Throughout my entire playthrough of Galactic Annihilation, I didn't see a single large body of water. Coincidence or is this just a way of removing the buggy naval without removing the buggy naval? I'm not being critical, I'm just curious.
GW uses the same templates as multiplayer defaults, none of which generate water levels over 50. media/ui/main/shared/js/star_system_templates.js
larger planets especially look bad and play bad, because you can see the flaws in the generation of how it over laps and just makes same size lakes everywhere, instead of you know like a big planet would have a huge ocean.
Right now the algorithm for terrain height is just simplex noise. I wouldn't say it's "flawed" because of it, since it's just generic gradient noise, but you're right that it will tend to generate peaks and troughs with a fairly constant frequency, which is pretty uninteresting in my opinion. And then there's this problem, which is a major issue in my opinion, and one that still hasn't really been addressed as far as I can tell. Hopefully all this stuff gets looked at in the future.
Yeah it doesnt look very good, and its not really fun to play on but im sure it will be improved at some point. But then again, most people cant even play those large planets lol. Still some large oceans would lead to some nice naval skirmishes. Oh man that picture, its so annoying when you play naval and there is 2 large oceans like that and they are separated by a little sand dunes like that. But yeah its very odd that it just makes the exactly same layout regardless of size
That's not really the point I was trying to make with that thread, and it also isn't really correct. I'm not very good at explaining that stuff though. Basically, the problem is that we've got tens of thousands of seeds to set only a few variables for every single planet, since it's using the same permutation table for every planet and just changing the gradients based on the seeds. Probably. The implications are that: any topology it can generate can be represented as a linear combination of just a handful of different topologies as a result, most potential layouts are impossible and many seeds will generate nearly the same planet as other seeds (this works at any scale). 16 & 33 are a good example.
