Box Off

[raidin...@gmail.com]

Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further. Here are the rules:

The standard size for it is 8x6, and the standard number of different kinds of tiles is 3 --- black, white and yellow. A different size can be used (for instance, 10x8), and also a different number of kinds of tiles can be used (for instance, 4 different kinds of tiles --- black, white, yellow and green.


Tiles are randomly laid out on the board to start the game. The object is to clear all the tiles off the board. Two same-colored tiles can be removed if they share an otherwise empty rectangle. Here are some examples on a 4x4 board:


_ _ _ _
_ b b _
_ _ _ _
_ _ _ _


In the above diagram, the two black tiles can be removed because they form a 2x1 rectangle.


_ _ _ _
w _ _ _
_ _ _ _
w _ _ _


In the above diagram, the two white tiles may be removed because they form the ends of an otherwise empty 1x3 rectangle. (If there were a tile between them --- including a white tile --- they could not be removed.)


_ _ _ y
_ _ _ _
_ _ _ _
_ y _ _


In the above diagram, the two yellow tiles may be removed because they form the corners of an otherwise empty 3x4 rectangle. If there were any other tiles within that rectangle --- including yellow tiles --- they could not be removed.


Here's an example game on a 6x3 board:


b b w b b w
b y y y w b
y w w w y y


The two white tiles at b1 and c1 are removed:


b b w y y w
b y y y w b
y _ _ w y y


Then the two black tiles at a2 and a3 are removed:


_ b w b b w
_ y y y w b
y _ _ w y y


Then the two yellow tiles at a1 and b2 are removed:


_ b w b b w
_ _ y y w b
_ _ _ w y y


Then the two yellow tiles at c2 and d2 are removed:


_ b w b b w
_ _ _ _ w b
_ _ _ w y y


Then the two black tiles at d3 and e3 are removed:


_ b w _ _ w
_ _ _ _ w b
_ _ _ w y y


Then the two white tiles at c3 and d1 are removed:


_ b _ _ _ w
_ _ _ _ w b
_ _ _ _ y y


Then the two yellow tiles at e1 and f1 are removed:


_ b _ _ _ w
_ _ _ _ w b
_ _ _ _ _ _


At this point nothing else can be done. The remaining black an white tiles interfere with each other. But the initial position could have been solved, had not a mistake been made. Do you see the mistake?

[Mark Steere]

Nice concept :)

Not an abstract game, but that's ok. There's plenty of bandwidth to spare now that Hazen ran away like a whipped dog. And now that nobody (including me) can come up with anything original. Some of us (including Luis The Mangy) are even intentionally unoriginal.

[Nick Bentley]

> Not an abstract game, but that's ok. There's plenty of bandwidth to spare now that Hazen ran away like a whipped dog. And now that nobody (including me) can come up with anything original. Some of us (including Luis The Mangy) are even intentionally unoriginal.

I may actually have an original thing or two to show in the next few months. No promises, but I've got a couple of weird ones in development.

[Mark Steere]

Looking forward, Nick :)

[raidin...@gmail.com]

On Tuesday, February 19, 2013 12:53:24 PM UTC-5, Mark Steere wrote:
> Nice concept :)

Thanks! I'm hoping Box Off will be more successful than my abstract games thus far. Comments and suggestions are welcome.

Steve

[Nick Reymann]

Neat! It reminds me of an old DOS game called Sixx which I used to play, except that in Sixx, you only need one or two consecutive "edges" of the rectangle to be empty, not the entire area of it. (Another way to visualize it is that there needs to be a clear path connecting two tiles, with only one 90 degree change in direction allowed.)

[gschm...@gmail.com]

On Tuesday, February 19, 2013 11:36:00 AM UTC-5, raidin...@gmail.com wrote:
> Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further. Here are the rules: The standard size for it is 8x6, and the standard number of different kinds of tiles is 3 --- black, white and yellow. A different size can be used (for instance, 10x8), and also a different number of kinds of tiles can be used (for instance, 4 different kinds of tiles --- black, white, yellow and green. But the initial position could have been solved, had not a mistake been made. Do you see the mistake?

