Strategy computer games are nowadays a very popular and exciting genre in the world of computer games.
Many succeeding commercial games were developed since the end 1980's and contributed to the growth and interest in computer games in general and strategy games in particular. Games such as Command & Conquer™ series (by Westwood and EA Studios), Civilization™ series (by Sid Meier), Warcraft™ and StarCraft™ series (by Blizzard) entered to the computer games' hall of fame, thanks to their inventiveness, artificial intelligence challenge and visual effects that they offer.
Almost every strategy computer game was based on the idea of an "electronic board game", a modern brother to the classic, "physical" board games such as chess, checkers,
backgammon, hexxagon and more. An interesting issue is the fact that both the classic board games and the modern strategy computer games are sharing many of the key elements that
make the players think and act strategically and tactically, according to the development of the game.
This issue is the basis for the project, and discussed further in the next pages. It is common to divide strategy games into two main types: "Abstract strategy", where there is "perfect information" regard to the game's state. An example for an "Abstract strategy" is Chess.
The other type is "Concrete strategy" where there is "incomplete information" regard to the game's state. This characteristic makes the game more interesting and surprising. An example for such a game is "Stratego".
This project aims to serve as an open source code framework, written under Microsoft .NET, for easy creation and expansion of "abstract strategy" games by providing operational
artificial intelligence algorithms and well-defined class libraries based on concepts taken from "the game theory" for decision making aspects.
PLEASE NOTICE: This document contains the complete theoretic information regard the subject. Yet, the source code of the project is NOT part of the document as well as the API manual. Please contact the author regard this extra materia.
Table of Contents
1. ABSTRACT
2. PROJECT GOALS
3. INTRODUCTION TO STRATEGY COMPUTER GAMES
3.1 Tools and Technologies
3.1.1 The Microsoft .NET Framework
3.1.2 Available Literature
3.1.3 Overview of Existing Open Source Projects
3.2 Strategy Games Theoretical Concepts
3.2.1 Turn-Based Strategy Games: Classic Model Overview
3.2.2 Modern Approach for Turn-Based and Real-Time Strategy Games
3.3 Artificial Intelligence Algorithms and Concepts for Strategy Computer Games
3.3.1 Choosing the Right Algorithm
3.3.2 Theoretic AI Concepts
3.3.3 Algorithms
4. FURTHER DIRECTIONS
5. BIBLIOGRAPHY
6. APPENDIX: THE IPHA API
6.1 Disclaimer for the code and the document
Project Objective and Core Themes
This work aims to develop an open-source framework using the Microsoft .NET environment to facilitate the creation and expansion of abstract strategy games, integrating artificial intelligence and decision-making concepts.
- Fundamental concepts of classic and modern strategy computer games.
- Implementation of artificial intelligence algorithms inspired by game theory.
- Development of multi-threaded .NET class libraries for strategy game construction.
- Methods for pathfinding, obstacle avoidance, and tactical decision-making in gaming environments.
- Integration of resource management and technology-based progression systems.
Excerpt from the Book
3.2 Strategy Games Theoretical Concepts
Turn-based strategy games, often abbreviated as "TBS", are an old genre of entertaining instrument with a long history of development and changes. Although chess (at least the so called "Western" chess variation) is the most famous and considered a historic representation for TBS game, we can find much older sources – back to 3100 BC (Predynastic Egypt period), such as the "Senet" game evolved by the ancient Egyptians.
Maybe the main reason that TBS games were developed especially in nationalistic and militant cultures is that TBS games force you to "think twice" before you order your warriors to do the next step. You always keep in mind that if you lose, the "assets" you're holding will be lost, and thereupon be remembered eternally in disgrace...
A very important keynote that will help you survive and eventually win the game is the fact you can always see the current state of the game. That is, where is yours and your opponent pawns are placed, how many pawns you and your opponents left, what are the risks that each of yours and your opponents pawns face with, what are the legal moves you and your opponent can do next, be aware to size of the "battlefield" and its restrictions (which is not being modified during the entire game) and so on. This type of knowledge is known in the world of artificial intelligence (AI) as "perfect information", where you have a complete picture regard the current state of all parameters of the game.
Summary of Chapters
1. ABSTRACT: Provides an overview of the popularity of strategy games and outlines the project's goal to create an open-source .NET framework for "abstract strategy" games.
2. PROJECT GOALS: Details the specific objectives of the author, including describing game fundamentals, AI algorithm implementation, and creating multi-threaded .NET class libraries.
3. INTRODUCTION TO STRATEGY COMPUTER GAMES: Serves as the technical and theoretical foundation, covering tools, literature, historical context of TBS games, and complex AI algorithms like pathfinding.
4. FURTHER DIRECTIONS: Discusses the "Seven Layers Model" of modern strategy games and identifies potential future improvements for the framework, such as editor tools.
5. BIBLIOGRAPHY: Lists the academic and industry sources referenced throughout the document, including key AI and game design literature.
6. APPENDIX: THE IPHA API: Contains the necessary legal disclaimer and usage conditions for the open-source code included with the project.
Keywords
Strategy games, Turn-Based Strategy, Real-Time Strategy, IPHA, .NET Framework, Artificial Intelligence, Pathfinding, Game Theory, Algorithm, Decision Making, Pawn, Fog of War, Resource Management, Object-Oriented Programming, Game Development.
Frequently Asked Questions
What is the primary purpose of this documentation?
The document serves as a guide and technical description for the IPHA framework, an open-source set of tools built on Microsoft .NET for developing strategy games, incorporating AI decision-making.
What are the main game genres covered in this work?
The focus is primarily on Turn-Based Strategy (TBS) and Real-Time Strategy (RTS) games, examining their historical roots and modern technical implementations.
What is the core research question addressed by the framework?
The work explores how to bridge the gap between abstract board game concepts and modern computational implementation by creating modular, intelligent agent systems.
Which programming technology is utilized for the implementation?
The framework is developed using the Microsoft .NET environment (C#), making it compatible with Windows and, via Mono, other platforms.
How is AI applied to game decision-making?
The AI uses "utility-based" algorithms and decision trees to evaluate moves, prioritize targets (attack, defense, harvest, build), and manage limited resources effectively.
What are the key technical challenges discussed in the framework?
Major challenges include pathfinding in dynamic environments, implementing "Fog of War", and managing "Next Target to Handle" (NTH) logic for AI agents.
How does the framework handle pathfinding for units?
It utilizes the Bresenham algorithm for simple, obstacle-free paths and an extended A* (A-Star) algorithm for complex, obstructed environments with dynamic risk assessment.
What is the role of "Assets Bank" and "Objectives Bank"?
These are data structures that store the current state of player units and resources, allowing the AI to calculate the utility of various tactical decisions during the game session.
How is the "Fog of War" conceptualized?
It is treated as a limitation of information for AI agents, where map data is only updated and processed when units are in visible range of the terrain.
What defines the "Retreat Under Pressure" (RUP) algorithm?
The RUP algorithm calculates a retreat factor based on a pawn's survival probability, taking into account its current hit points, speed, and asset value versus the perceived threat level.
- Quote paper
- Jacob Barkai (Author), 2009, IPHA - An Open Source Framework for Development of Strategy Games based on Microsoft Technology, Munich, GRIN Verlag, https://www.grin.com/document/132539
