Generating an attribute space for analyzing balance in single unit RTS game combat

Achieving balance in a large real-time strategy (RTS) game requires consideration of all contributing factors. We propose an attribute space representation as a common framework for reasoning about balance in the combat scenarios found in these games. Attribute space search is used to identify balanced configurations and game features such as special abilities and terrain act as modifiers to a basis attribute space. The approach is evaluated through the development of an attribute space representation for single unit conflict. An efficient and accurate predictive model is developed corresponding to a simulation of a common RTS combat representation employing the typical attributes of range, speed, health and damage. This attribute space is used to relate the properties of each attribute to the resulting game balance and to identify new balance points when individual attributes are enhanced.