SoulCandy Posted February 23, 2023 Share Posted February 23, 2023 Like most of you, I'm probably MOST excited to see all of the hand-crafted details, easter eggs, and planet surfaces in KSP 2. Design w/ intention - That will always be more valuable to me than what I'm about to suggest. Ok, that said... It would be interesting to start a KSP 2 campaign in a procedurally generated solar system for a whole separate reason - CHAOS! Trying to land on an impossibly close-to-the-sun planet that completes a full rotation around the sun in minutes rather than hours, days, or years. Black holes. Floating islands in the sky poking in an out of cloud-cover. Extraordinarily deep marine trenches breathtakingly high mountain tops. Not to mention, all of the numerous combinations of orbital planes, star sizes, planet biomes, etc etc that would force players to radically adapt to their unique scenario. Just a thought for a distant future release (or mod). Link to comment Share on other sites More sharing options...
TheOtherDave Posted February 23, 2023 Share Posted February 23, 2023 Interesting idea! WRT placing the planets in stable orbits, the problem is that we don't really have a way to know if a system is stable indefinitely. Like, IIRC, we can calculate that (absent a fly-by or something by some other star or planet) our solar system will be pretty stable for the next few million years or so, but at some point the math gets too chaotic to really know for sure. AFAIK, all we can really do is place planets where we've seen them IRL (either in our solar system or others) and run the clock forward for a while to ensure nothing goes crazy too quickly. To be clear, I'm not expecting KSP 2 games to last millions of years — it's just that life probably needs relatively stable solar systems to get going, so you don't want to start in a system that wouldn't have lasted more than a few thousand years since the kerbals have ostensibly been evolving and living there for millions of years. WRT how numerous and how heavy the planets should be, I don't know. WRT how to determine whether a planet should have an atmosphere and such, I don't know, but I'm pretty sure that's something which could be googled. Also, the "frost line" is probably worth looking up. WRT generating the planets themselves, back in the late 90s and early 00s there was a website called something like "the super fun and bouncy and stretchy site" which had a ton of algorithms. IIRC, the algorithm it had for generating planets was roughly: Start with a sphere Pick a random plane that intersects the sphere (it's fine if it happens to go through the center, but don't make them all do that or your planet will be entirely too symmetric) Shift the part of the sphere on one side of the plane in a random direction by a random amount (subject to reasonable constraints) Repeat steps 2-3 probably somewhere between 1k-100k times (IIRC, the more iterations you have, the less it matters what the "reasonable constraints" are in step 3) I don't remember if you can start with shapes other than a sphere for moons or asteroids too small to pull themselves into a sphere, but IIRC it makes pretty nice "raw" planets. It doesn't account for weathering or cratering, nor do I know any algorithms for figuring that out, other than to say they'll both at least partially depend on the planet's atmosphere and age. Also, keep in mind that if there's going to be an ocean or "long-term lake", the surface below sea (or lake) level will weather differently than the areas above sea (or lake) level. I'd imagine you could maybe drop it into existing weather prediction software to do some statistical modeling WRT where rain would normally fall, then use that to inform where you model rivers cutting into the surface and such, but I've never tried it so I don't know. It might also help you determine where things like rainforests or desserts would naturally form, too, but it's not an area I have any expertise in... I just vaguely remember that algorithm. Of course, it only accounts for "height maps", not different soil compositions and such. Also, I have no clue how to know what a "realistic" resource distribution would be, nor how to balance that against what's needed to make the game fun. Link to comment Share on other sites More sharing options...
Bej Kerman Posted February 23, 2023 Share Posted February 23, 2023 Just now, TheOtherDave said: WRT placing the planets in stable orbits, the problem is that we don't really have a way to know if a system is stable indefinitely. Like, IIRC, we can calculate that (absent a fly-by or something by some other star or planet) our solar system will be pretty stable for the next few million years or so, but at some point the math gets too chaotic to really know for sure. AFAIK, all we can really do is place planets where we've seen them IRL (either in our solar system or others) and run the clock forward for a while to ensure nothing goes crazy too quickly. To be clear, I'm not expecting KSP 2 games to last millions of years — it's just that life probably needs relatively stable solar systems to get going, so you don't want to start in a system that wouldn't have lasted more than a few thousand years since the kerbals have ostensibly been evolving and living there for millions of years. When you always hear "rails" and see orbits represented as ellipses, it's easy to forget that there are no rails holding everything together and that the only thing holding planets in their ellipses is a force pulling on them, making no attempt to lock them in an ellipse. Link to comment Share on other sites More sharing options...
Recommended Posts