a1 and b2 should not have been removed. Here's is my solution:

b b w b b w
b y y y w b
y w w w y y

b b w b b w
b y y y w b

y - - w y y

b b w b b w
b y y y w b

y - - w - -

b b w b b w

b y y y - b
y - - - - -

b b w - - w
b y y y - b
y - - - - -

b b w - - w
b - y y - b
- - - - - -

- - w - - w
b - y y - b
- - - - - -

- - w - - w
b - - - - b
- - - - - -

- - w - - w
- - - - - -
- - - - - -

- - - - - -
- - - - - -
- - - - - -

[raidin...@gmail.com]

On Friday, February 22, 2013 10:57:21 AM UTC-5, Nick Reymann wrote:
> Neat! It reminds me of an old DOS game called Sixx which I used to play, except that in Sixx, you only need one or two consecutive "edges" of the rectangle to be empty, not the entire area of it. (Another way to visualize it is that there needs to be a clear path connecting two tiles, with only one 90 degree change in direction allowed.)

Interesting. If you get a chance to try Box Off, I'd be curious to know how you think it compares to Sixx in terms of complexity. I recommend playing Box Off on a 12x8 board with 4 different kinds of tiles.

[raidin...@gmail.com]

On Friday, February 22, 2013 1:18:03 PM UTC-5, gschm...@gmail.com wrote:
> On Tuesday, February 19, 2013 11:36:00 AM UTC-5, raidin...@gmail.com wrote:
>
> > Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further. Here are the rules: The standard size for it is 8x6, and the standard number of different kinds of tiles is 3 --- black, white and yellow. A different size can be used (for instance, 10x8), and also a different number of kinds of tiles can be used (for instance, 4 different kinds of tiles --- black, white, yellow and green. But the initial position could have been solved, had not a mistake been made. Do you see the mistake?
>
>
>
> a1 and b2 should not have been removed. Here's is my solution:
>
>
>
> b b w b b w
>
> b y y y w b
>
> y w w w y y
>
>
>
> b b w b b w
>
> b y y y w b
>
> y - - w y y
>
>
>
> b b w b b w
>
> b y y y w b
>
> y - - w - -

I'm not sure you understand the rules correctly. It looks like you're removing the w's at d1 and e2? If so, this is illegal because the rectangle isn't otherwise empty (there's a y at d2).

[gschm...@gmail.com]

On Friday, February 22, 2013 11:21:43 PM UTC-5, raidin...@gmail.com wrote:
> I'm not sure you understand the rules correctly. It looks like you're removing the w's at d1 and e2? If so, this is illegal because the rectangle isn't otherwise empty (there's a y at d2).

Oops. Actually, I do understand the rules but in my haste messed things up. (I solved it by first circling the rectangular groups around corner pairs, then constructing the order that leads to that configuration). Here's my corrected solution...

b b w b b w
b y y y w b
y w w w y y

b b w b b w
b y y y w b
y - - w y y

b b w b b w
b y y y w b
y - - w - -

b b w b b w

b y - - w b

y - - w - -

b b w - - w

b y - - w b

y - - w - -

b b - - - -
b y - - w b

y - - w - -

b b - - - -
b y - - - b

y - - - - -

- b - - - -

- y - - - b

y - - - - -

- b - - - -

- - - - - b

- - - - - -

- - - - - -

[raidin...@gmail.com]

Yes that's correct.

[gschm...@gmail.com]

I might be interested in making a (free) computer version of Box Off (both Axiom & Zillions). Is that alright with you?

[raidin...@gmail.com]

On Saturday, February 23, 2013 5:25:08 PM UTC-5, gschm...@gmail.com wrote:
> I might be interested in making a (free) computer version of Box Off (both Axiom & Zillions). Is that alright with you?

Yes that'd be great!

[gschm...@gmail.com]

OK then, I've posted an initial computer version of the puzzle (Axiom PC game) at:
http://www.datafilehost.com/download-5c9e9295.html

Please download it and give it a try.

"raidin" (and anyone else who's interested) please read the following file (which includes my contact email) and contact me:
C:\Program Files\BoxOff\BoxOff\Readme.txt

If anyone would like to submit 8x6 puzzles to me, please do and I will include them in the final version of the puzzle game.

Thanks!

[Daniel S.]

interesting puzzle!

I guess to generate them you just work backwards?
Is there a way to be sure that there is only 1 solution when generating the puzzle?
given an even number of b, w, and y respectively will there always be a solution for a filled board?
is there a way to adjust the difficulty without changing the board size?

There's also rec.puzzles. You might try getting feedback there as well.

