This is a collection of small one-player puzzle games.
This manual is copyright 2004-2010 Simon Tatham. All rights reserved. You may distribute this documentation under the MIT licence. See appendix A for the licence text in full.
I wrote this collection because I thought there should be more small desktop toys available: little games you can pop up in a window and play for two or three minutes while you take a break from whatever else you were doing. And I was also annoyed that every time I found a good game on (say) Unix, it wasn't available the next time I was sitting at a Windows machine, or vice versa; so I arranged that everything in my personal puzzle collection will happily run on both, and have more recently done a port to Mac OS X as well. When I find (or perhaps invent) further puzzle games that I like, they'll be added to this collection and will immediately be available on both platforms. And if anyone feels like writing any other front ends – PocketPC, Mac OS pre-10, or whatever it might be – then all the games in this framework will immediately become available on another platform as well.
The actual games in this collection were mostly not my invention; they are re-implementations of existing game concepts within my portable puzzle framework. I do not claim credit, in general, for inventing the rules of any of these puzzles. (I don't even claim authorship of all the code; some of the puzzles have been submitted by other authors.)
This collection is distributed under the MIT licence (see appendix A). This means that you can do pretty much anything you like with the game binaries or the code, except pretending you wrote them yourself, or suing me if anything goes wrong.
The most recent versions, and source code, can be found at http://www.chiark.greenend.org.uk/~sgtatham/puzzles/
.
Please report bugs to anakin@pobox.com
. You might find it helpful to read this article before reporting a bug:
http://www.chiark.greenend.org.uk/~sgtatham/bugs.html
Patches are welcome. Especially if they provide a new front end (to make all these games run on another platform), or a new game.
This chapter describes features that are common to all the games.
These actions are all available from the ‘Game’ menu and via keyboard shortcuts, in addition to any game-specific actions.
(On Mac OS X, to conform with local user interface standards, these actions are situated on the ‘File’ and ‘Edit’ menus instead.)
The Load and Save operations preserve your entire game history (so you can save, reload, and still Undo and Redo things you had done before saving).
Some games (such as Solo) are capable of solving a game ID you have typed in from elsewhere. Other games (such as Rectangles) cannot solve a game ID they didn't invent themself, but when they did invent the game ID they know what the solution is already. Still other games (Pattern) can solve some external game IDs, but only if they aren't too difficult.
The ‘Solve’ command adds the solved state to the end of the undo chain for the puzzle. In other words, if you want to go back to solving it yourself after seeing the answer, you can just press Undo.
There are two ways to save a game specification out of a puzzle and recreate it later, or recreate it in somebody else's copy of the same puzzle.
The ‘Specific’ and ‘Random Seed’ options from the ‘Game’ menu (or the ‘File’ menu, on Mac OS X) each show a piece of text (a ‘game ID’) which is sufficient to reconstruct precisely the same game at a later date.
You can enter either of these pieces of text back into the program (via the same ‘Specific’ or ‘Random Seed’ menu options) at a later point, and it will recreate the same game. You can also use either one as a command line argument (on Windows or Unix); see section 2.4 for more detail.
The difference between the two forms is that a descriptive game ID is a literal description of the initial state of the game, whereas a random seed is just a piece of arbitrary text which was provided as input to the random number generator used to create the puzzle. This means that:
(Use the ‘About’ menu option to find out the version number of the program. Programs with the same version number running on different platforms should still be random-seed compatible.)
A descriptive game ID starts with a piece of text which encodes the parameters of the current game (such as grid size). Then there is a colon, and after that is the description of the game's initial state. A random seed starts with a similar string of parameters, but then it contains a hash sign followed by arbitrary data.
If you enter a descriptive game ID, the program will not be able to show you the random seed which generated it, since it wasn't generated from a random seed. If you enter a random seed, however, the program will be able to show you the descriptive game ID derived from that random seed.
Note that the game parameter strings are not always identical between the two forms. For some games, there will be parameter data provided with the random seed which is not included in the descriptive game ID. This is because that parameter information is only relevant when generating puzzle grids, and is not important when playing them. Thus, for example, the difficulty level in Solo (chapter 11) is not mentioned in the descriptive game ID.
These additional parameters are also not set permanently if you type in a game ID. For example, suppose you have Solo set to ‘Advanced’ difficulty level, and then a friend wants your help with a ‘Trivial’ puzzle; so the friend reads out a random seed specifying ‘Trivial’ difficulty, and you type it in. The program will generate you the same ‘Trivial’ grid which your friend was having trouble with, but once you have finished playing it, when you ask for a new game it will automatically go back to the ‘Advanced’ difficulty which it was previously set on.
The ‘Type’ menu, if present, may contain a list of preset game settings. Selecting one of these will start a new random game with the parameters specified.
The ‘Type’ menu may also contain a ‘Custom’ option which allows you to fine-tune game parameters. The parameters available are specific to each game and are described in the following sections.
(This section does not apply to the Mac OS X version.)
The games in this collection deliberately do not ever save information on to the computer they run on: they have no high score tables and no saved preferences. (This is because I expect at least some people to play them at work, and those people will probably appreciate leaving as little evidence as possible!)
However, if you do want to arrange for one of these games to default to a particular set of parameters, you can specify them on the command line.
The easiest way to do this is to set up the parameters you want using the ‘Type’ menu (see section 2.3), and then to select ‘Random Seed’ from the ‘Game’ or ‘File’ menu (see section 2.2). The text in the ‘Game ID’ box will be composed of two parts, separated by a hash. The first of these parts represents the game parameters (the size of the playing area, for example, and anything else you set using the ‘Type’ menu).
If you run the game with just that parameter text on the command line, it will start up with the settings you specified.
For example: if you run Cube (see chapter 4), select ‘Octahedron’ from the ‘Type’ menu, and then go to the game ID selection, you will see a string of the form ‘o2x2#338686542711620
’. Take only the part before the hash (‘o2x2
’), and start Cube with that text on the command line: ‘cube o2x2
’.
If you copy the entire game ID on to the command line, the game will start up in the specific game that was described. This is occasionally a more convenient way to start a particular game ID than by pasting it into the game ID selection box.
(You could also retrieve the encoded game parameters using the ‘Specific’ menu option instead of ‘Random Seed’, but if you do then some options, such as the difficulty level in Solo, will be missing. See section 2.2 for more details on this.)
(This section only applies to the Unix port.)
In addition to being able to specify game parameters on the command line (see section 2.4), there are various other options:
--game
--load
--generate
n
If game parameters are specified on the command-line, they will be used to generate the game IDs; otherwise a default set of parameters will be used.
The most common use of this option is in conjunction with --print
, in which case its behaviour is slightly different; see below.
--print
wx
h
On each page of puzzles, there will be w across and h down. If there are more puzzles than w×h, more than one page will be printed.
If --generate
has also been specified, the invented game IDs will be used to generate the printed output. Otherwise, a list of game IDs is expected on standard input (which can be descriptive or random seeds; see section 2.2), in the same format produced by --generate
.
For example:
net --generate 12 --print 2x3 7x7w | lpr
will generate two pages of printed Net puzzles (each of which will have a 7×7 wrapping grid), and pipe the output to the lpr
command, which on many systems will send them to an actual printer.
There are various other options which affect printing; see below.
--save
file-prefix [ --save-suffix
file-suffix ]
If --generate
has also been specified, the invented game IDs will be used to generate the printed output. Otherwise, a list of game IDs is expected on standard input (which can be descriptive or random seeds; see section 2.2), in the same format produced by --generate
.
