Realistic Planetary Simulation

A while back I started prototyping planet simulation, but while digging through research papers I hit roadblocks, and eventually, life got in the way. This thread is about those roadblocks as well as planet simulation in general.

I started trying to simulate plate tectonics. First, reading up on tectonic plates is a special kind of pain as researchers like saying completely different answers with complete confidence. I found out a few things:

  • We don’t know how the tectonic plates first formed.
  • We don’t know when they formed (we have a bunch of wildly different guesses on the scale of billions of years), but we know it was many millions of years after life developed.
  • We haven’t done that much research into tectonic plates on non-earth planets, and earth is the only planet we know to have them.

Now, this is what I gathered from reading some papers. If I missed something feel free to correct me.

This paper here suggests that plate tectonics might not even be needed for life, but other scientists disagree.


Plate tectonics is one of the systems that I think can be handled semi-randomly.

In evolution, there are a lot of patterns of what we expect life to evolve like, so we will need a very detailed algorithm to stay consistent with those patterns.

For plate tectonics, I would presume most players and even most devs don’t know much about how it works, or what patterns they would want to see if follow. At most, I would expect people would expect a plate tectonic simply to do: Continents should move around, sometimes mountains should form, sometimes mountains should shrink again. Etc. So if it’s dynamically creating and separating continents, mountains, islands, etc. then we can get away with doing it semi-randomly behind the scenes.


Agreed completely. You don’t have to stress yourself out covering the nitty-gritty of plate tectonics; as Nick mentions, we want plate tectonics as a means for a dynamic continent and terrain generator. If doesn’t have to be perfect.

If nothing else and in terms of conflicting studies, stick with research that you personally think is logical enough and choose an interpretation you think would lead to interesting gameplay.