UC Santa Cruz

In this thesis, I describe a Monte Carlo Tree Search (MCTS) powered tool

that I created to help assess the impact of various design choices for in-development

games built on the Unity platform. MCTS shows promise for playing many games,

but the games must be engineered to oer a compatible interface. To circumvent this

obstacle, I developed a support library for augmenting Unity games, as well as exper-

iment templates in Jupyter Notebook for running machine playtesting experiments. I

also propose ways for designers to use this tool to ask and answer designs questions. To

illustrate this, I successfully integrated the library with It's Alive!, a game I am cur-

rently developing, as well as 2D Roguelike, an open source tutorial game available from

the Unity asset store.3 The integration took fewer than 100 lines of code (see Appendix

B). I demonstrate the tools capability to answer both game design and player modeling

questions, as well as provide the results of the system validation experiments.