For example:
net --generate 12 --save game --save-suffix .sav
will generate twelve Net saved-game files with the names game0.sav
to game11.sav
.
--version
The following options are only meaningful if --print
is also specified:
--with-solutions
--scale
n
--colour
(Note: the Windows version of this game is called NETGAME.EXE
to avoid clashing with Windows's own NET.EXE
.)
I originally saw this in the form of a Flash game called FreeNet [1], written by Pavils Jurjans; there are several other implementations under the name NetWalk. The computer prepares a network by connecting up the centres of squares in a grid, and then shuffles the network by rotating every tile randomly. Your job is to rotate it all back into place. The successful solution will be an entirely connected network, with no closed loops. As a visual aid, all tiles which are connected to the one in the middle are highlighted.
[1] http://www.jurjans.lv/stuff/net/FreeNet.htm
This game can be played with either the keyboard or the mouse. The controls are:
The following controls are not necessary to complete the game, but may be useful:
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
The grid generation in Net has been carefully arranged so that the barriers are independent of the rest of the grid. This means that if you note down the random seed used to generate the current puzzle (see section 2.2), change the Barrier probability parameter, and then re-enter the same random seed, you should see exactly the same starting grid, with the only change being the number of barriers. So if you're stuck on a particular grid and need a hint, you could start up another instance of Net, set up the same parameters but a higher barrier probability, and enter the game seed from the original Net window.
This is another one I originally saw as a web game. This one was a Java game [2], by Paul Scott. You have a grid of 16 squares, six of which are blue; on one square rests a cube. Your move is to use the arrow keys to roll the cube through 90 degrees so that it moves to an adjacent square. If you roll the cube on to a blue square, the blue square is picked up on one face of the cube; if you roll a blue face of the cube on to a non-blue square, the blueness is put down again. (In general, whenever you roll the cube, the two faces that come into contact swap colours.) Your job is to get all six blue squares on to the six faces of the cube at the same time. Count your moves and try to do it in as few as possible.
Unlike the original Java game, my version has an additional feature: once you've mastered the game with a cube rolling on a square grid, you can change to a triangular grid and roll any of a tetrahedron, an octahedron or an icosahedron.
[2] http://www3.sympatico.ca/paulscott/cube/cube.htm
This game can be played with either the keyboard or the mouse.
Left-clicking anywhere on the window will move the cube (or other solid) towards the mouse pointer.
The arrow keys can also used to roll the cube on its square grid in the four cardinal directions. On the triangular grids, the mapping of arrow keys to directions is more approximate. Vertical movement is disallowed where it doesn't make sense. The four keys surrounding the arrow keys on the numeric keypad (‘7’, ‘9’, ‘1’, ‘3’) can be used for diagonal movement.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
The old ones are the best: this is the good old ‘15-puzzle’ with sliding tiles. You have a 4×4 square grid; 15 squares contain numbered tiles, and the sixteenth is empty. Your move is to choose a tile next to the empty space, and slide it into the space. The aim is to end up with the tiles in numerical order, with the space in the bottom right (so that the top row reads 1,2,3,4 and the bottom row reads 13,14,15,space).
This game can be controlled with the mouse or the keyboard.
A left-click with the mouse in the row or column containing the empty space will move as many tiles as necessary to move the space to the mouse pointer.
The arrow keys will move a tile adjacent to the space in the direction indicated (moving the space in the opposite direction).
(All the actions described in section 2.1 are also available.)
The only options available from the ‘Custom...’ option on the ‘Type’ menu are Width and Height, which are self-explanatory. (Once you've changed these, it's not a ‘15-puzzle’ any more, of course!)
Another sliding tile puzzle, visually similar to Fifteen (see chapter 5) but with a different type of move. This time, there is no hole: all 16 squares on the grid contain numbered squares. Your move is to shift an entire row left or right, or shift an entire column up or down; every time you do that, the tile you shift off the grid re-appears at the other end of the same row, in the space you just vacated. To win, arrange the tiles into numerical order (1,2,3,4 on the top row, 13,14,15,16 on the bottom). When you've done that, try playing on different sizes of grid.
I might have invented this game myself, though only by accident if so (and I'm sure other people have independently invented it). I thought I was imitating a screensaver I'd seen, but I have a feeling that the screensaver might actually have been a Fifteen-type puzzle rather than this slightly different kind. So this might be the one thing in my puzzle collection which represents creativity on my part rather than just engineering.
Left-clicking on an arrow will move the appropriate row or column in the direction indicated. Right-clicking will move it in the opposite direction.
Alternatively, use the cursor keys to move the position indicator around the edge of the grid, and use the return key to move the row/column in the direction indicated.
(All the actions described in section 2.1 are also available.)
The parameters available from the ‘Custom...’ option on the ‘Type’ menu are:
Twiddle is a tile-rearrangement puzzle, visually similar to Sixteen (see chapter 6): you are given a grid of square tiles, each containing a number, and your aim is to arrange the numbers into ascending order.
In basic Twiddle, your move is to rotate a square group of four tiles about their common centre. (Orientation is not significant in the basic puzzle, although you can select it.) On more advanced settings, you can rotate a larger square group of tiles.
I first saw this type of puzzle in the GameCube game ‘Metroid Prime 2’. In the Main Gyro Chamber in that game, there is a puzzle you solve to unlock a door, which is a special case of Twiddle. I developed this game as a generalisation of that puzzle.
To play Twiddle, click the mouse in the centre of the square group you wish to rotate. In the basic mode, you rotate a 2×2 square, which means you have to click at a corner point where four tiles meet.
In more advanced modes you might be rotating 3×3 or even more at a time; if the size of the square is odd then you simply click in the centre tile of the square you want to rotate.
Clicking with the left mouse button rotates the group anticlockwise. Clicking with the right button rotates it clockwise.
You can also move an outline square around the grid with the cursor keys; the square is the size above (2×2 by default, or larger). Pressing the return key or space bar will rotate the current square anticlockwise or clockwise respectively.
(All the actions described in section 2.1 are also available.)
Twiddle provides several configuration options via the ‘Custom’ option on the ‘Type’ menu:
You have a grid of squares, with numbers written in some (but not all) of the squares. Your task is to subdivide the grid into rectangles of various sizes, such that (a) every rectangle contains exactly one numbered square, and (b) the area of each rectangle is equal to the number written in its numbered square.
Credit for this game goes to the Japanese puzzle magazine Nikoli [3]; I've also seen a Palm implementation at Puzzle Palace [4]. Unlike Puzzle Palace's implementation, my version automatically generates random grids of any size you like. The quality of puzzle design is therefore not quite as good as hand-crafted puzzles would be, but on the plus side you get an inexhaustible supply of puzzles tailored to your own specification.
[3] http://www.nikoli.co.jp/puzzles/7/index_text-e.htm
[4] http://www.puzzle.gr.jp/puzzle/sikaku/palm/index.html.en
This game is played with the mouse or cursor keys.
Left-click any edge to toggle it on or off, or left-click and drag to draw an entire rectangle (or line) on the grid in one go (removing any existing edges within that rectangle). Right-clicking and dragging will allow you to erase the contents of a rectangle without affecting its edges.
Alternatively, use the cursor keys to move the position indicator around the board. Pressing the return key then allows you to use the cursor keys to drag a rectangle out from that position, and pressing the return key again completes the rectangle. Using the space bar instead of the return key allows you to erase the contents of a rectangle without affecting its edges, as above.
When a rectangle of the correct size is completed, it will be shaded.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
The default expansion factor of zero means that Rectangles will simply generate a grid of the size you ask for, and do nothing further. If you set an expansion factor of (say) 0.5, it means that each dimension of the grid will be expanded to half again as big after generation. In other words, the initial grid will be 2/3 the size in each dimension, and will be expanded to its full size without adding any more rectangles.
