In my recent quest for a sweet little map to mess around with, I reached a couple of conclusions, not all of which were particularly intuitive:
a) On the extremely intuitive, uninstructive, but entertaining side, however awesome setting values on the minimum temperature so high the in-game editor won't allow them might be, it's worth noting that murderous but not obscenely high values are fun in that the dwarves survive longer, and in some cases can actually go about their business for some time in temperatures hot enough to melt lead in the shade. (Not sure what the exact values are for STANDARD_FLESH temperature damage, but I have to imagine zero-variance temperature just at the brink of them would be fun.)
b) It seems to be really difficult to predict where sand will show up. The frustrating thing is that, unlike whether layer 2/3 is chert or flint (or indeed what color the sand will be), there's a tantalizing element of non-randomness to it. The only way I've seen posted to guarantee lots and lots of sand is basically to turn the world into Arrakis, which (while fun) has various intrinsic problems. Which leads to the next observation:
c) Erosion seems to have fairly little relationship with sand production, but there is at least one thing it governs above and beyond the map's highs and lows: at extremely low erosion levels, every non-volcanic area will be covered in sedimentary rock, often one or two layers' worth. And oddly enough, at 0 erosion it seems like every mountain biome has a solid layer of rock salt on its surface. What this suggests is that erosion, acting (probably) by itself, wears down terrain both in the physical sense (thinning sediment and making the typical bottom igneous intrusive layer(s) bigger) and along a sort of index for the sedimentary layer. In the zero-erosion world I generated, mudstone and sandstone were also extremely abundant; limestone seemed to be far and away more prominent than chalk or dolomite; and all of this was everywhere except the mountains buried in lots and lots of soil. I was on a specific quest at the time map-wise, so I never wound up embarking in it - but I would bet an extremely low-erosion world would look a lot different than the usual.
(d) On the one hand, forests are pretty consistently worthwhile and keeping conditions favorable for them makes the chances of finding a good embark point without tree growth. On the other hand, it seems like even a fairly slightly greater proportion of non-savage/evil forest than the default strongly tilts the fragile balance between elves and humans strongly in elves' favor. In maps with more or less ubiquitous tree cover, humans die out quickly enough they should count as fanciful, and you're in effect left with a single trading partner for everything that isn't wood.
I've finally got the nice, rich chalk/magma/sand site I was after, but not without repeatedly burning down the world, salting it, and apparently causing no humans to grow there again. What stupid worldgen tricks have you run into?
Author
Topic: Stupid worldgen tricks (Read 981 times)