Procedural generation

From CatchChallenger wiki
Jump to: navigation, search

The map can be generated by Procedural generation but it's possible to plan IA with auto learning based on rating system to generate each time better map.



The normal Heightmap algorithm do a very curved Heightmap. But to maximize the compression and keep the correct style of the map you need drop the curved style.

Some transformation todo to rectuangularize the base terrain layer map


With machine learning/deep learning it's possible to analyse the map structure to after generate a content automatically.

Voronoi diagram

20 points and their Voronoi cells
Map with Voronoi polygon + heigh map = Terrain
Map with Voronoi polygon after pixellisation

In mathematics, a Voronoi diagram is a partitioning of a plane into regions based on distance to points in a specific subset of the plane. That set of points (called seeds, sites, or generators) is specified beforehand, and for each seed there is a corresponding region consisting of all points closer to that seed than to any other. These regions are called Voronoi cells. The Voronoi diagram of a set of points is dual to its Delaunay triangulation. Wikipedia

After you need pixelise it to use it into tile, you can use 1 tile, 2 tile step (or more).

Transition and height


Do the transition border into the zone only do gap between zone.

Based on mask, terrain spot, altitude line like into map.



Some species is into lot of biomes, other into specific biomes. And some is endemic.

Wikipedia quote:

Biodiversity, a portmanteau of "biological diversity," generally refers to the variety and variability of life on Earth. According to the United Nations Environment Programme, biodiversity typically measures variation at the genetic, the species, and the ecosystem level. Terrestrial biodiversity tends to be greater near the equator, which seems to be the result of the warm climate and high primary productivity. Biodiversity is not distributed evenly on Earth, and is richest in the tropics. These tropical forest ecosystems cover less than 10 per cent of earth's surface, and contain about 90 percent of the world's species. Marine biodiversity tends to be highest along coasts in the Western Pacific, where sea surface temperature is highest and in the mid-latitudinal band in all oceans. There are latitudinal gradients in species diversity. Biodiversity generally tends to cluster in hotspots, and has been increasing through time, but will be likely to slow in the future.

Into catchchallenger the simulation can't be do because the species don't evolve for now (it's game, fixed content), can't have selection natural. But we can have correct distribution of the current monster defined by mapweight (endemic variables), luckweight (rate of it into the place).

CatchChallenger support multiple monsters zones, then the kind of monster can be defined by map or with finer granularity by arbitrary group of tile

The problem:

 Put X marker to this numbers
 - The proportion of the number marked with the others number need to be respected
 - The sum of marker * marked number should be near of the median * number of marker