Setting an expansion factor of around 0.5 tends to make the game more difficult, and also (in my experience) rewards a less deductive and more intuitive playing style. If you set it too high, though, the game simply cannot generate more than a few rectangles to cover the entire grid, and the game becomes trivial.
This game combines the grid generation of Net (see chapter 3) with the movement of Sixteen (see chapter 6): you have a Net grid, but instead of rotating tiles back into place you have to slide them into place by moving a whole row at a time.
As in Sixteen, control is with the mouse or cursor keys. See section 6.1.
The available game parameters have similar meanings to those in Net (see section 3.2) and Sixteen (see section 6.2).
Netslide was contributed to this collection by Richard Boulton.
You have a grid of squares, which must all be filled in either black or white. Beside each row of the grid are listed the lengths of the runs of black squares on that row; above each column are listed the lengths of the runs of black squares in that column. Your aim is to fill in the entire grid black or white.
I first saw this puzzle form around 1995, under the name ‘nonograms’. I've seen it in various places since then, under different names.
Normally, puzzles of this type turn out to be a meaningful picture of something once you've solved them. However, since this version generates the puzzles automatically, they will just look like random groupings of squares. (One user has suggested that this is actually a good thing, since it prevents you from guessing the colour of squares based on the picture, and forces you to use logic instead.) The advantage, though, is that you never run out of them.
This game is played with the mouse.
Left-click in a square to colour it black. Right-click to colour it white. If you make a mistake, you can middle-click, or hold down Shift while clicking with any button, to colour the square in the default grey (meaning ‘undecided’) again.
You can click and drag with the left or right mouse button to colour a vertical or horizontal line of squares black or white at a time (respectively). If you click and drag with the middle button, or with Shift held down, you can colour a whole rectangle of squares grey.
You can also move around the grid with the cursor keys. Pressing the return key will cycle the current cell through empty, then black, then white, then empty, and the space bar does the same cycle in reverse.
(All the actions described in section 2.1 are also available.)
The only options available from the ‘Custom...’ option on the ‘Type’ menu are Width and Height, which are self-explanatory.
You have a square grid, which is divided into as many equally sized sub-blocks as the grid has rows. Each square must be filled in with a digit from 1 to the size of the grid, in such a way that
You are given some of the numbers as clues; your aim is to place the rest of the numbers correctly.
Under the default settings, the sub-blocks are square or rectangular. The default puzzle size is 3×3 (a 9×9 actual grid, divided into nine 3×3 blocks). You can also select sizes with rectangular blocks instead of square ones, such as 2×3 (a 6×6 grid divided into six 3×2 blocks). Alternatively, you can select ‘jigsaw’ mode, in which the sub-blocks are arbitrary shapes which differ between individual puzzles.
Another available mode is ‘killer’. In this mode, clues are not given in the form of filled-in squares; instead, the grid is divided into ‘cages’ by coloured lines, and for each cage the game tells you what the sum of all the digits in that cage should be. Also, no digit may appear more than once within a cage, even if the cage crosses the boundaries of existing regions.
If you select a puzzle size which requires more than 9 digits, the additional digits will be letters of the alphabet. For example, if you select 3×4 then the digits which go in your grid will be 1 to 9, plus ‘a
’, ‘b
’ and ‘c
’. This cannot be selected for killer puzzles.
I first saw this puzzle in Nikoli [5], although it's also been popularised by various newspapers under the name ‘Sudoku’ or ‘Su Doku’. Howard Garns is considered the inventor of the modern form of the puzzle, and it was first published in Dell Pencil Puzzles and Word Games. A more elaborate treatment of the history of the puzzle can be found on Wikipedia [6].
[5] http://www.nikoli.co.jp/puzzles/1/index_text-e.htm
[6] http://en.wikipedia.org/wiki/Sudoku
To play Solo, simply click the mouse in any empty square and then type a digit or letter on the keyboard to fill that square. If you make a mistake, click the mouse in the incorrect square and press Space to clear it again (or use the Undo feature).
If you right-click in a square and then type a number, that number will be entered in the square as a ‘pencil mark’. You can have pencil marks for multiple numbers in the same square. Squares containing filled-in numbers cannot also contain pencil marks.
The game pays no attention to pencil marks, so exactly what you use them for is up to you: you can use them as reminders that a particular square needs to be re-examined once you know more about a particular number, or you can use them as lists of the possible numbers in a given square, or anything else you feel like.
To erase a single pencil mark, right-click in the square and type the same number again.
All pencil marks in a square are erased when you left-click and type a number, or when you left-click and press space. Right-clicking and pressing space will also erase pencil marks.
Alternatively, use the cursor keys to move the mark around the grid. Pressing the return key toggles the mark (from a normal mark to a pencil mark), and typing a number in is entered in the square in the appropriate way; typing in a 0 or using the space bar will clear a filled square.
(All the actions described in section 2.1 are also available.)
Solo allows you to configure two separate dimensions of the puzzle grid on the ‘Type’ menu: the number of columns, and the number of rows, into which the main grid is divided. (The size of a block is the inverse of this: for example, if you select 2 columns and 3 rows, each actual block will have 3 columns and 2 rows.)
If you tick the ‘X’ checkbox, Solo will apply the optional extra constraint that the two main diagonals of the grid also contain one of every digit. (This is sometimes known as ‘Sudoku-X’ in newspapers.) In this mode, the squares on the two main diagonals will be shaded slightly so that you know it's enabled.
If you tick the ‘Jigsaw’ checkbox, Solo will generate randomly shaped sub-blocks. In this mode, the actual grid size will be taken to be the product of the numbers entered in the ‘Columns’ and ‘Rows’ boxes. There is no reason why you have to enter a number greater than 1 in both boxes; Jigsaw mode has no constraint on the grid size, and it can even be a prime number if you feel like it.
If you tick the ‘Killer’ checkbox, Solo will generate a set of of cages, which are randomly shaped and drawn in an outline of a different colour. Each of these regions contains a smaller clue which shows the digit sum of all the squares in this region.
You can also configure the type of symmetry shown in the generated puzzles. More symmetry makes the puzzles look prettier but may also make them easier, since the symmetry constraints can force more clues than necessary to be present. Completely asymmetric puzzles have the freedom to contain as few clues as possible.
Finally, you can configure the difficulty of the generated puzzles. Difficulty levels are judged by the complexity of the techniques of deduction required to solve the puzzle: each level requires a mode of reasoning which was not necessary in the previous one. In particular, on difficulty levels ‘Trivial’ and ‘Basic’ there will be a square you can fill in with a single number at all times, whereas at ‘Intermediate’ level and beyond you will have to make partial deductions about the set of squares a number could be in (or the set of numbers that could be in a square). At ‘Unreasonable’ level, even this is not enough, and you will eventually have to make a guess, and then backtrack if it turns out to be wrong.
Generating difficult puzzles is itself difficult: if you select one of the higher difficulty levels, Solo may have to make many attempts at generating a puzzle before it finds one hard enough for you. Be prepared to wait, especially if you have also configured a large puzzle size.
You have a grid of covered squares, some of which contain mines, but you don't know which. Your job is to uncover every square which does not contain a mine. If you uncover a square containing a mine, you lose. If you uncover a square which does not contain a mine, you are told how many mines are contained within the eight surrounding squares.
This game needs no introduction; popularised by Windows, it is perhaps the single best known desktop puzzle game in existence.
