Rational Level Design as design process optimization

A tool for decision-making in early game design

As development schedules become shorter and the competition in the gaming industry rises , there is a dire need for new strategies to optimize workflows and design approaches. This thesis attempts to tackle this problem by using the recently emerging technique of Rational Level Design (RLD), a guideline and evaluation tool. It is suggested to use RLD in the early stage of game development, applying its technique as a simulation tool to predict the impact of design decisions. Due to the rational approach, the simulations return measurable, comparable and comprehendible data, that allows for decisions to be backed up with hard facts. The application of RLD in game development is explained, followed by the exemlrary evaluation of the project „Draw Cowboy“ that was realized using this approach. In a critical evaluation is is discussed whether and how development errors of „Draw Cowboy“ could have been prevented using this method.

In the past decade, computer games have evolved from a niche product to a trendsetting mass entertainment. In form of apps and mini-games, they have conquered social media, smartphones, portable music players and tablets. This has opened up a new and thriving sector of game development and boosted the traditional sector. Large scale (AAA) games have budgets of several million euros and teams of several hundred people, that sometimes work for years on the same game. Apps and minigames have low budgets, small teams and tight development scedules. This rapid development calls for new strategies to optimize the development workflow, reduce errors and improve the gameplay experience. This thesis attempts to modify the method of Rational Level Design (RLD) to tackle this problem. In game development the revision of decisions, is one of the most time-consuming factors. Using RLD to simulate the impact of possible decisions and collect data to back this up, is an approach to optimize the development of games. What does Rational Level Design do? The technique of RLD that is already used in the game industry and has proven its reliability. Focusing on design principles and breaking them down into small, measurable particles, or impact factors, game levels are „decoded“ and can easily be compared. Gameplay elements and mechanics are put into relation and evaluated in their behavior throughout the level or game. This enables the game/level designer to connect sections, that scored badly during playtesting, with the behavior of gameplay elements and mechanics. Having identified these sections, the sections can be revised and altered to resemble sections that worked well. How can it be used in early game development? Using RLD in early game development (RLD-ED) does not differ much from its use during a later stage. The difference is, that the evaluation will focus on comparing values of different scenarios, for example using a heavily armed, slow enemy versus one that is lightly armed and agile. Comparing the resulting values and curved will visualize the different outcome of both decisions. Having hard facts, it is possible to make an informed decision, that can possibly prevent detours and errors in design. Does it work? As demonstrated in this thesis, RLD-ED has the potential to prevent bad decisions, when done thoroughly and correctly. As RLD-ED has to be done for a larger amount of scenarios, it is probable that development time will be shortened due to the amount of work that is neccessary. However, considering the fact that detours can make up for 25% of the development time or more, it is an approach worth considering. RLD is a technique that is not affiliated with any software, making it an affordable option for small studios.

Masterarbeit von Suzanne Schanda, betreut von Prof. Jens Müller und Prof. Dr. Thomas RIst