Skip to main content

Université de Montréal

Organization2 reports

2 reports, presentations and talks by Université de Montréal in the Game Industry Library, the newest “Democratizing Game Modding with GenAI: A Case Study of StarCharM” (2025).

Documents

Cover of Democratizing Game Modding with GenAI: A Case Study of StarCharM
Whitepaper35 pages

Democratizing Game Modding with GenAI: A Case Study of StarCharM

StarCharM serves as a transformative framework for democratizing game modding by leveraging generative artificial intelligence to lower the technical barriers associated with content creation.

  • StarCharM uses GPT-4o and multi-stage prompt chaining to automate the creation of complex JSON configuration files, including NPC schedules, dialogue, and trait-based gift preferences for Stardew Valley.
  • The framework successfully lowers technical barriers for non-technical users, enabling them to create narrative-driven content through natural language inputs and iterative, scaffolded workflows.
  • 2025 evaluations reveal a tension between accessibility and community standards, specifically citing concerns over content homogenization and the production of repetitive, robotic dialogue.
Game DevelopmentAIGame Design+2
Polytechnique MontréalOct 2025
Cover of SyDRA: An Approach to Understand Game Engine Architecture
Whitepaper28 pages

SyDRA: An Approach to Understand Game Engine Architecture

The Subsystem-Dependency Recovery Approach, or SyDRA, provides a structured methodology for navigating the inherent complexity of game engine architectures.

  • The Subsystem-Dependency Recovery Approach (SyDRA) improves developer efficiency and accuracy in architectural tasks by visualizing complex dependency models derived from file-include relationships.
  • Empirical analysis of ten open-source game engines identified that core components, specifically low-level rendering and resource management, frequently exhibit high coupling and poor folder-level cohesion.
  • SyDRA functions as a diagnostic tool that enables teams to predict the downstream impacts of code modifications and identify specific refactoring opportunities more effectively than manual inspection.
Game DevelopmentGame EnginesGlobal
Concordia UniversityJul 2024

Do you work at Université de Montréal? Get a free report of how these documents are read and cited here.