Real-time player adaptable PCG using gameplay analysis techniques

Abstract

With the ever-increasing popularity of the gaming industry, game development companies are constantly focusing on methods to achieve successful game projects, which typically entail creating unique and engaging experiences for the players. Gaming companies rely on a team of talented and experienced individuals such as game designers, scriptwriters, artists, and programmers to develop high-quality assets and innovative gameplay. However, this approach can lead to several budgeting, management and quality challenges[1]. This project aims to explore and demonstrate methods to mitigate the high reliance on said individuals to create entertaining games by focusing on real-time player adaptable procedural content generation (PCG) using gameplay analysis techniques. PCG encompasses several methods in video game development to automatically generate game content such as maps, levels and loot through algorithms rather than manually creating each component. This project focused on using Markov Chains [2] to generate levels in a 2D Shoot 'Em Up game. The levels created are procedurally generated using the Binary Space Partitioning algorithm to create a dungeon-like map that the player can explore. The levels are generated based on the player's performance in the primary level by considering several level and player parameters using Markov Chains. The game is adapted to either increase or decrease the difficulty by manipulating the probabilities of the content generated to make it a more engaging and playable experience aiming to improve the players' performance leaning towards a somewhat 'optimal' level difficulty. The results obtained from the evaluation that was carried out were then interpreted, showing positive results indicating that the proposed system produced more appealing gameplay tailored to the users' abilities than if there was more controlled content.