Procedural level generation in 2D roguelite games
Sepänmaa, Tomi (2024)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024120433047
https://urn.fi/URN:NBN:fi:amk-2024120433047
Tiivistelmä
In this thesis, procedural level generation in 2D roguelite games is studied. The brief history of video games and where it all began is explored. Additionally, the birth of the roguelite genre is investigated, and it is examined how a game is defined as a roguelite.
The purpose of this thesis is to familiarize with procedural level generation and explore its use cases. Additionally, different kinds of algorithms, their processes and use cases are examined.
For this thesis, three commercial video games and their procedural level generation are analyzed. From these analyzes, speculations are made about which algorithms are used and how the code works. Additionally, valuable information about KILLBEAT’s level generation was provided through an interview.
In conclusion, procedural level generation is a powerful tool if it is used in the right cases and projects. Procedural level generation is the right tool if enormous amounts of levels are needed, but the contents are controlled by the developers.
This topic offers broad possibilities for further research, such as further developing existing algorithms or even developing entirely new ones. Researching generative AI can also be a significant opportunity for procedural level generation.
The purpose of this thesis is to familiarize with procedural level generation and explore its use cases. Additionally, different kinds of algorithms, their processes and use cases are examined.
For this thesis, three commercial video games and their procedural level generation are analyzed. From these analyzes, speculations are made about which algorithms are used and how the code works. Additionally, valuable information about KILLBEAT’s level generation was provided through an interview.
In conclusion, procedural level generation is a powerful tool if it is used in the right cases and projects. Procedural level generation is the right tool if enormous amounts of levels are needed, but the contents are controlled by the developers.
This topic offers broad possibilities for further research, such as further developing existing algorithms or even developing entirely new ones. Researching generative AI can also be a significant opportunity for procedural level generation.