[raidin...@gmail.com]

On Monday, February 25, 2013 12:06:52 AM UTC-5, Daniel S. wrote:
> interesting puzzle!
>
>
>
> I guess to generate them you just work backwards?
>
> Is there a way to be sure that there is only 1 solution when generating the puzzle?
>
> given an even number of b, w, and y respectively will there always be a solution for a filled board?

No, some initial setups will be mathematically impossible to solve. The *smaller* the board size, the more likely on average it is to be impossible, due to reduced options for working your way out of jams.

>
> is there a way to adjust the difficulty without changing the board size?

Yes, increase the number of different kinds of stones (from 3 to 4).

[gschm...@gmail.com]

On Monday, February 25, 2013 12:38:19 AM UTC-5, raidin...@gmail.com wrote: > On Monday, February 25, 2013 12:06:52 AM UTC-5, Daniel S. wrote: > interesting puzzle! > > > > I guess to generate them you just work backwards? > > Is there a way to be sure that there is only 1 solution when generating the puzzle? > > given an even number of b, w, and y respectively will there always be a solution for a filled board? No, some initial setups will be mathematically impossible to solve. The *smaller* the board size, the more likely on average it is to be impossible, due to reduced options for working your way out of jams. > > is there a way to adjust the difficulty without changing the board size? Yes, increase the number of different kinds of stones (from 3 to 4). > > > > There's also rec.puzzles. You might try getting feedback there as well.

First off, the version I posted plays a slightly different game which I will fix of course.

I'm looking for effective ways of creating new puzzles, so if anyone has any suggestions, I'd like to hear them. I don't think working backwards necessarily works since you at least have to make sure you're not creating any irregular gaps in the process that can't later accomodate a box off move (e.g. a 1x1 box, or a 2x1 "L").

[Daniel S.]

> I'm looking for effective ways of creating new puzzles, so if anyone has any suggestions, I'd like to hear them. I don't think working backwards necessarily works since you at least have to make sure you're not creating any irregular gaps in the process that can't later accomodate a box off move (e.g. a 1x1 box, or a 2x1 "L").

work backwards.
allow gaps.
maybe even put orange stones in the gaps and instruct to remove them at the start of the puzzle.
it might have aesthetic value even.

[gschm...@gmail.com]

Welcome to the new Google Groups! Revert to the old Google Groups. Learn about the new features you’ll find.
Groups
POST REPLY
1 of 99+ (99+)
OverviewDiscussion

My groups
Home
Starred
Announcements
Google Groups Announcements
Recently viewed
Google Maps TSP Solver
comp.lang.forth
comp.ai.games
rec.games.abstract
Recent searches
Recently posted to
rec.games.abstract
Favorites
Click on a group’s star icon to add it to your favorites
©2013 Google
Privacy - Terms of Service - Google Home
rec.games.abstract ›
Box Off
19 posts by 6 authors in rec.games.abstract



Feb 19

raidin...@gmail.com Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further. Here are the rules: The standard size for it is 8x6, and the standard number of different kinds of tiles is


Feb 19

Mark Steere Nice concept :) Not an abstract game, but that's ok. There's plenty of bandwidth to spare now that Hazen ran away like a whipped dog. And now that nobody (including me) can come up with anything original. Some of us (including Luis The Mangy) are e


Feb 19

Nick Bentley I may actually have an original thing or two to show in the next few months. No promises, but I've got a couple of weird ones in development.


Feb 19

Mark Steere Looking forward, Nick :)


Feb 19

raidin...@gmail.com On Tuesday, February 19, 2013 12:53:24 PM UTC-5, Mark Steere wrote: > Nice concept :) Thanks! I'm hoping Box Off will be more successful than my abstract games thus far. Comments and suggestions are welcome. Steve


Feb 22

Nick Reymann Neat! It reminds me of an old DOS game called Sixx which I used to play, except that in Sixx, you only need one or two consecutive "edges" of the rectangle to be empty, not the entire area of it. (Another way to visualize it is that there needs to


Feb 22

me On Tuesday, February 19, 2013 11:36:00 AM UTC-5, raidin...@gmail.com wrote: > Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further. Here are the rules: The standard


Feb 22

raidin...@gmail.com Interesting. If you get a chance to try Box Off, I'd be curious to know how you think it compares to Sixx in terms of complexity. I recommend playing Box Off on a 12x8 board with 4 different kinds of tiles.


Feb 22

raidin...@gmail.com I'm not sure you understand the rules correctly. It looks like you're removing the w's at d1 and e2? If so, this is illegal because the rectangle isn't otherwise empty (there's a y at d2).


Feb 23

me Oops. Actually, I do understand the rules but in my haste messed things up. (I solved it by first circling the rectangular groups around corner pairs, then constructing the order that leads to that configuration). Here's my corrected solution...


Feb 23

raidin...@gmail.com Yes that's correct.


Feb 23

me I might be interested in making a (free) computer version of Box Off (both Axiom & Zillions). Is that alright with you?


Feb 23

raidin...@gmail.com Yes that'd be great!


Feb 24 (23 hours ago)

me OK then, I've posted an initial computer version of the puzzle (Axiom PC game) at: http://www.datafilehost.com/download-5c9e9295.html Please download it and give it a try. "raidin" (and anyone else who's interested) please read the following file (wh


12:06 AM (15 hours ago)

Daniel S. interesting puzzle! I guess to generate them you just work backwards? Is there a way to be sure that there is only 1 solution when generating the puzzle? given an even number of b, w, and y respectively will there always be a solution for a filled bo


12:38 AM (15 hours ago)

raidin...@gmail.com No, some initial setups will be mathematically impossible to solve. The *smaller* the board size, the more likely on average it is to be impossible, due to reduced options for working your way out of jams.




me (gschm...@gmail.com change) Post reply

10:40 AM (5 hours ago)

On Monday, February 25, 2013 12:38:19 AM UTC-5, raidin...@gmail.com wrote: > On Monday, February 25, 2013 12:06:52 AM UTC-5, Daniel S. wrote: > interesting puzzle! > > > > I guess to generate them you just work backwards? > > Is there a way to be sure that there is only 1 solution when generating the puzzle? > > given an even number of b, w, and y respectively will there always be a solution for a filled board? No, some initial setups will be mathematically impossible to solve. The *smaller* the board size, the more likely on average it is to be impossible, due to reduced options for working your way out of jams. > > is there a way to adjust the difficulty without changing the board size? Yes, increase the number of different kinds of stones (from 3 to 4). > > > > There's also rec.puzzles. You might try getting feedback there as well.

First off, the version I posted plays a slightly different game which I will fix of course.

I'm looking for effective ways of creating new puzzles, so if anyone has any suggestions, I'd like to hear them. I don't think working backwards necessarily works since you at least have to make sure you're not creating any irregular gaps in the process that can't later accomodate a box off move (e.g. a 1x1 box, or a 2x1 "L").

Daniel S. Post reply

1:46 PM (2 hours ago)

me (gschm...@gmail.com change) Post reply

3:51 PM (less than a minute ago)

- hide quoted text -

That works too, although it may result in a slightly easier puzzle since there would be fewer tiles remaining.

I think this works...

Work backwards, but don't make any moves that result in an odd sized gap. That works because any even size gap can be filled in by some combination of 2x1 boxes (or larger).

This leads to the following strategy. Don't make a move which produces an odd sized gap.

(I strongly suspect there is a "parity" aspect to this puzzle)

[Phil Carmody]

raidin...@gmail.com writes:
> Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further.

If they do, they're quitters.

Any solitaire where you either finish or not can be turned into a
2-player game where you take turns, and the last player to move is
either the winner or loser.

Phil
--
"In a world of magnets and miracles"
-- Insane Clown Posse, Miracles, 2009. Much derided.
"Magnets, how do they work"
-- Pink Floyd, High Hopes, 1994. Lauded as lyrical geniuses.

[gschm...@gmail.com]

On Thursday, March 7, 2013 6:29:44 AM UTC-5, Phil Carmody wrote:
> raidin...@gmail.com writes:
>
> > Box Off is a solitaire game, so technically this post is off topic. 2-Player perfect information game purists will want to read no further.
>
>
>
> If they do, they're quitters.

> Any solitaire where you either finish or not can be turned into a
>
> 2-player game where you take turns, and the last player to move is
>
> either the winner or loser.
>

Good point.

Another thought is that two players play side by side on identical boards and attempt to clear the board. Each player is disallowed from playing the other player's previous move. I briefly tried it. It may work, but there may be flaws that I haven't considered.

[gschm...@gmail.com]

I have made available a freeware version of BoxOff at:

http://www.freewarefiles.com/BoxOff_program_85505.html

The puzzle game supports three sizes, 8x6, 8x12, and 12x15.

Enjoy!

-- Greg