Finnish Snakes

Here’s a pair of Finnish Snakes. One puzzle’s shaded givens (dark gray cells) are the others’ blank givens (light gray cells), and vice versa (ignoring the head and tail). Sticking to the symmetric layout and duality restrictions probably made both puzzles a little worse than what they could’ve been alone, though. I also couldn’t get the lengths match up.

I was going to use white/black circles instead of coloured cells, but couldn’t bother in the end.

Finnish Snake 1

Finnish Snake 2a


  • Shade some cells to form a snake.
  • The snake must use all dark gray cells, and none of the light gray cells.
  • The head and tail (numbered cells) are given, and indicate the snake’s length (49 and 47 cells long, respectively).
  • The snake doesn’t touch itself, even by corners.

EDIT: On the second snake, fixed the clue at R1C5.


Today’s puzzle is a Fillomino variant. Yet again I don’t know the creator of this puzzle type, or if the name matches the original. If you know, please tell me in the comments, etc. (Edit: Thanks!)

I quite like this variation, there’s something satisfying in how it tends to flow nicely.

Tatamino 1


  • Divide the grid into polyominoes of size 1, 2 or 3.
  • Polyominoes of the same size may not share an edge.
  • A clue tells the size of the polyomino the cell belongs in.

For a harder version, remove the clues from R2C5 and R4C7.

TSLI / Heyawake

I’m not sure if today’s puzzle enritely qualifies as a double, since one puzzle uses numbers while the other doesn’t.

TSLI is a LITS variant. TSLI is to LITS as Pata is to Tapa, hence the naming.

I actually made the TSLI first, and as an afterthought decided to make a Heyawake with the same grid. It’s the first Heyawake I’ve made, and considering how much trouble I have solving Heyawakes, it turned out surprisingly good.



TSLI rules:

  • Shade some cells to form a continuous wall that contains no 2×2 squares of shaded cells.
  • Each region contains exactly four unshaded cells, and they must form an L, I, T or S tetromino. Tetrominoes of the same shape may not share an edge.


Heyawake 1

Heyawake rules:

  • A region’s clue tells exactly how many shaded cells it contains. Unclued regions may have any number of shaded cells (including zero).
  • Two shaded cells may not share an edge.
  • The unshaded cells must form a single, continuous polyomino.
  • No continuous line of unshaded cells may pass over two or more region borders.
  • To clarify the last rule, the scenario below is illegal, but the one below it legal.



EDIT: Turns out both puzzles were broken, both had slight ambiguities. That’s what happens when I make puzzles while delirious with fever, and have no test-solvers to boot…

The fix isn’t elegant, but I like the rest of the TSLI too much to completely retool the lower left corner. The Heyawake was reclued too.

(The dot means that the cell is unshaded.)

Outside Nurikabe

This is the sibling to today’s Outside Nurikabe on the GMPuzzles blog. It’s harder, but shouldn’t be as bifurcation-heavy as the first one was.

Outside Clue Nurikabe 2


  • Unlike in regular Nurikabe, the clue numbers have been placed outside the grid.
  • The clues will appear in the first unshaded cell in the clue’s row/column, counted from the clue’s direction.
  • Standard Nurikabe rules then apply:
    • Shade some cells to form a continuous wall that contains no 2×2 squares of shaded cells.
    • The clue numbers hint at connected groups of unshaded cells.
    • Every group has exactly one clue, which tells the group’s size.


I’m not sure who came up with this puzzle type or what it was originally called. If someone knows, please let me know too.

Neighbors 3

Neighbors 2

Neighbors 1


  • Divide the grid into connected regions (polyminoes) of equal size.
  • Each region has exactly one clue.
  • A clue tells how many regions the clue’s region touches (shares an edge with).


Here’s a Pata.

Pata 1


  • Shade some cells to form a continuous wall that contains no 2×2 squares.
  • The clue numbers tell the size of continuous groups of blank cells in the eight cells around the clues. Two groups must have at least one shaded cell between them. The clue cells themselved are blank.

Treasure Field

A friend came up with the rules, and I made this.

Treasure Field 1


  • Place a treasure chest in every row and column. Two chests may not share an edge, but they may touch by corners.
  • A clue with an arrow points to the direction of the nearest chest . Two arrows mean there are two chests equidistant to the clue in the directions the two arrows point to.
  • The distance metric used is Manhattan distance.
  • Additionally, the clues are unambiguous: from the clue’s perspective, the nearest chest can’t lie on a diagonal (or the clue itself).

To help clarify the rules, here’s the solution to the puzzle my friend made, with the clue arrows colour-coded according to which chest they hint at.

The shaded arrowless cells in the corners are empty, i.e. don’t contain treasure.

Double Back

Today’s puzzle is a Double Back. The size of almost all regions is a multiple of three, and the one exception to the rule looks like the number three. The colouring is for aesthetical purposes.

Double Back 1


  • Draw a single, non-intersecting loop that visits every cell exactly once, and every region exactly twice.


Here’s an easy/medium LITS. Getting the puzzle to wrap up was the hardest part.



  • Shade some cells to form a tetromino in each region.
  • Two tetrominoes of the same shape may not share an edge.
  • The shaded cells must also form a connected wall that contains no 2×2 squares of shaded cells.

EDIT: I modified two of the central regions, the puzzle solves more smoothly now.

Alternating Masyu

Today’s puzzle might be somewhat vulnerable to metalogic due to the sparse clues, but I think the actual logic involved is pretty interesting.

Alternating Masyu 1


  • Draw a single non-intersecting loop that runs through the centers of the cells and visits all pearls.
  • On black pearls the loop must make a right turn, and the loop must go straight through the cells before and after the pearl.
  • On white pearls the loop goes straight through, and the loop must make a right turn on at least one of the cells before and after the pearl.
  • Or, see Masyu rules.
  • Additionally, the loop must alternate between visiting black and white pearls: it can’t visit two pearls of the same colour in a row.