This version of it has an unusual property. By default, it will generate its mine positions in such a way as to ensure that you never need to guess where a mine is: you will always be able to deduce it somehow. So you will never, as can happen in other versions, get to the last four squares and discover that there are two mines left but you have no way of knowing for sure where they are.
This game is played with the mouse.
If you left-click in a covered square, it will be uncovered.
If you right-click in a covered square, it will place a flag which indicates that the square is believed to be a mine. Left-clicking in a marked square will not uncover it, for safety. You can right-click again to remove a mark placed in error.
If you left-click in an uncovered square, it will ‘clear around’ the square. This means: if the square has exactly as many flags surrounding it as it should have mines, then all the covered squares next to it which are not flagged will be uncovered. So once you think you know the location of all the mines around a square, you can use this function as a shortcut to avoid having to click on each of the remaining squares one by one.
If you uncover a square which has no mines in the surrounding eight squares, then it is obviously safe to uncover those squares in turn, and so on if any of them also has no surrounding mines. This will be done for you automatically; so sometimes when you uncover a square, a whole new area will open up to be explored.
You can also use the cursor keys to move around the minefield. Pressing the return key in a covered square uncovers it, and in an uncovered square will clear around it (so it acts as the left button), pressing the space bar in a covered square will place a flag (similarly, it acts as the right button).
All the actions described in section 2.1 are also available.
Even Undo is available, although you might consider it cheating to use it. If you step on a mine, the program will only reveal the mine in question (unlike most other implementations, which reveal all of them). You can then Undo your fatal move and continue playing if you like. The program will track the number of times you died (and Undo will not reduce that counter), so when you get to the end of the game you know whether or not you did it without making any errors.
(If you really want to know the full layout of the grid, which other implementations will show you after you die, you can always use the Solve menu option.)
The options available from the ‘Custom...’ option on the ‘Type’ menu are:
%
sign on the end in which case the game will arrange for that proportion of the squares in the grid to be mines.
Beware of setting the mine count too high. At very high densities, the program may spend forever searching for a solvable grid.
You have a grid of coloured squares, which you have to clear by highlighting contiguous regions of more than one coloured square; the larger the region you highlight, the more points you get (and the faster you clear the arena).
If you clear the grid you win. If you end up with nothing but single squares (i.e., there are no more clickable regions left) you lose.
Removing a region causes the rest of the grid to shuffle up: blocks that are suspended will fall down (first), and then empty columns are filled from the right.
Same Game was contributed to this collection by James Harvey.
This game can be played with either the keyboard or the mouse.
If you left-click an unselected region, it becomes selected (possibly clearing the current selection).
If you left-click the selected region, it will be removed (and the rest of the grid shuffled immediately).
If you right-click the selected region, it will be unselected.
The cursor keys move a cursor around the grid. Pressing the Space or Enter keys while the cursor is in an unselected region selects it; pressing Space or Enter again removes it as above.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
If you turn it off, the game generator will not try to guarantee soluble grids; it will, however, still ensure that there are at least 2 squares of each colour on the grid at the start (since a grid with exactly one square of a given colour is definitely insoluble). Grids generated with this option disabled may contain more large areas of contiguous colour, leading to opportunities for higher scores; they can also take less time to generate.
You have a grid of squares, some light and some dark. Your aim is to light all the squares up at the same time. You can choose any square and flip its state from light to dark or dark to light, but when you do so, other squares around it change state as well.
Each square contains a small diagram showing which other squares change when you flip it.
This game can be played with either the keyboard or the mouse.
Left-click in a square to flip it and its associated squares, or use the cursor keys to choose a square and the space bar or Enter key to flip.
If you use the ‘Solve’ function on this game, it will mark some of the squares in red. If you click once in every square with a red mark, the game should be solved. (If you click in a square without a red mark, a red mark will appear in it to indicate that you will need to reverse that operation to reach the solution.)
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a set of coloured pegs, and have to reproduce a predetermined sequence of them (chosen by the computer) within a certain number of guesses.
Each guess gets marked with the number of correctly-coloured pegs in the correct places (in black), and also the number of correctly-coloured pegs in the wrong places (in white).
This game is also known (and marketed, by Hasbro, mainly) as a board game ‘Mastermind’, with 6 colours, 4 pegs per row, and 10 guesses. However, this version allows custom settings of number of colours (up to 10), number of pegs per row, and number of guesses.
Guess was contributed to this collection by James Harvey.
This game can be played with either the keyboard or the mouse.
With the mouse, drag a coloured peg from the tray on the left-hand side to its required position in the current guess; pegs may also be dragged from current and past guesses to copy them elsewhere. To remove a peg, drag it off its current position to somewhere invalid.
Right-clicking in the current guess adds a ‘hold’ marker; pegs that have hold markers will be automatically added to the next guess after marking.
Alternatively, with the keyboard, the up and down cursor keys can be used to select a peg colour, the left and right keys to select a peg position, and the space bar or Enter key to place a peg of the selected colour in the chosen position. ‘D’ or Backspace removes a peg, and ‘H’ adds a hold marker.
When the guess is complete, the smaller feedback pegs will be highlighted; clicking on these (or moving the peg cursor to them with the arrow keys and pressing the space bar or Enter key) will mark the current guess, copy any held pegs to the next guess, and move the ‘current guess’ marker.
If you correctly position all the pegs the solution will be displayed below; if you run out of guesses (or select ‘Solve...’) the solution will also be revealed.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu. The default game matches the parameters for the board game ‘Mastermind’.
Note that this doesn't allow blank pegs in the solution; if you really wanted that, use one extra colour.
A number of pegs are placed in holes on a board. You can remove a peg by jumping an adjacent peg over it (horizontally or vertically) to a vacant hole on the other side. Your aim is to remove all but one of the pegs initially present.
This game, best known as ‘Peg Solitaire’, is possibly one of the oldest puzzle games still commonly known.
To move a peg, drag it with the mouse from its current position to its final position. If the final position is exactly two holes away from the initial position, is currently unoccupied by a peg, and there is a peg in the intervening square, the move will be permitted and the intervening peg will be removed.
Vacant spaces which you can move a peg into are marked with holes. A space with no peg and no hole is not available for moving at all: it is an obstacle which you must work around.
You can also use the cursor keys to move a position indicator around the board. Pressing the return key while over a peg, followed by a cursor key, will jump the peg in that direction (if that is a legal move).
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
A normal set of dominoes – that is, one instance of every (unordered) pair of numbers from 0 to 6 – has been arranged irregularly into a rectangle; then the number in each square has been written down and the dominoes themselves removed. Your task is to reconstruct the pattern by arranging the set of dominoes to match the provided array of numbers.
This puzzle is widely credited to O. S. Adler, and takes part of its name from those initials.
Left-clicking between any two adjacent numbers places a domino covering them, or removes one if it is already present. Trying to place a domino which overlaps existing dominoes will remove the ones it overlaps.
Right-clicking between two adjacent numbers draws a line between them, which you can use to remind yourself that you know those two numbers are not covered by a single domino. Right-clicking again removes the line.
You can also use the cursor keys to move a cursor around the grid. When the cursor is half way between two adjacent numbers, pressing the return key will place a domino covering those numbers, or pressing the space bar will lay a line between the two squares. Repeating either action removes the domino or line.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You are given a number of points, some of which have lines drawn between them. You can move the points about arbitrarily; your aim is to position the points so that no line crosses another.
I originally saw this in the form of a Flash game called Planarity [7], written by John Tantalo.
[7] http://home.cwru.edu/~jnt5/Planarity
To move a point, click on it with the left mouse button and drag it into a new position.
