In my previous post, Coding a Game of Memory in Delphi – OOP Model, I’ve been developing the model, aka the back end, for the Memory (Match Up, Concentration, …) game. The idea was to separate the game logic from the user interface (aka the front end). As a result a few classes were introduced: TPlayer, TField and, of course, the main class/object TMemoryGame implementing all the code required to run the game.
Having only the model does not help us much if we actually want to play the game. Therefore, this time, we go into building the user interface in Delphi.
Since the TMemoryGame class is framework agnostic (and not platform specific), it is up for you to decide if you would like to do a classic Windows VCL application, a FireMonkey Android mobile game or something that works on a Mac. To make it quick and simple, I’ll go old-style VCL school.
Memory, Match Up, Concentration … there are many names for a simple card game I’m certain you’ve been playing with your friends at some point during your childhood. I’m also certain you are still playing it from time to time (at least I do with my kids). Just a few months ago, I’ve tried my “luck” against a robot in CosmoCaixa, Barcelona (image).
The rules of the game are simple: cards are laid face down on a surface and two, per turn, are flipped face up. If the flipped cards are a match pair (same looking, same rank, save value) the player claims (wins) the pair and plays again. If they are not a match, cards are flipped face down again, and the next player takes turn. The game ends when all the pairs have been claimed and the player with the most claimed pairs is the winner. If all players have the same number of claimed pairs we can agree to have a tie, or to have the last player be the winner.
I’ve always been a fan of such simple games – from my point of view they are a perfect pick if you want to start learning programming – have fun and sharpen your developer skills at the same time.
While there are Delphi implementations of the game you can find online – most of them have heavily mixed the visual presentation of the game (user interface) with the model (implementation of the game logic).
In my version of Memory, I’d like to separate the user interface (front end) from the game logic (back end) as much as possible. I want to create a game model in OOP style – where the game logic does not interact (or as less as possible) with the front end.
Say you are developing a game like Chess, Go, Checkers, Tic-Tac-Toe or Memory. In each of those games the game board is a rectangle looking playfield of different size (rows x columns). Tic-Tac-Toe is 3×3, Checkers is 8×8, while Go can be 19×19 or 13×13 and similar.
In a game with an arbitrary number of game fields you might want to have the board look as closely to square as possible (rectangle where height and width are the same). Think of Memory. Let’s say we have 24 cards, that is 12 pairs. If you want to place them in a rectangular grid, most similar to square, you would go for 4 x 6 (or 6 x 4) board size (as it would look more square like than 3 x 8 and 2 x 12 or 1 x 24 would be too wide).