Austin Edwards
Game Designer
Overview
Exploring Procedural Generation in Unreal Engine is an independent research project focused on analyzing different procedural generation methods and their effectiveness in level and world design. The study examines three techniques, Voxel Generation, Wave Function Collapse, and Unreal Engine's Procedural Content Generation (PCG), to evaluate their viability, ease of use, and development time. By investigating how these methods have been used in existing games and understanding their core principles, the project aims to assess their practical application in real-world development scenarios.
​
The research emphasizes rapid implementation, mirroring the short turnaround times often required in professional environments. Each method was tested within Unreal Engine to determine its strengths, limitations, and suitability for different design needs. This iterative approach provided insight into how procedural generation can streamline content creation while balancing creative control and technical constraints. Through this study, a clearer understanding of procedural workflows emerged, offering valuable takeaways for developers looking to integrate these techniques into their own projects.
Role: Data Analyst & Programmer
Engine: Unreal Engine 4
Platforms: PC
Duration: 2 Month
Team Size: 1
Contributions
Data Analyst
-
Evaluated the efficiency, usability, and performance of Voxel Generation, Wave Function Collapse, and Unreal Engine Procedural Content Generation (PCG)
-
Analyzed case studies of games utilizing each method to identify trends and best practices
-
Measured implementation time and resource demands to assess feasibility in professional development settings
​
Programmer
-
Implemented and tested procedural generation techniques within Unreal Engine under rapid development constraints
-
Developed tools and scripts to streamline integration and adjust generation parameters dynamically
-
Optimized procedural workflows to balance automation with creative control
Voxel Generation

Basic voxel world generation proved to be straight forward with UE but lacked diversity.

Voxel generation proved to be highly effective in regard to terrain manipulation.

Block size increase was easy but would force heavy shifts in overall game design going forward.

Basic voxel world generation proved to be straight forward with UE but lacked diversity.
Voxel Generation is a highly functional method in Unreal Engine, allowing for procedural terrain that creates a new game world each time the game runs. It significantly reduces development time but comes with limitations that require careful planning, as it often becomes the primary generation method. Testing different voxel scales and amounts helps refine performance and visual quality, while adjusting styles ensures optimal efficiency. Fixed locations can be integrated into randomly generated worlds to maintain structure within dynamic environments. Modifying voxel terrain further enhances flexibility in level design. If its constraints align with a project’s needs, voxel generation is a powerful tool for creating immersive, ever-changing game worlds.
Wave Function Collapse

A building creating using wave function collapse that was spawned through a placement system

Collecting of Wave Function Collapse creating building assets that differing depending on the developer's required conditions

Through a series of colored pipes various placement patterns could be found allowing developer changes as needed

A building creating using wave function collapse that was spawned through a placement system
Wave Function Collapse (WFC) is a promising but under-supported method for procedural generation in Unreal Engine. It can create functional assets and basic levels, but currently, it is limited to structures like buildings and pipes unless significant time is spent developing a custom tileset, which is simply limited by available time. WFC can enhance level design by designing assets for manual or random placement and refining textures for a more natural look. Integrating it with other procedural techniques improves asset variety and player experience. Ongoing testing with personal asset tilesets while monitoring Unreal Engine’s support will help maximize WFC’s potential.
UE Procedural Content Genration

Pathways through forest generated base upon developer requirements

Through the use of external variables like height maps PCG is able to generation certain objects

Basic forest pathways generation with stricter boundaries that needed to be followed

Pathways through forest generated base upon developer requirements
Unreal Engine’s Procedural Content Generation (PCG) is the most well-supported method for quickly filling levels with assets, thanks to its integration within UE5. It enables the creation of navigable boundaries and pathways, ensuring a structured yet immersive environment. Asset placement is optimized for both immersion and performance, while generated landmarks and distinct biomes enhance navigation and world clarity. However, PCG often relies on other Unreal Engine systems, such as terrain manipulation, making it difficult to use independently. For developers looking to efficiently populate and control large-scale asset placement, PCG is an extremely effective tool.
Take Aways
Combining different procedural generation methods can help mitigate each technique’s limitations, creating a more flexible and efficient development pipeline. Initially, the goal was to integrate Voxel Generation for terrain, Wave Function Collapse (WFC) for asset placement, and Procedural Content Generation (PCG) for level design. However, due to each method’s constraints—Voxel Generation restricting development choices, WFC’s asset limitations, and PCG’s reliance on Unreal Engine’s ecosystem—this approach proved unfeasible. Instead, Unreal Engine’s Landscaping and height mapping tools replaced Voxel Generation, while WFC and PCG remained integral to the process.
​
Moving forward, refining these methods is essential, such as improving UV textures for WFC to enhance visual consistency and creating assets specifically designed for procedural generation to avoid implementation issues. Rather than focusing solely on combining methods, dedicating time to understanding each technique independently will provide deeper insight into their strengths and weaknesses. Developing smaller games centered around procedural generation will further highlight their impact on game design, while continued experimentation may reveal new ways to integrate these methods more effectively. By addressing these challenges and exploring innovative solutions, procedural generation can become a more powerful tool for creating dynamic and engaging game worlds.