(All the actions described in section 2.1 are also available.)
There is only one parameter available from the ‘Custom...’ option on the ‘Type’ menu:
A number of balls are hidden in a rectangular arena. You have to deduce the positions of the balls by firing lasers positioned at the edges of the arena and observing how their beams are deflected.
Beams will travel straight from their origin until they hit the opposite side of the arena (at which point they emerge), unless affected by balls in one of the following ways:
Beams that are reflected appear as a ‘R’; beams that hit balls head-on appear as ‘H’. Otherwise, a number appears at the firing point and the location where the beam emerges (this number is unique to that shot).
You can place guesses as to the location of the balls, based on the entry and exit patterns of the beams; once you have placed enough balls a button appears enabling you to have your guesses checked.
Here is a diagram showing how the positions of balls can create each of the beam behaviours shown above:
1RHR----
|..O.O...|
2........3
|........|
|........|
3........|
|......O.|
H........|
|.....O..|
12-RH---
As shown, it is possible for a beam to receive multiple reflections before re-emerging (see turn 3). Similarly, a beam may be reflected (possibly more than once) before receiving a hit (the ‘H’ on the left side of the example).
Note that any layout with more than 4 balls may have a non-unique solution. The following diagram illustrates this; if you know the board contains 5 balls, it is impossible to determine where the fifth ball is (possible positions marked with an x
):
--------
|........|
|........|
|..O..O..|
|...xx...|
|...xx...|
|..O..O..|
|........|
|........|
--------
For this reason, when you have your guesses checked, the game will check that your solution produces the same results as the computer's, rather than that your solution is identical to the computer's. So in the above example, you could put the fifth ball at any of the locations marked with an x
, and you would still win.
Black Box was contributed to this collection by James Harvey.
To fire a laser beam, left-click in a square around the edge of the arena. The results will be displayed immediately. Clicking or holding the left button on one of these squares will highlight the current go (or a previous go) to confirm the exit point for that laser, if applicable.
To guess the location of a ball, left-click within the arena and a black circle will appear marking the guess; click again to remove the guessed ball.
Locations in the arena may be locked against modification by right-clicking; whole rows and columns may be similarly locked by right-clicking in the laser square above/below that column, or to the left/right of that row.
The cursor keys may also be used to move around the grid. Pressing the Enter key will fire a laser or add a new ball-location guess, and pressing Space will lock a cell, row, or column.
When an appropriate number of balls have been guessed, a button will appear at the top-left corner of the grid; clicking that (with mouse or cursor) will check your guesses.
If you click the ‘check’ button and your guesses are not correct, the game will show you the minimum information necessary to demonstrate this to you, so you can try again. If your ball positions are not consistent with the beam paths you already know about, one beam path will be circled to indicate that it proves you wrong. If your positions match all the existing beam paths but are still wrong, one new beam path will be revealed (written in red) which is not consistent with your current guesses.
If you decide to give up completely, you can select Solve to reveal the actual ball positions. At this point, correctly-placed balls will be displayed as filled black circles, incorrectly-placed balls as filled black circles with red crosses, and missing balls as filled red circles. In addition, a red circle marks any laser you had already fired which is not consistent with your ball layout (just as when you press the ‘check’ button), and red text marks any laser you could have fired in order to distinguish your ball layout from the correct one.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a grid of squares. Your aim is to draw a diagonal line through each square, and choose which way each line slants so that the following conditions are met:
Credit for this puzzle goes to Nikoli [8].
[8] http://www.nikoli.co.jp/puzzles/39/index.htm
(in Japanese)
Left-clicking in a blank square will place a \
in it (a line leaning to the left, i.e. running from the top left of the square to the bottom right). Right-clicking in a blank square will place a /
in it (leaning to the right, running from top right to bottom left).
Continuing to click either button will cycle between the three possible square contents. Thus, if you left-click repeatedly in a blank square it will change from blank to \
to /
back to blank, and if you right-click repeatedly the square will change from blank to /
to \
back to blank. (Therefore, you can play the game entirely with one button if you need to.)
You can also use the cursor keys to move around the grid. Pressing the return or space keys will place a \
or a /
, respectively, and will then cycle them as above.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a grid of squares. Some are filled in black; some of the black squares are numbered. Your aim is to ‘light up’ all the empty squares by placing light bulbs in some of them.
Each light bulb illuminates the square it is on, plus all squares in line with it horizontally or vertically unless a black square is blocking the way.
To win the game, you must satisfy the following conditions:
Non-numbered black squares may have any number of lights adjacent to them.
Credit for this puzzle goes to Nikoli [9].
Light Up was contributed to this collection by James Harvey.
[9] http://www.nikoli.co.jp/puzzles/32/index-e.htm
(beware of Flash)
Left-clicking in a non-black square will toggle the presence of a light in that square. Right-clicking in a non-black square toggles a mark there to aid solving; it can be used to highlight squares that cannot be lit, for example.
You may not place a light in a marked square, nor place a mark in a lit square.
The game will highlight obvious errors in red. Lights lit by other lights are highlighted in this way, as are numbered squares which do not (or cannot) have the right number of lights next to them.
Thus, the grid is solved when all non-black squares have yellow highlights and there are no red lights.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
This is a hint rather than an instruction. If the grid generator is unable to generate a puzzle to this precise specification, it will increase the proportion of black squares until it can.
You are given a map consisting of a number of regions. Your task is to colour each region with one of four colours, in such a way that no two regions sharing a boundary have the same colour. You are provided with some regions already coloured, sufficient to make the remainder of the solution unique.
Only regions which share a length of border are required to be different colours. Two regions which meet at only one point (i.e. are diagonally separated) may be the same colour.
I believe this puzzle is original; I've never seen an implementation of it anywhere else. The concept of a four-colouring puzzle was suggested by Owen Dunn; credit must also go to Nikoli and to Verity Allan for inspiring the train of thought that led to me realising Owen's suggestion was a viable puzzle. Thanks also to Gareth Taylor for many detailed suggestions.
To colour a region, click the left mouse button on an existing region of the desired colour and drag that colour into the new region.
(The program will always ensure the starting puzzle has at least one region of each colour, so that this is always possible!)
If you need to clear a region, you can drag from an empty region, or from the puzzle boundary if there are no empty regions left.
Dragging a colour using the right mouse button will stipple the region in that colour, which you can use as a note to yourself that you think the region might be that colour. A region can contain stipples in multiple colours at once. (This is often useful at the harder difficulty levels.)
You can also use the cursor keys to move around the map: the colour of the cursor indicates the position of the colour you would drag (which is not obvious if you're on a region's boundary, since it depends on the direction from which you approached the boundary). Pressing the return key starts a drag of that colour, as above, which you control with the cursor keys; pressing the return key again finishes the drag. The space bar can be used similarly to create a stippled region. Double-pressing the return key (without moving the cursor) will clear the region, as a drag from an empty region does: this is useful with the cursor mode if you have filled the entire map in but need to correct the layout.
If you press L during play, the game will toggle display of a number in each region of the map. This is useful if you want to discuss a particular puzzle instance with a friend – having an unambiguous name for each region is much easier than trying to refer to them all by names such as ‘the one down and right of the brown one on the top border’.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
In ‘Unreasonable’ mode, the program will feel free to generate puzzles which are as hard as it can possibly make them: the only constraint is that they should still have a unique solution. Solving Unreasonable puzzles may require guessing and backtracking.
You are given a grid of dots, marked with yellow lines to indicate which dots you are allowed to connect directly together. Your aim is to use some subset of those yellow lines to draw a single unbroken loop from dot to dot within the grid.
