GEEvo: Game Economy Generation and Balancing with Evolutionary Algorithms

TL;DR

GEEvo (Game Economy Evolution), a framework to generate graph-based game economies and balancing both, newly generated or existing economies, is proposed and a case study evaluating damage balancing for two fictional economies of two popular game character classes is conducted.

Abstract

Game economy design significantly shapes the player experience and progression speed. Modern game economies are becoming increasingly complex and can be very sensitive to even minor numerical adjustments, which may have an unexpected impact on the overall gaming experience. Consequently, thorough manual testing and fine-tuning during development are essential. Unlike existing works that address algorithmic balancing for specific games or genres, this work adopts a more abstract approach, focusing on game balancing through its economy, detached from a specific game. We propose GEEvo (Game Economy Evolution), a framework to generate graph-based game economies and balancing both, newly generated or existing economies. GEEvo uses a two-step approach where evolutionary algorithms are used to first generate an economy and then balance it based on specified objectives, such as generated resources or damage dealt over time. We define different objectives by differently parameterizing the fitness function using data from multiple simulation runs of the economy. To support this, we define a lightweight and flexible game economy simulation framework. Our method is tested and benchmarked with various balancing objectives on a generated dataset, and we conduct a case study evaluating damage balancing for two fictional economies of two popular game character classes.

Figures (4)

• Figure 1: The process of GEEvo is two-step: First, a game designer models an existing game economy using our simulation framework or creates one with the generator. Second, the designer sets an objective based on which the balancer then optimizes the economy graph's weights. In an iterative process the designer evaluates the weights found and, if needed, may reconfigure the balancer. This may involve specifying static weights for e.g., a particular narrative context or enhancing the influence of probabilistic elements within the economy.
• Figure 2: Example of a game economy using the proposed framework and two simulations of it, each with a different configuration. The graph (a) shows the economy from the game Minecraft to craft torches from wood and coal in an automation setting. (b) and (c) show the monitoring of the pool nodes simulating the economy in (a). By only changing the amount of coal needed to craft torches, the entire economy behaves differently.
• Figure 3: The structure of the balancer in detail: The balancer iteratively optimizes an economy's weights toward a balancing objective $x$. Therefore, it adjusts the weights based on the fitness $f_t$ per time step $t$ in relation to $x$ through crossovers and mutations. $f_t$ is calculated based on the result $s_t$ of multiple simulation runs of the economy.
• Figure 4: Economy graphs for the case study to balance the damage dealing of a mage (a) and an archer (b). The values to be balanced are the weights on the edges. Fixed values are represented by absolute values.