Connect-4 👾

• This project was generated with JAVAFX, playing Connect-4 with "the computer" or "two players."

Authors:

Habiba Osama zaky

Mai Ahmed Hussien

Basel Ahmed Awad

Ali Hassan Alsharawy

UI

Table of Contents

• User Guide and features
• Heuristic used for Minimax Algorithm

User Guide and features

• First home screen :

  • The user chooses to play with computer or two players

• Second Game scene :

  • The user defines the number of levels for the minimax algorithm max level for without pruning 8 and with pruning 10

  • Choose to play with pruning or without pruning

  

• Third Game Scene :

  • The score of the computer and the player score are shown

  • The Restart button

  • Reset the game and start with the same initial conditions

  • The Back button go back to home

  

  • The showTree button view the tree of the evaluation of the minimax algorithm where the red arrow indicates the chosen column

  

  With pruning the **X** means the this branche is cut off
  

  

Assumptions:

• our game the player continue playing until the board is full

• The cpu is red player

• Consider overlapping as points (if there is five piecies beside each othermeans 2 points)

  

Heuristic used for Minimax Algorithm

• Check every group of 4 in the whole board (horizontally, vertically, in the positive diagonal, int the negative diagonal) and certain combinations of the 4 group has a weight:

  • 4 computer pieces (point) -> 100

  • 3 computer pieces and an empty place (candidate point)-> 20

  • 2 computer pieces and an empty place (candidate point)-> 6

  • 4 opponent pieces (point) -> -150

  • 3 opponent pieces and an empty place (candidate point)-> -60

  • 2 opponent pieces and an empty place (candidate point) -> -6

  • middle column -> 2 for every piece

  • 2 columns beside middle column left and right of it -> 1 for every piece (as the middle col increase the chance for the cpu to win more points)

• Algorithm is defensive

• The evaluation called at the terminal nodes. When the terminal state is reached, we call the evaluateScore method from the Evaluation class. This method evaluates the whole heuristic.