Some of the spaces between the lines contain numbers. These numbers indicate how many of the lines around that space form part of the loop. The loop you draw must correctly satisfy all of these clues to be considered a correct solution.
In the default mode, the dots are arranged in a grid of squares; however, you can also play on triangular or hexagonal grids, or even more exotic ones.
Credit for the basic puzzle idea goes to Nikoli [10].
Loopy was originally contributed to this collection by Mike Pinna, and subsequently enhanced to handle various types of non-square grid by Lambros Lambrou.
[10] http://www.nikoli.co.jp/puzzles/3/index-e.htm
(beware of Flash)
Click the left mouse button on a yellow line to turn it black, indicating that you think it is part of the loop. Click again to turn the line yellow again (meaning you aren't sure yet).
If you are sure that a particular line segment is not part of the loop, you can click the right mouse button to remove it completely. Again, clicking a second time will turn the line back to yellow.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You are a small green ball sitting in a grid full of obstacles. Your aim is to collect all the gems without running into any mines.
You can move the ball in any orthogonal or diagonal direction. Once the ball starts moving, it will continue until something stops it. A wall directly in its path will stop it (but if it is moving diagonally, it will move through a diagonal gap between two other walls without stopping). Also, some of the squares are ‘stops’; when the ball moves on to a stop, it will stop moving no matter what direction it was going in. Gems do not stop the ball; it picks them up and keeps on going.
Running into a mine is fatal. Even if you picked up the last gem in the same move which then hit a mine, the game will count you as dead rather than victorious.
This game was originally implemented for Windows by Ben Olmstead [11], who was kind enough to release his source code on request so that it could be re-implemented for this collection.
[11] http://xn13.com/
You can move the ball in any of the eight directions using the numeric keypad. Alternatively, if you click the left mouse button on the grid, the ball will begin a move in the general direction of where you clicked.
If you use the ‘Solve’ function on this game, the program will compute a path through the grid which collects all the remaining gems and returns to the current position. A hint arrow will appear on the ball indicating the direction in which you should move to begin on this path. If you then move in that direction, the arrow will update to indicate the next direction on the path. You can also press Space to automatically move in the direction of the hint arrow. If you move in a different direction from the one shown by the arrow, the hint arrows will stop appearing because you have strayed from the provided path; you can then use ‘Solve’ again to generate a new path if you want to.
All the actions described in section 2.1 are also available. In particular, if you do run into a mine and die, you can use the Undo function and resume playing from before the fatal move. The game will keep track of the number of times you have done this.
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a grid of squares, some of which contain trees. Your aim is to place tents in some of the remaining squares, in such a way that the following conditions are met:
This puzzle can be found in several places on the Internet, and was brought to my attention by e-mail. I don't know who I should credit for inventing it.
Left-clicking in a blank square will place a tent in it. Right-clicking in a blank square will colour it green, indicating that you are sure it isn't a tent. Clicking either button in an occupied square will clear it.
If you drag with the right button along a row or column, every blank square in the region you cover will be turned green, and no other squares will be affected. (This is useful for clearing the remainder of a row once you have placed all its tents.)
You can also use the cursor keys to move around the grid. Pressing the return key over an empty square will place a tent, and pressing the space bar over an empty square will colour it green; either key will clear an occupied square.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a set of islands distributed across the playing area. Each island contains a number. Your aim is to connect the islands together with bridges, in such a way that:
There are some configurable alternative modes, which involve changing the parallel-bridge limit to something other than 2, and introducing the additional constraint that no sequence of bridges may form a loop from one island back to the same island. The rules stated above are the default ones.
Credit for this puzzle goes to Nikoli [12].
Bridges was contributed to this collection by James Harvey.
[12] http://www.nikoli.co.jp/puzzles/14/index-e.htm
To place a bridge between two islands, click the mouse down on one island and drag it towards the other. You do not need to drag all the way to the other island; you only need to move the mouse far enough for the intended bridge direction to be unambiguous. (So you can keep the mouse near the starting island and conveniently throw bridges out from it in many directions.)
Doing this again when a bridge is already present will add another parallel bridge. If there are already as many bridges between the two islands as permitted by the current game rules (i.e. two by default), the same dragging action will remove all of them.
If you want to remind yourself that two islands definitely do not have a bridge between them, you can right-drag between them in the same way to draw a ‘non-bridge’ marker.
If you think you have finished with an island (i.e. you have placed all its bridges and are confident that they are in the right places), you can mark the island as finished by left-clicking on it. This will highlight it and all the bridges connected to it, and you will be prevented from accidentally modifying any of those bridges in future. Left-clicking again on a highlighted island will unmark it and restore your ability to modify it.
You can also use the cursor keys to move around the grid: if possible the cursor will always move orthogonally, otherwise it will move towards the nearest island to the indicated direction. Pressing the return key followed by a cursor key will lay a bridge in that direction (if available); pressing the space bar followed by a cursor key will lay a ‘non-bridge’ marker.
You can mark an island as finished by pressing the return key twice.
Violations of the puzzle rules will be marked in red:
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
High expansion factors usually mean easier puzzles with fewer possible islands; low expansion factors can create lots of tightly-packed islands.
You have a square grid; each square may contain a digit from 1 to the size of the grid, and some squares have clue signs between them. Your aim is to fully populate the grid with numbers such that:
There are two modes for this game, ‘Unequal’ and ‘Adjacent’.
In ‘Unequal’ mode, the clue signs are greater-than symbols indicating one square's value is greater than its neighbour's. In this mode not all clues may be visible, particularly at higher difficulty levels.
In ‘Adjacent’ mode, the clue signs are bars indicating one square's value is numerically adjacent (i.e. one higher or one lower) than its neighbour. In this mode all clues are always visible: absence of a bar thus means that a square's value is definitely not numerically adjacent to that neighbour's.
In ‘Trivial’ difficulty level (available via the ‘Custom’ game type selector), there are no greater-than signs in ‘Unequal’ mode; the puzzle is to solve the Latin square only.
At the time of writing, the ‘Unequal’ mode of this puzzle is appearing in the Guardian weekly under the name ‘Futoshiki’.
Unequal was contributed to this collection by James Harvey.
Unequal shares much of its control system with Solo.
To play Unequal, simply click the mouse in any empty square and then type a digit or letter on the keyboard to fill that square. If you make a mistake, click the mouse in the incorrect square and press Space to clear it again (or use the Undo feature).
If you right-click in a square and then type a number, that number will be entered in the square as a ‘pencil mark’. You can have pencil marks for multiple numbers in the same square. Squares containing filled-in numbers cannot also contain pencil marks.
The game pays no attention to pencil marks, so exactly what you use them for is up to you: you can use them as reminders that a particular square needs to be re-examined once you know more about a particular number, or you can use them as lists of the possible numbers in a given square, or anything else you feel like.
To erase a single pencil mark, right-click in the square and type the same number again.
All pencil marks in a square are erased when you left-click and type a number, or when you left-click and press space. Right-clicking and pressing space will also erase pencil marks.
As for Solo, the cursor keys can be used in conjunction with the digit keys to set numbers or pencil marks. You can also use the 'M' key to auto-fill every numeric hint, ready for removal as required, or the 'H' key to do the same but also to remove all obvious hints.
Alternatively, use the cursor keys to move the mark around the grid. Pressing the return key toggles the mark (from a normal mark to a pencil mark), and typing a number in is entered in the square in the appropriate way; typing in a 0 or using the space bar will clear a filled square.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a rectangular grid containing a number of dots. Your aim is to draw edges along the grid lines which divide the rectangle into regions in such a way that every region is 180° rotationally symmetric, and contains exactly one dot which is located at its centre of symmetry.
