Paul, Maybe the problem is actually that difficult - there may be no easy solutions to the calculation, and it might indeed take hours to find a valid solution.
But anyway - here's a thought, in case it helps. Have you considered, for each pair: - putting the pair on adjacent empty squares. - moving one (or both!) away from the other until it hits an obstacle or a 'used' point (see next line). Therefore there's a clear path between them. - marking the 'clear' path between them as 'used' Rinse and repeat. Don't know if that's helpful, but you never know. That should eliminate the 'pick all possible squares and pathfind' part of it, I think. Ian On Wed, Mar 11, 2009 at 4:59 PM, Paul Steven <paul_ste...@btinternet.com> wrote: > Thanks Anthony > > Perhaps I am overcomplicating things but I am under the impression I need to > ensure there is a clear path between each pair as I add them. > > My current algorithm is as follows: > > Step 1. Choose a random tile position (Random Row, Random Column) > > Step 2. Create an array of all possible tiles that this random tile can be > paired with. To create this array, I use a path finding function to check > there is a clear path between the 2 tiles > > Step 3. Choose one of the possible tiles from the array (currently I am just > choosing a random one) and update grid data to reflect the 2 paired tiles > are not on the grid. > > Repeat from Step 1 again.... > > This works fine until around 26 pairs have been placed then it gets stuck as > it cannot find any more tiles that have a clear path. > > I have even added code to ensure that no isolated empty tiles are created as > a result of creating a pairing. > > I have tried putting several timed break points in my code to restart the > entire process if it fails to complete the entire grid. However it just > doesn't seem to ever be able to generate an entire grid. > > Perhaps this process is really time consuming and I need to leave it for > hours to generate a grid but I wouldn't have thought so. > > Any advice much appreciated! > > Thanks > > Paul > > > -----Original Message----- > From: flashcoders-boun...@chattyfig.figleaf.com > [mailto:flashcoders-boun...@chattyfig.figleaf.com] On Behalf Of Anthony Pace > Sent: 11 March 2009 16:31 > To: Flash Coders List > Subject: Re: [Flashcoders] Advice on creating random grid of pairs for a > game > > Sounds like one of my old high-school projects, > > I see several ways of doing what you want that make sense: > > * -random population always of an array 81 units long (80 if you > want the last to always be blank) > o --you could create a function that loops 41(or 40 if last > always blank)times tracking the positions in a temp array > and then returning the completed array (if 41st loop > choosing unfilled position with a 0 for blank card/wild card) > o --there are other ways of doing this; yet, they use more > loops and if statements. > * -random population of an array 81 units long and then shuffling > the "deck" sort of > o --shuffling by swapping only positions that have not been > swapped (you could even isolate left and right sides to > simulate a card deck getting shuffled by hand) > o --shuffling by swapping from 1 to 81 to ensure all positions > get swapped > o --there are most definitely more ways of doing this > * -static population and then shuffling the "deck" same as above. > > As well, although I think you may not need to be told this, when you > want to display the positions, you need to know and track which row > level of the grid you are on and what column position you are in, in > order to properly find the x and y values; thus, in your case testing if > the modulus of 9 equals 0 will be handy. > > I hope the concepts help, > Anthony > > > Paul Steven wrote: >> I am trying to write the code to populate a 9 by 9 grid of tiles with 40 >> pairs of matching tiles. >> >> I have 4 different types of tiles and I want to create a grid that has 40 >> matching pairs and 1 blank tile. >> >> I do not want the pairs all to be next to each other but there needs to be >> at least one solution that enables a player to match each and every pair. > To >> match a pair there needs to be a clear path between each member of the > pair. >> >> 1,2,1 >> 2,0,3 >> 4,4,3 >> >> In this simple example, a user could clear all tiles by matching in the >> following order: >> >> 4 - 4 >> 3 - 3 >> 2 - 2 >> 1 - 1 >> >> Note that until they have matched the pair of 2's they cannot match the > 1's >> as there would be no clear path between them. >> >> All my attempts so far to create this grid is resulting in script errors > due >> to my code being unable to find a solution. My code basically gets about > 60% >> of the board created then finds it cannot any more clear paths to create > the >> remainder of the grid. >> >> I would really appreciate any help cracking this function. >> >> Thanks >> >> Paul >> >> >> _______________________________________________ >> Flashcoders mailing list >> Flashcoders@chattyfig.figleaf.com >> http://chattyfig.figleaf.com/mailman/listinfo/flashcoders >> >> > _______________________________________________ > Flashcoders mailing list > Flashcoders@chattyfig.figleaf.com > http://chattyfig.figleaf.com/mailman/listinfo/flashcoders > > _______________________________________________ > Flashcoders mailing list > Flashcoders@chattyfig.figleaf.com > http://chattyfig.figleaf.com/mailman/listinfo/flashcoders > _______________________________________________ Flashcoders mailing list Flashcoders@chattyfig.figleaf.com http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
