Jump to content

  • Log In with Google      Sign In   
  • Create Account


auto generate mahjong level with at least 1 way to win


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
8 replies to this topic

#1 newwork01   Members   -  Reputation: 112

Like
2Likes
Like

Posted 24 March 2014 - 03:21 AM

Hello everybody, Im new in game programing.

 

I decided to make a game like mahjong but simpler than it !

 

there are n tiles marked with char A-Z, they are arranged into many layer of pyramid shape . The bottom layer will have mxm tiles , the top layer will have 2x2 tiles . Here is the sample :

 

532Qyx7.jpg

 

Game rule : User will pick 2 tiles, if they have same char then remove both . User win the game when all tiles are removed.

 

The problem is : How to auto generate the level randomly with at least 1 way to win . Please help , im stuck with this problem ! Thank a lot guy !



Sponsor:

#2 frob   Moderators   -  Reputation: 18874

Like
2Likes
Like

Posted 24 March 2014 - 03:56 AM

There are several different Mah-Jongg rules out there.  It is a multi-person game much like Rummy and similar card game variants.

 

If you just mean the solitaire matching game (which is not Mah-Jongg but sometimes called that) then you can build up a layout in reverse.  Start by placing a pair of tiles, then place another pair of tiles, then another, then another, until all the tiles are placed. It is guaranteed winnable since you can exactly reverse the order. Just be careful to build in a somewhat regular manner to avoid adjacent tiles that you cannot place.


Check out my personal indie blog at bryanwagstaff.com.

#3 newwork01   Members   -  Reputation: 112

Like
0Likes
Like

Posted 24 March 2014 - 04:07 AM

There are several different Mah-Jongg rules out there.  It is a multi-person game much like Rummy and similar card game variants.

 

If you just mean the solitaire matching game (which is not Mah-Jongg but sometimes called that) then you can build up a layout in reverse.  Start by placing a pair of tiles, then place another pair of tiles, then another, then another, until all the tiles are placed. It is guaranteed winnable since you can exactly reverse the order. Just be careful to build in a somewhat regular manner to avoid adjacent tiles that you cannot place.

Thank for your help ! But can you give me the sample of placing those tiles because there are many layer for botton to top that make a pyramidal form ? Do i have to start placing those tiles at the top layer or the bottom layer or random layer(each tile at different layer) ?



#4 frob   Moderators   -  Reputation: 18874

Like
2Likes
Like

Posted 24 March 2014 - 04:38 AM


But can you give me the sample of placing those tiles because there are many layer for botton to top that make a pyramidal form ? Do i have to start placing those tiles at the top layer or the bottom layer or random layer(each tile at different layer) ?

I don't have an algorithm, but I could invent a likely one pretty easily.

 

Start with an empty board. Select a desired layout. Pick two valid tile locations (note that mid-air is probably not a valid tile location at this point). Place a tile in each. Repeat until layout is complete.

 

The space picking algorithm will need to follow the reverse rules of playing the game. If you can only take from endpoints or the highest level, then you can only place from where they could be taken. You would want to statistically prefer to place on short rows and bias against nearly full rows, otherwise you may get trapped in a position where you only have two adjacent positions on a single row.  If your layout rules allow odd patterns and holes, you will need to account for that in your space picking algorithm.

 

Worst case is you get stuck and try again with a new random seed, potentially taking a few seconds in a bad situation where many board combinations are tried and rejected.  


Check out my personal indie blog at bryanwagstaff.com.

#5 jbadams   Senior Staff   -  Reputation: 17247

Like
1Likes
Like

Posted 24 March 2014 - 04:39 AM

You'll be starting with a blank play-field, so you'll be building the layers from the bottom upwards by adding pairs of tiles to any valid and available spaces until the board is complete and ready to play.

 

Because tiles in upper layers are placed on top of tiles in lower layers, obviously you can't place tiles into a second (or third, fourth, etc.) layer until the layer beneath them has enough tiles.

 

 

Looking at your example image above, consider that you could not place the bottom-right tile in the second layer until all four tiles that it overlaps have already been placed.

 

 

//EDIT: Beaten to it by frob -- but hopefully between the two of us you'll have an idea of how to approach this.



#6 newwork01   Members   -  Reputation: 112

Like
0Likes
Like

Posted 24 March 2014 - 07:17 PM

Thank for helping me so far , I will try your methods !

 

By the way, I found this algorithm :

http://www.formauri.es/personal/pgimeno/mj/mjsol.html

but i can't understand this algorithm , can you guys help me explain this . Thank a lot guys !



#7 jbadams   Senior Staff   -  Reputation: 17247

Like
0Likes
Like

Posted 25 March 2014 - 03:29 AM


By the way, I found this algorithm :
http://www.formauri.es/personal/pgimeno/mj/mjsol.html
but i can't understand this algorithm , can you guys help me explain this . Thank a lot guys !

Were you aware that the linked page describes an approach for solving existing puzzles rather than for generating puzzles?  We can still help you to understand it if you wish though! smile.png

 

Did you understand any of the explanation, or is there a specific point at which you got stuck?  Are you having trouble with some of the terminology, or do you understand all of that but are stuck on the logic of how the algorithm works?



#8 Bacterius   Crossbones+   -  Reputation: 8158

Like
1Likes
Like

Posted 25 March 2014 - 05:35 AM


Were you aware that the linked page describes an approach for solving existing puzzles rather than for generating puzzles?  We can still help you to understand it if you wish though!

 

If the algorithm to solve an arbitrary puzzle is sufficiently constructive, it is often not hard to reverse it to produce an algorithm to generate a solvable puzzle smile.png (though such an algorithm can probably more readily be obtained by simply reversing the game rules, as frob suggests).


The slowsort algorithm is a perfect illustration of the multiply and surrender paradigm, which is perhaps the single most important paradigm in the development of reluctant algorithms. The basic multiply and surrender strategy consists in replacing the problem at hand by two or more subproblems, each slightly simpler than the original, and continue multiplying subproblems and subsubproblems recursively in this fashion as long as possible. At some point the subproblems will all become so simple that their solution can no longer be postponed, and we will have to surrender. Experience shows that, in most cases, by the time this point is reached the total work will be substantially higher than what could have been wasted by a more direct approach.

 

- Pessimal Algorithms and Simplexity Analysis


#9 jbadams   Senior Staff   -  Reputation: 17247

Like
0Likes
Like

Posted 25 March 2014 - 05:37 AM


If the algorithm to solve an arbitrary puzzle is sufficiently constructive, it is often not hard to reverse it to produce an algorithm to generate a solvable puzzle smile.png  

Absolutely, that's true. smile.png






Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS