never think too much about avoiding uniqueness moves though. My fences can usually have some uniqueness shortcuts too. But if I'm happy with the non-uniqueness logical path, I'll just keep it in.

Here it is a kinda backbreakign one, where it's really hard to avoid that area. and the way to do it needs some thinking, so I don't mind it opening up the puzzle.

More out curiosity, you guys think it would have been better if I had avoided that uniqueness move by closing the loop through placing white balls on R6C3, R7C1 and R8C3 instead of R6C1 and R6C3?

That would make it a little easier to solve through internal logic. (Or maybe it doesn't, it's tricky to judge having already solved it.)

But I do think that's preferable over having it at about the same difficulty by allowing for the meta-logic of uniqueness. It makes for a more 'pure' puzzle, but I guess that's a matter of taste.

By the way, the uniqueness clue I noticed is in a different spot and it showed when I was filling in some trivial starting moves. It probably only shows when doing things in a certain order. I'll describe it in white text:

Spoiler ->After marking the line going from R1C5 to R1C9, and the two downward line segments from R2C6 and R1C9, you can tell that R4C7 and R4C8 can't be crossed by vertical lines because there's 2 ways to link them together in the R2C7-R3C8 box.<- End spoiler

It's a totally unnecessary clue of course, but like I said in my previous post, if it's staring me in the face, it's hard to ignore.

Bram wrote:

More out curiosity, you guys think it would have been better if I had avoided that uniqueness move by closing the loop through placing white balls on R6C3, R7C1 and R8C3 instead of R6C1 and R6C3?

It's a matter of taste imo. I liked it the way it was, and was just wondering what your intentions were when creating it.

Marcin

Forseti, I don't agree with you on that
Explanation> The C8 vertical line would extend into the R7C8 black ball and thus has to turn in R3C8. <End Explanation

That's true of course.

->At the time I spotted it though, I had my eyes fixed on the top right quadrant of the puzzle trying to get all I could from that cluster of balls. I hadn't done anything with the bottom right yet so the 2 line segments coming up from R7C8 weren't filled in at that time. The uniquess issue topside happened to be more obvious to me at the time than the collision waiting to happen downwards.<-

It's a very, very minor clue, and, like I posted earlier, it only shows when doing things in a very specific order. It also becomes completely unnecessary about 5 seconds later. It certainly doesn't take away from the puzzle at all.