Automated Bug Detection in Modern Gaming Ecosystems

Xuanhao Hong

doi:10.54097/6gavmk49

Authors

Xuanhao Hong

DOI:

https://doi.org/10.54097/6gavmk49

Keywords:

Bug detection, Machine learning, Game quality assurance, Automated testing.

Abstract

The explosive growth and rapidly growing technology sophistication have turned bug detection into an urgent issue in the modern gaming industry. Poor bug detection may even destroy the player experience and the commercial success. Traditional bug detection approaches are mostly manual-based, and they are inefficient, costly, and ineffective for the exploding size and sophistication of game environments. In this paper, a survey and technical analysis are conducted of four representative game bug detection approaches based on public dataset validation, including text, image, static analysis, and dynamic analysis approaches. The analysis shows that different game types may benefit from different approaches, which have their own pros and cons for different bug types. Meanwhile, different combination approaches of these four types could also reduce the false positive rate effectively. The study provides practical hints for the game quality assurance team. They could trade off the accuracy, latency, and domain applicability.

Downloads

Download data is not yet available.

References

[1] Newzoo. Global Games Market Report 2024[R]. Amsterdam: Newzoo, 2024.

[2] McKinsey & Company. The Economics of AAA Game Development[R]. New York: McKinsey & Company, 2023.

[3] Ariyurek, S., Betin-Can, A., & Surer, E. Automated Video Game Testing Using Synthetic and Humanlike Agents[J]. IEEE Transactions on Games, 2021, 13(1): 50-67.

[4] Drachen, A., Sifa, R., Bauckhage, C., & Thurau, C. Guns, swords and data: Clustering of player behavior in computer games in the wild[C]//2012 IEEE Conference on Computational Intelligence and Games (CIG). 2012: 163-170.

[5] Azizi, E., & Zaman, L. Astrobug: Automatic Game Bug Detection Using Deep Learning[J]. IEEE Transactions on Games, 2024, 16(4): 793-806.

[6] Wilkins, B., & Stathis, K. Learning to Identify Perceptual Bugs in 3D Video Games[EB/OL]. 2022. https://doi.org/10.48550/arXiv.2202.12884.

[7] Nantes, A., Brown, R., & Maire, F. Neural network-based detection of virtual environment anomalies[J]. Neural Comput & Applic, 2013, 23(6): 1711-1728.

[8] Gudmundsson, S. F., et al. Human-Like Playtesting with Deep Learning[C]//2018 IEEE Conference on Computational Intelligence and Games (CIG). 2018: 1-8.

[9] Bergdahl, J., Gordillo, C., Tollmar, K., & Gisslén, L. Augmenting Automated Game Testing with Deep Reinforcement Learning[C] 2020: 600-603.

[10] MacCormick, D., & Zaman, L. Echoing the Gameplay: Analyzing Gameplay Sessions across Genres by Reconstructing Them from Recorded Data[J]. International Journal of Human-Computer Interaction, 2023, 39(1): 52-84.

[11] Xue, F. Automated Mobile Apps Testing from Visual Perspective[C]//SIGSOFT 2020. 2020: 577-581. https://doi.org/10.1145/3395363.3402644.

[12] Shirzadehhajimahmood, S., et al. Using an Agent-Based Approach for Robust Automated Testing of Computer Games[C]//A-TEST 2021. 2021: 1-8.