Evil influence? Only for your productivity.
January 18, 2012 8:35 AM   Subscribe

Masyu, also known as Pearls, is an NP-complete logic puzzle created by the makers of Sudoku. Brandon McPhail provides a few free puzzles to get your feet wet on his web site (Java applet). Once you've mastered those, UCLICK Games offers a free daily puzzle (Flash) with the past month of archives available too.

If you have an iOS device, TootSweet's free Masyu app includes a large number of puzzles. The company also makes an easy collection and a challenging one, both free as well.
posted by The Winsome Parker Lewis (28 comments total) 20 users marked this as a favorite

 
The first link (to Wikipedia) is unavailable today as part of the SOPA blackout. The article is mirrored here for your convenience. It contains a good summary of the rules and history of Masyu.
posted by The Winsome Parker Lewis at 8:35 AM on January 18, 2012


Evil influence, indeed! This game is totally going to SOPA any hope I had of productivity today.
posted by jph at 8:50 AM on January 18, 2012


I kinda wish more of these Nikoli games were in the sgt-puzzles collection.

While Masyu might be NP-complete, it seems like it's the only puzzle I can actually solve on the World Puzzle Championship tests.
posted by pwnguin at 8:57 AM on January 18, 2012 [1 favorite]


I kinda wish more of these Nikoli games were in the sgt-puzzles collection.

Yeah, when I went on a Masyu binge earlier this week I fantasized about coding one up, but I'm not quite technically proficient.
posted by muddgirl at 9:00 AM on January 18, 2012


Or rather I can do the coding, but the theory behind puzzle generation is a bit beyond me.

One 'issue' with Masyu puzzles is that solutions aren't necessarily unique.
posted by muddgirl at 9:00 AM on January 18, 2012


This is great!

One 'issue' with Masyu puzzles is that solutions aren't necessarily unique.

I'd noticed that the exact placement of lines doesn't always form a unique solution, but I haven't yet come across any that would allow multiple options for the order in which the pearls are connected. (I've only played a handful so far.)
posted by nobody at 9:12 AM on January 18, 2012


Never mind. "Big X" on the java applet has rotational symmetry, so I guess that's allowed, too.
posted by nobody at 9:28 AM on January 18, 2012


The first link (to Wikipedia) is unavailable today as part of the SOPA blackout.

Isn't the solution to the puzzle given in the (mirrored) link incorrect? There are two black cells in a row, so the requirement that "Black circles must be turned upon, but the loop must travel straight through the next and previous cells in its path" is violated: for each of them, the loop fails to travel straight through either the next or the previous cell it visits.
posted by kenko at 10:28 AM on January 18, 2012


I don't see any errors in this solution. Which set of black circles are you concerned about?
posted by muddgirl at 10:36 AM on January 18, 2012


kenko: I'm not sure I understand. There are no places in that puzzle where two black circles are in a row (not consecutively, at least). I double-checked the solution on that page and it's completely valid.

The rules as described on that page aren't very clear. I think this page does a better job explaining things, with illustrations.
posted by The Winsome Parker Lewis at 10:37 AM on January 18, 2012


The term "cells" includes blank spaces. The black and white circles occupy the cells; they aren't cells themselves.
posted by wanderingmind at 10:37 AM on January 18, 2012


I have to confess that I was terribly confused by the Masyu rules for several hours the first time I played.
posted by muddgirl at 10:51 AM on January 18, 2012


Oh god... this is the sort of game I love until I hate it. And then keep playing it. And the resort to writing a script to solve it faster than I can so that actually doing it myself seems like a waste of time.

Sadly, doing the same for Plants vs Zombies is beyond my programming skills...
posted by sodium lights the horizon at 10:58 AM on January 18, 2012


Fascinating. This game annoys me in precisely the same way that Sudoku and Picross do. Requires the same sort of patient deduction, I guess.
posted by NMcCoy at 10:59 AM on January 18, 2012


well damn, these are fantastic. thank you!
posted by peachfuzz at 11:14 AM on January 18, 2012


Oh, I love all these 'not sudoku' puzzles. So much more interesting than vanilla Sudoku, whose popularity I find inexplicable. Masyu, Nurikabe, and Hidato are all so much more fun! Masyu puzzles were frustrating at first because it took a long time before I was able to 'see' the connections.
posted by word_virus at 11:14 AM on January 18, 2012


Good. Now could someone explain the rules of Matisyahu?
posted by Mr. Bad Example at 12:01 PM on January 18, 2012


NMcCoy: "Requires the same sort of patient deduction, I guess."

Eh, actually, once you figure out the deduction heuristics (two white circles on the board edge form a halfpipe) to these sorts of games, they get pretty easy. NP-completeness is technically true but local reasoning is often still applicable for these games so the logic chain rarely gets huge.
posted by pwnguin at 1:18 PM on January 18, 2012


this is the sort of game I love until I hate it

Me too!

And then keep playing it.

Me too, LOL!

And then resort to writing a script to solve it faster than I can


................you win.
posted by Ian A.T. at 1:22 PM on January 18, 2012


Has anyone solved "mathgrant's big one" on the java applet? I keep running into a couple of problem spots near the end. Either it has no solution, or I'm making an unwarranted assumption somewhere along the line.
posted by DevilsAdvocate at 6:03 PM on January 18, 2012


It has a solution. If still hve it up if you want to take a screengrab of where you are and I can give you a hint.
posted by muddgirl at 6:45 PM on January 18, 2012


I can't type. I still have the screen open so I can check what you have.
posted by muddgirl at 6:45 PM on January 18, 2012


I got it, finally. It (meaning "mathgrant's big one" specifically) is fascinating in that in one sense, it's actually quite easy in that just about everything is forced. At the same time, it's (obviously) quite hard in that if you do make a mistake by making what you think is a forced move that isn't, it's incredibly difficult to figure out just where you went wrong.

Thanks for the offer of help!
posted by DevilsAdvocate at 8:18 PM on January 18, 2012


Can we all at least agree that 日本式 needs to go fuck itself?
Hepburn represent?
No?
Oh...okay. :{
posted by GoingToShopping at 11:13 PM on January 18, 2012


"the makers of Sudoku"? a maker of Sudoku, sure, but if you are implying they invented sudoku, you are incorrect.
posted by BurnChao at 2:50 AM on January 19, 2012


Thanks for the correction, BurnChao. I'm not fully up to speed on my puzzle designers/publishers. Probably should've researched that one a bit more.
posted by The Winsome Parker Lewis at 6:46 AM on January 19, 2012


Sure, no problem.
posted by BurnChao at 8:55 PM on January 20, 2012


I am now officially addicted to this. It took me a few puzzles to see how those three simple rules (white rules, black rules, path rule) form all the logic necessary to complete large puzzles, and I don't need to look at the whole thing and attempt to plan the big-picture of the path.

This is awesome. It's like one-player chess.

Thanks!
posted by MustardTent at 6:15 AM on January 23, 2012


« Older Stop the Wall - A group of NYC based Internet comp...  |  The rise and fall of personal ... Newer »


This thread has been archived and is closed to new comments