301
DF Gameplay Questions / Re: Creating ☼Advanced World Generation☼ docs, have questions.
« on: April 03, 2011, 01:54:41 pm »
The evil/good thing seems to behave in a strange unintuitive manner.
I think that what you probably want to do if you want, say. 1/5th of the ocean to be evil is go crank up the "minimum final region count" for oceans to at least 5, the minimum ocean square count to however much ocean you want, then set "evil for large regions" to the approximate size of the evil region you want. I *think* that the thing will pick the ocean region that's closest to the size of whatever you set for large region evil squares.
So, I think that if you have large ocean regions with 1500, 2500, and 3500 squares each, and no other large regions other than oceans, and you set evil squares for large regions to 2100, I think it will pick the 2500 region to make evil.
Of course you have to make sure you have enough elevations in the range 0-99 that are separate from each other so you'd probably also need to crank up elevation variance and maybe adjust the mesh to be a large size and increase the weighted range for 0-20.
Of course the parameters don't let you specifically say "only make oceans evil" so you're probably still left to chance and may have to manually reject a few worlds before you get what you want.
As you can see this crap is complicated and that's why I was getting kind of sick of not having good docs to refer to. Heck, I wrote most of the docs at this point and I still need it as a reference.
If you come up with something good, please post it to the Worldgen Examples page talking about how you had to adjust the parameters to get this to work. Sometimes the only way for people to figure out complicated stuff like this is to look at what other people came up with.
Anyway, things are improving but we could still use people to look over the docs and try to identify anything that's incorrect. (For example, I had the mesh size thing wrong and recently fixed it.) So anyone who has time please read over the docs. You'll probably end up learning something anyway....
I think that what you probably want to do if you want, say. 1/5th of the ocean to be evil is go crank up the "minimum final region count" for oceans to at least 5, the minimum ocean square count to however much ocean you want, then set "evil for large regions" to the approximate size of the evil region you want. I *think* that the thing will pick the ocean region that's closest to the size of whatever you set for large region evil squares.
So, I think that if you have large ocean regions with 1500, 2500, and 3500 squares each, and no other large regions other than oceans, and you set evil squares for large regions to 2100, I think it will pick the 2500 region to make evil.
Of course you have to make sure you have enough elevations in the range 0-99 that are separate from each other so you'd probably also need to crank up elevation variance and maybe adjust the mesh to be a large size and increase the weighted range for 0-20.
Of course the parameters don't let you specifically say "only make oceans evil" so you're probably still left to chance and may have to manually reject a few worlds before you get what you want.
As you can see this crap is complicated and that's why I was getting kind of sick of not having good docs to refer to. Heck, I wrote most of the docs at this point and I still need it as a reference.
If you come up with something good, please post it to the Worldgen Examples page talking about how you had to adjust the parameters to get this to work. Sometimes the only way for people to figure out complicated stuff like this is to look at what other people came up with.
Anyway, things are improving but we could still use people to look over the docs and try to identify anything that's incorrect. (For example, I had the mesh size thing wrong and recently fixed it.) So anyone who has time please read over the docs. You'll probably end up learning something anyway....