This puzzle was invented by Nikoli [13], under the name ‘Tentai Show’; its name is commonly translated into English as ‘Spiral Galaxies’.
Galaxies was contributed to this collection by James Harvey.
[13] http://www.nikoli.co.jp/en/puzzles/astronomical_show/
Left-click on any grid line to draw an edge if there isn't one already, or to remove one if there is. When you create a valid region (one which is closed, contains exactly one dot, is 180° symmetric about that dot, and contains no extraneous edges inside it) it will be highlighted automatically; so your aim is to have the whole grid highlighted in that way.
During solving, you might know that a particular grid square belongs to a specific dot, but not be sure of where the edges go and which other squares are connected to the dot. In order to mark this so you don't forget, you can right-click on the dot and drag, which will create an arrow marker pointing at the dot. Drop that in a square of your choice and it will remind you which dot it's associated with. You can also right-click on existing arrows to pick them up and move them, or destroy them by dropping them off the edge of the grid. (Also, if you're not sure which dot an arrow is pointing at, you can pick it up and move it around to make it clearer. It will swivel constantly as you drag it, to stay pointed at its parent dot.)
You can also use the cursor keys to move around the grid squares and lines. Pressing the return key when over a grid line will draw or clear its edge, as above. Pressing the return key when over a dot will pick up an arrow, to be dropped the next time the return key is pressed; this can also be used to move existing arrows around, removing them by dropping them on a dot or another arrow.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a grid of squares, some of which contain digits, and the rest of which are empty. Your job is to fill in digits in the empty squares, in such a way that each connected region of squares all containing the same digit has an area equal to that digit.
(‘Connected region’, for the purposes of this game, does not count diagonally separated squares as adjacent.)
For example, it follows that no square can contain a zero, and that two adjacent squares can not both contain a one. No region has an area greater than 9 (because then its area would not be a single digit).
Credit for this puzzle goes to Nikoli [14].
Filling was contributed to this collection by Jonas Kölker.
[14] http://www.nikoli.co.jp/en/puzzles/fillomino/
To play Filling, simply click the mouse in any empty square and then type a digit on the keyboard to fill that square. By dragging the mouse, you can select multiple squares to fill with a single keypress. If you make a mistake, click the mouse in the incorrect square and press 0, Space, Backspace or Enter to clear it again (or use the Undo feature).
You can also move around the grid with the cursor keys; typing a digit will fill the square containing the cursor with that number, or typing 0, Space, or Enter will clear it. You can also select multiple squares for numbering or clearing by using the return key, before typing a digit to fill in the highlighted squares (as above).
(All the actions described in section 2.1 are also available.)
Filling allows you to configure the number of rows and columns of the grid, through the ‘Type’ menu.
You have a square grid; each square may contain a digit from 1 to the size of the grid. The grid is divided into blocks of varying shape and size, with arithmetic clues written in them. Your aim is to fully populate the grid with digits such that:
Note that a block may contain more than one digit the same (provided the identical ones are not in the same row and column). This rule is precisely the opposite of the rule in Solo's ‘Killer’ mode (see chapter 11).
This puzzle appears in the Times under the name ‘KenKen’.
Keen shares much of its control system with Solo (and Unequal).
To play Keen, simply click the mouse in any empty square and then type a digit on the keyboard to fill that square. If you make a mistake, click the mouse in the incorrect square and press Space to clear it again (or use the Undo feature).
If you right-click in a square and then type a number, that number will be entered in the square as a ‘pencil mark’. You can have pencil marks for multiple numbers in the same square. Squares containing filled-in numbers cannot also contain pencil marks.
The game pays no attention to pencil marks, so exactly what you use them for is up to you: you can use them as reminders that a particular square needs to be re-examined once you know more about a particular number, or you can use them as lists of the possible numbers in a given square, or anything else you feel like.
To erase a single pencil mark, right-click in the square and type the same number again.
All pencil marks in a square are erased when you left-click and type a number, or when you left-click and press space. Right-clicking and pressing space will also erase pencil marks.
As for Solo, the cursor keys can be used in conjunction with the digit keys to set numbers or pencil marks. Use the cursor keys to move a highlight around the grid, and type a digit to enter it in the highlighted square. Pressing return toggles the highlight into a mode in which you can enter or remove pencil marks.
Pressing M will fill in a full set of pencil marks in every square that does not have a main digit in it.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a square grid. On each square of the grid you can build a tower, with its height ranging from 1 to the size of the grid. Around the edge of the grid are some numeric clues.
Your task is to build a tower on every square, in such a way that:
In harder or larger puzzles, some towers will be specified for you as well as the clues round the edge, and some edge clues may be missing.
This puzzle appears on the web under various names, particularly ‘Skyscrapers’, but I don't know who first invented it.
Towers shares much of its control system with Solo, Unequal and Keen.
To play Towers, simply click the mouse in any empty square and then type a digit on the keyboard to fill that square with a tower of the given height. If you make a mistake, click the mouse in the incorrect square and press Space to clear it again (or use the Undo feature).
If you right-click in a square and then type a number, that number will be entered in the square as a ‘pencil mark’. You can have pencil marks for multiple numbers in the same square. A square containing a tower cannot also contain pencil marks.
The game pays no attention to pencil marks, so exactly what you use them for is up to you: you can use them as reminders that a particular square needs to be re-examined once you know more about a particular number, or you can use them as lists of the possible numbers in a given square, or anything else you feel like.
To erase a single pencil mark, right-click in the square and type the same number again.
All pencil marks in a square are erased when you left-click and type a number, or when you left-click and press space. Right-clicking and pressing space will also erase pencil marks.
As for Solo, the cursor keys can be used in conjunction with the digit keys to set numbers or pencil marks. Use the cursor keys to move a highlight around the grid, and type a digit to enter it in the highlighted square. Pressing return toggles the highlight into a mode in which you can enter or remove pencil marks.
Pressing M will fill in a full set of pencil marks in every square that does not have a main digit in it.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
You have a grid of white squares, all of which contain numbers. Your task is to colour some of the squares black (removing the number) so as to satisfy all of the following conditions:
Credit for this puzzle goes to Nikoli [15] who call it Hitori.
Singles was contributed to this collection by James Harvey.
[15] http://www.nikoli.com/en/puzzles/hitori/index.html
(beware of Flash)
Left-clicking on an empty square will colour it black; left-clicking again will replace the number. Right-clicking will add a circle (useful for indicating that a cell is definitely not black).
You can also use the cursor keys to move around the grid. Pressing the return or space keys will turn a square black or add a circle respectively, and pressing the key again will replace the number or remove the circle.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
A rectangular grid has been filled with a mixture of magnets (that is, dominoes with one positive end and one negative end) and blank dominoes (that is, dominoes with two neutral poles). These dominoes are initially only seen in silhouette. Around the grid are placed a number of clues indicating the number of positive and negative poles contained in certain columns and rows.
Your aim is to correctly place the magnets and blank dominoes such that all the clues are satisfied, with the additional constraint that no two similar magnetic poles may be orgothonally adjacent (since they repel). Neutral poles do not repel, and can be adjacacent to any other pole.
Credit for this puzzle goes to Janko [16].
Magnets was contributed to this collection by James Harvey.
[16] http://www.janko.at/Raetsel/Magnete/index.htm
Left-clicking on an empty square places a magnet at that position with the positive pole on the square and the negative pole on the other half of the magnet; left-clicking again reverses the polarity, and a third click removes the magnet.
