The Summit VR

Environment design was highly centered around procedural generation. This picture shows off biome blending, a feature that took a long time to implement, and never actually was fully finished; I generated biomes using a Voronoi diagram, because I wanted their sizes and types to vary randomly. In most (if not all) procedurally generated biomes, an extra layer of noise is added to indicate temperature, another for aridity, so high aridity low temperature environments would be tundra, and low aridity low temperature environments would be mountains, et cetera. I wanted deserts to be next to mountains, so I tried something different; blending a Voronoi diagram with several colors was incredibly difficult to do, though, at the border between more than two different biomes.

Mechanically, the game was inspired a lot by the genre of survival games. More importantly, though, I made sure everything was diegetic; there were no menus, no UI, the only thing close to it was this tablet you could obtain with a mini-map of your surrounding area, and heart and drumstick models under your wrist to display your health. I wanted to make the game as immersive as possible. As you can imagine, this caused some hiccups (which are clear in the picture). I solved some of them; this was actually mid-implementation of my tree-chopping solution, in which I transitioned from chunks of the tree trunk model to a basic, primitive, and standard log "item" that you can see in the picture.

Working alone, I had to do every aspect of the game's development by myself, and this included asset generation. I watched some tutorials, used references, and created every model you see in the game (with the exception of the hands): some of them ended up getting rigged and animated, like this one, a deer.

The game was engineered for VR first and foremost; I kept things generally robust in case I wanted to port to a desktop version, but it was important to me for things to be diegetic and immersive.

Procedural generation was the largest part of this project, and my biggest time investment; not because it was difficult, but because it was really, really fun. It was so gratifying starting with a randomly offset plane mesh and shifting it from that to making mountains, different biomes, and eventually even multithreaded terrain generation on an infinite scale.