Performance Comparison of WebAssembly and Phaser in Procedural Maze Generation

Phaser and WebAssembly are both excellent options for web game development, each offering unique advantages. This paper specifically focuses on procedural maze generation and compares the performance of WebAssembly (using C++ and SDL) and Phaser across various levels of complexity, defined by maze dimensions. The study aims to assess execution time efficiency and identify performance trends across different scenarios. Using the Depth-First Search (DFS) algorithm, we conducted multiple test runs at increasing maze sizes to evaluate the scalability and computational efficiency of both technologies. The results indicate that WebAssembly enables near-native execution of C++ code in web browsers, leading to significant performance advantages in larger maze dimensions. Specifically, WebAssembly exhibited a more consistent, linear increase in execution time, making it a strong candidate for handling computationally intensive tasks. In contrast, Phaser performed slightly better in smaller maze dimensions but demonstrated greater variability and steeper increases in execution time as complexity grew. The findings suggest that while Phaser is an accessible and user-friendly framework for game development, its performance may not scale as efficiently as WebAssembly when dealing with complex procedural generation tasks. However, the study also acknowledges limitations, such as fewer test runs for larger mazes and potential differences in implementation efficiency between the two technologies. Overall, WebAssembly appears better suited for tasks requiring high-performance execution, whereas Phaser remains a viable option for less demanding applications where ease of use and rapid development are prioritized.