Baumeister T, Hofmann P, Klein L, Fey D (2025)
Publication Language: English
Publication Type: Conference contribution
Publication year: 2025
Graphics Processing Units (GPUs) have become essential in various fields beyond gaming, especially in parallel computing and artificial intelligence. Teaching GPU architecture and programming concepts presents unique challenges in higher education due to their complexity and the black-box nature of traditional approaches. This paper presents µGPUSim, a block-based, gamified GPU simulator designed to facilitate GPU programming education. The simulator employs a simplified architecture and visual programming approach, eliminating syntax barriers and providing immediate visual feedback. We describe the design and implementation of the simulator using the Godot game engine, as well as its gamification elements including achievement badges, star ratings, and persistent progress tracking. The simulator was evaluated across multiple student cohorts in computer architecture courses at our university. Results show high acceptance rates and improved motivation among students, with the gamified elements significantly contributing to their learning experience. We discuss the balance between educational content and gamification, and provide insights into further improvements based on student and instructor feedback.
APA:
Baumeister, T., Hofmann, P., Klein, L., & Fey, D. (2025). Block-Based Learning: A Gamified Simulator for GPU Programming Education. In Proceedings of the 4th EAI Conference on Computer Science and Education in Computer Science. Bratislava, SK.
MLA:
Baumeister, Tobias, et al. "Block-Based Learning: A Gamified Simulator for GPU Programming Education." Proceedings of the 4th EAI Conference on Computer Science and Education in Computer Science, Bratislava 2025.
BibTeX: Download