Right-clicking on an empty square places a blank domino there. Right-clicking again places two question marks on the domino, signifying 'this cannot be blank' (which can be useful to note deductions while solving, and right-clicking again empties the domino.
You can also use the cursor keys to move a cursor around the grid. Pressing the return key will lay a domino with a positive pole at that position; pressing again reverses the polarity and then removes the domino, as with left-clicking. Using the space bar allows placement of blank dominoes and cannot-be-blank hints, as for right-clicking.
(All the actions described in section 2.1 are also available.)
These parameters are available from the ‘Custom...’ option on the ‘Type’ menu.
(Grids with at least one odd dimension tend to be easier to solve)
This software is copyright 2004-2010 Simon Tatham.
Portions copyright Richard Boulton, James Harvey, Mike Pinna, Jonas Kölker, Dariusz Olszewski, Michael Schierl, Lambros Lambrou and Bernd Schmidt.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the ‘Software’), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED ‘AS IS’, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Black Box: Chapter 19
Bridges: Chapter 26
bugs: Chapter 1
command line: Section 2.2, Section 2.4, Section 2.5
common features: Chapter 2
controls: Section 2.1
controls, for Black Box: Section 19.1
controls, for Bridges: Section 26.1
controls, for Cube: Section 4.1
controls, for Dominosa: Section 17.1
controls, for Fifteen: Section 5.1
controls, for Filling: Section 29.1
controls, for Flip: Section 14.1
controls, for Galaxies: Section 28.1
controls, for Guess: Section 15.1
controls, for Inertia: Section 24.1
controls, for Keen: Section 30.1
controls, for Light Up: Section 21.1
controls, for Loopy: Section 23.1
controls, for Magnets: Section 33.1
controls, for Map: Section 22.1
controls, for Mines: Section 12.1
controls, for Net: Section 3.1
controls, for Netslide: Chapter 9
controls, for Pattern: Section 10.1
controls, for Pegs: Section 16.1
controls, for Rectangles: Section 8.1
controls, for Same Game: Section 13.1
controls, for Singles: Section 32.1
controls, for Sixteen: Section 6.1
controls, for Slant: Section 20.1
controls, for Solo: Section 11.1
controls, for Tents: Section 25.1
controls, for Towers: Section 31.1
controls, for Twiddle: Section 7.1
controls, for Unequal: Section 27.1
controls, for Untangle: Section 18.1
copy: Section 2.1
copyright: Appendix A
Cube: Chapter 4
‘Custom’, menu option: Section 2.3
default parameters, specifying: Section 2.4
Dominosa: Chapter 17
Edit menu: Section 2.1
exit: Section 2.1
feedback: Chapter 1
Fifteen: Chapter 5
File menu: Section 2.1
Filling: Chapter 29
Flip: Chapter 14
format, ID: Section 2.2
four-colouring: Chapter 22
FreeNet: Chapter 3
Futoshiki: Chapter 27
Galaxies: Chapter 28
game ID: Section 2.2
game ID, format: Section 2.2
game ID, generating: Section 2.5
Game menu: Section 2.1, Section 2.2
generating game IDs: Section 2.5
Guess: Chapter 15
Hitori: Chapter 32
ID format: Section 2.2
ID, game: Section 2.2
Inertia: Chapter 24
initial state: Section 2.2
Janko: Chapter 33
Keen: Chapter 30
KenKen: Chapter 30
keys: Section 2.1
keys, for Black Box: Section 19.1
keys, for Cube: Section 4.1
keys, for Fifteen: Section 5.1
keys, for Flip: Section 14.1
keys, for Guess: Section 15.1
keys, for Inertia: Section 24.1
keys, for Net: Section 3.1
keys, for Same Game: Section 13.1
Latin square: Chapter 27
licence: Appendix A
licence, MIT: Chapter 1, Appendix A
Light Up: Chapter 21
Linux: Chapter 1, Section 2.5
load: Section 2.1, Section 2.5
Loopy: Chapter 23
Mac OS X: Chapter 1, Section 2.1, Section 2.2, Section 2.4
Magnets: Chapter 33
Map: Chapter 22
Mastermind: Chapter 15
Mines: Chapter 12
MIT licence: Chapter 1, Appendix A
Net: Chapter 3
NETGAME.EXE
: Chapter 3
Netslide: Chapter 9
NetWalk: Chapter 3
new game: Section 2.1
Nikoli: Chapter 8, Chapter 11, Chapter 20, Chapter 21, Chapter 23, Chapter 26, Chapter 28, Chapter 29, Chapter 32
nonograms: Chapter 10
parameters: Section 2.2, Section 2.3
parameters, for Black Box: Section 19.2
parameters, for Bridges: Section 26.2
parameters, for Cube: Section 4.2
parameters, for Dominosa: Section 17.2
parameters, for Fifteen: Section 5.2
parameters, for Filling: Section 29.2
parameters, for flip: Section 14.2
parameters, for Galaxies: Section 28.2
parameters, for Guess: Section 15.2
parameters, for Inertia: Section 24.2
parameters, for Keen: Section 30.2
parameters, for Light Up: Section 21.2
parameters, for Loopy: Section 23.2
parameters, for Magnets: Section 33.2
parameters, for Map: Section 22.2
parameters, for Mines: Section 12.2
parameters, for Net: Section 3.2
parameters, for Netslide: Chapter 9
parameters, for Pattern: Section 10.2
parameters, for Pegs: Section 16.2
parameters, for Rectangles: Section 8.2
parameters, for Same Game: Section 13.2
parameters, for Singles: Section 32.2
parameters, for Sixteen: Section 6.2
parameters, for Slant: Section 20.2
parameters, for Solo: Section 11.2
parameters, for Tents: Section 25.2
parameters, for Towers: Section 31.2
parameters, for Twiddle: Section 7.2
parameters, for Unequal: Section 27.2
parameters, for Untangle: Section 18.2
patches: Chapter 1
Pattern: Chapter 10
Pegs: Chapter 16
Planarity: Chapter 18
PostScript: Section 2.5
preferences, specifying default: Section 2.4
preset: Section 2.3
printing, on Unix: Section 2.5
printing, on Windows: Section 2.1
15-puzzle: Chapter 5
Puzzle Palace: Chapter 8
quit: Section 2.1
Random Seed: Section 2.2
Rectangles: Chapter 8
redo: Section 2.1
restart game: Section 2.1
Same Game: Chapter 13
save: Section 2.1, Section 2.5
shortcuts (keyboard): Section 2.1
shortcuts (keyboard), for Black Box: Section 19.1
shortcuts (keyboard), for Cube: Section 4.1
shortcuts (keyboard), for Fifteen: Section 5.1
shortcuts (keyboard), for Flip: Section 14.1
shortcuts (keyboard), for Guess: Section 15.1
shortcuts (keyboard), for Inertia: Section 24.1
shortcuts (keyboard), for Net: Section 3.1
shortcuts (keyboard), for Same Game: Section 13.1
Singles: Chapter 32
Sixteen: Chapter 6
Skyscrapers: Chapter 31
Slant: Chapter 20
Solitaire, Peg: Chapter 16
Solo: Chapter 11
solve: Section 2.1
source code: Chapter 1
‘Specific’, menu option: Section 2.2
state, initial: Section 2.2
Tents: Chapter 25
Towers: Chapter 31
Twiddle: Chapter 7
Type menu: Section 2.3
undo: Section 2.1
Unequal: Chapter 27
Unix: Chapter 1, Section 2.5
Untangle: Chapter 18
version: Section 2.2
website: Chapter 1
Windows: Chapter 1, Chapter 3