Publication: A Metric for Automatically Flagging Problem Levels in Games from Prototype Walkthrough Data

Playtesting early and often is important for all game developers, but especially for the growing number of indie teams producing commercial games; however, playtesting game prototypes remains an expensive and time-consuming process. In this paper, we present a new game metric, automatically generated from prototype walkthrough data, which flags problematic levels so that developers know where to invest their effort in fixing the game. Created during the development of the commercial game Angus Hates Aliens, in collaboration with indie developer Team Stendec, our death-related problem level likelihood indicator (DPLI) is interpretable and actionable, i.e., it easily allowed the developer to know where to fix the game levels. Finally, DPLI correlated to enjoyment ratings for the game levels, suggesting that it was a good indicator of problems in the context of our prototype evaluation.

Downloads

PDF

Participants

Max Birk
University of Saskatchewan
Christoph Luerig
Regan Mandryk
University of Saskatchewan

Citation

Birk, M., Luerig, C., Mandryk, R.L. 2015. A Metric for Automatically Flagging Problem Levels in Games from Prototype Walkthrough Data. In Academic MindTrek'15, Tampere, Finland. To appear.

BibTeX

@inproceedings {375-AhA_mindtrek_submitted_cr2,
author= {Max Birk and Christoph Luerig and Regan Mandryk},
title= {A Metric for Automatically Flagging Problem Levels in Games from Prototype Walkthrough Data},
booktitle= {Academic MindTrek'15},
year= {2015},
address= {Tampere, Finland}
}