A level designer typically creates the levels of a game to cater for a certain set of objectives, or mission. But in procedural content generation, it is common to treat the creation of missions and the generation of levels as two separate concerns. This often leads to generic levels that allow for various missions. However, this also creates a generic impression for the player, because the potential for synergy between the objectives and the level is not utilised. Following up on the mission-space generation concept, as described by Dormans, we explore the possibilities of procedurally generating a level from a designer-made mission. We use a generative grammar to transform a mission into a level in a mixed-initiative design setting. We provide two case studies, dungeon levels for a rogue-like game, and platformer levels for a metroidvania game. The generators differ in the way they use the mission to generate the space, but are created with the same tool for content generation based on model transformations. We discuss the differences between the two generation processes and compare it with a parameterized approach.
LINK
This paper addresses the procedural generation of levels for collaborative puzzle-platform games. To address this issue, we distinguish types of multiplayer interaction, focusing on two-player collaboration, and identify relevant game mechanics for a puzzle-platform game, addressing player movement, interaction with moving game objects, and physical interaction involving both players. These are further formalized as game design patterns. To test the feasibility of the approach, a level generator has been implemented based on a rule-based approach, using the existing tool called Ludoscope and a prototype game developed in the Unity game engine. The level generation procedure results in over 3.7 million possible playable level variations that can be generated automatically. Each of these levels encourages or even requires both players to engage in collaborative gameplay.
Grammar-based procedural level generation raises the productivity of level designers for games such as dungeon crawl and platform games. However, the improved productivity comes at cost of level quality assurance. Authoring, improving and maintaining grammars is difficult because it is hard to predict how each grammar rule impacts the overall level quality, and tool support is lacking. We propose a novel metric called Metric of Added Detail (MAD) that indicates if a rule adds or removes detail with respect to its phase in the transformation pipeline, and Specification Analysis Reporting (SAnR) for expressing level properties and analyzing how qualities evolve in level generation histories. We demonstrate MAD and SAnR using a prototype of a level generator called Ludoscope Lite. Our preliminary results show that problematic rules tend to break SAnR properties and that MAD intuitively raises flags. MAD and SAnR augment existing approaches, and can ultimately help designers make better levels and level generators.
