Nowadays more and more game development company – mostly indie – start to use procedural content generation in their games. This means that they don't create all the content with a development team by hand, but uses a cerfully fine tuned algorithm with some random element instead. This has the advantege of endless replayability offered to the player. And this is many games key feature in the industry.
This unknown element that gives the generation random nature makes possible that the player is always facing new challanges every single time the game starts. This prevents that we never offer rehearsed, repetitive and in the end boring gameplay.
Another big advantege of this technology is that it can learn and adapt to the user playstyle making a better and personalized experience.
This technology used mostly for level/map, loot and enemy encounter generation, but can be used for creating anything in theory.
The task set by this thesis is to implement a procedural dungeon generation software that makes all variables used for generation settable. Thereby allowing the examination of multiple configuration. These maps will be examined by several aspects and compared with each other and with the results of other algorithms in this topic.
The solution must contain a videogame by which we can compare the generated maps with each other.