Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Trekkin]

I've been monkeying around with world gen trying to build a site that's just absurdly rich in metals, as well as absurdly deep, and I'm not quite understanding what factors influence the material layer depths. I've seen places with no more than 10 volcanism go from soil to granite; I've seen 5 level deep diorite on high-volcanism tiles with little soil. There has to be some rhyme or reason to it, but darned if I can figure out what it is.

So are any parameters, either explicit in world gen or implicit as a result of other such parameters, influence how layers are put together, or is it just random?

[They Got Leader]

Look at this wiki page on advanced world generation. You want to find MINERAL SCARCITY for minerals. I forgot where the elevation variants where.

[Trekkin]

Mineral scarcity I can fix, but my issue is with the relative layer depths, since they control which types of minerals have a chance to spawn.

[i2amroy]

If the thing you are looking for here is just a deep site, then simply turn up the "Levels Above Layer X:" setting, and you will have more stone in-between your cavern layers.

For more minerals you want Mineral scarcity, as outlined above.

For controlling what type of minerals spawn, you can get a little more diversity by decreasing the mineral scarcity option, but for the majority of it is just tied up in the arcane mysteries of the world generator's RNG, and isn't very easily edited. Personally I just suggest that you go geth DFHack and use the "prospect all" command on your sites to see what they have before embarking, then just keep generating worlds until you get one that works.

[They Got Leader]

If the thing you are looking for here is just a deep site, then simply turn up the "Levels Above Layer X:" setting, and you will have more stone in-between your cavern layers.

Aha. That was the command I was missing!

[Trekkin]

Yeah. I guess I should have explained:

I'm trying to build very deep sites rich in every metal in the game. Thus mineral scarcity at 100 and depth at over 100 z-levels total, and I tend to get tens of thousands of squares of native gold, cassiterite, garnierite, and so forth, and that's of course adjustable through depth, since it's igneous intrusive. What trips me up is that sedimentary layers seem to come and go almost at will; what is to all appearances a normal, soil-covered, aquifer-bearing sedimentary layer will turn out to be all granite under the dirt.

So even if there's no way to change the depth of specifically the sedimentary layer, is there some utility that will at least verify its presence from the embark screen?

[i2amroy]

Yeah. I guess I should have explained:

I'm trying to build very deep sites rich in every metal in the game. Thus mineral scarcity at 100 and depth at over 100 z-levels total, and I tend to get tens of thousands of squares of native gold, cassiterite, garnierite, and so forth, and that's of course adjustable through depth, since it's igneous intrusive. What trips me up is that sedimentary layers seem to come and go almost at will; what is to all appearances a normal, soil-covered, aquifer-bearing sedimentary layer will turn out to be all granite under the dirt.

So even if there's no way to change the depth of specifically the sedimentary layer, is there some utility that will at least verify its presence from the embark screen?

As I mentioned, using the "prospect all" command at the embark screen through DFHack should allow you to see exactly what is present there.

[Trekkin]

As I mentioned, using the "prospect all" command at the embark screen through DFHack should allow you to see exactly what is present there.

Reading comprehension fail. That's excellent! Thank you!

[CaptainArchmage]

I think the question is more about increasing the number of sedimentary layers you get. Setting mineral scarcity at 100 increases the number of minerals you get, but doesn't help you get more sedimentary layers. I think that's attached to the Volcanism rating; if the volcanism is between 0 and 20 you get sedimentary layers. Has anyone gotten more than one type of sedimentary layer rock on one embark on the same biome?

[Urist Da Vinci]

SCIENCE!

I made a copy of my DF 0.34.11 folder and removed most of the stone raws except for 1 layer stone of each type (sedimentary, metamorphic, igneous intrusive, and igneous extrusive). These stones were renamed after their types to make reading dfhack easier. The sedimentary stone was also the only flux.

I then generated a world where the volcanic variance was low, and exported the volcanism map from legends mode. On the volcanism map, 255 (white) is high volcanic, whereas 0 (black) is low volcanic. The map had large blobs of light and dark, with grey shades in between.

I then used dfhack prospect to examine sites in the world and compare the layer stones to the volcanism levels from the map. I also tried to note other patterns. The flux finder is not reliable.

Results:
(all results assume dfhack prospect is trustworthy)

1. After surveying the map, I observed that sites in the top 20% of volcanism (200-255) have igneous extrusive, and sites in the bottom 40% of volcanism have sedimentary (0-100).

2. The remainder of the map (100-200 or the upper-middle 40%) can have sedimentary, or metamorphic, or igneous intrusive, but NOT igneous extrusive. The logic here is a bit fuzzy and unknown. I GUESS that the game is randomly choosing from the available stone types, perhaps with some weighting.

3. ALL sites in the world have metamorphic and igneous intrusive underground, below the variable surface layers.

4. Soil depth varies with elevation (i.e. mountain vs swamp), and is not related to stone types.

5. Due to max number of subregions or something, an region might look like it should contain a variety of layer types, but instead the entire region contains one set of layer types. If your volcanic variance is high (perhaps to cause volcanoes to appear worldwide), this might cause regions to assume a higher or lower volcanism than you'd prefer.


If I retested this, I would add a dozen metamorphic stones, but keep only 1 of each of the other stone types, to bias and reveal the presence of random selection at low volcanism.

[Urist Da Vinci]

I repeated the prospecting with a world with full vanilla stones, and a world that had only one type of sedimentary.

1. The top 20% of volcanism leads to areas always having igneous extrusive. The stone layer closest to the surface is always igneous extrusive here. Some of the stone layers underground will also be igneous extrusive. As always, there is metamorphic and igneous intrusive stones deeper underground.

2. The remaining 80% seems to be weighted or affected by other variables. Low volcanism almost always leads to surface sedimentary stone, whereas medium-high volcanism often leads to surface metamorphic or igneous intrusive stone. This isn't a hard rule like result #1.

3. Sedimentary stone can be found anywhere that doesn't have igneous extrusive, with varying probability. However, you always only get 1 layer of 1 sedimentary stone per region, so your embark should straddle different regions if you want more than 1 type of sedimentary stone on the map. The sedimentary stone layer is typically 4-6 z-levels thick, regardless of where in the world it is found.

[Trekkin]

Urist da Vinci, your science is fantastic as always. When you say the top 20% of volcanism, do you mean 80-100 or the top 20% of whatever the range is set to? Because the latter might explain my 0-10 volcanism worlds going from dirt to granite.

[Urist Da Vinci]

Urist da Vinci, your science is fantastic as always. When you say the top 20% of volcanism, do you mean 80-100 or the top 20% of whatever the range is set to? Because the latter might explain my 0-10 volcanism worlds going from dirt to granite.

I meant 80-100, which also happens to be the near-white on the volcanism map (200-255 luminance).

*generates a 0-1 volcanism world*
The volcanism map is completely black, but I can find patches where there is no sedimentary stone when prospecting with dfhack.
*examines other exported maps from legends mode*
*repeats for 30-31, 50-51, 75-76, 80-100 volcanism*

OK, it appears that for any volcanism level under 80, right down to 0 volcanism, it is most likely to be sedimentary but there is a chance that it will have metamorphic or igneous intrusive rock at the surface. It also appears that the chance is independant of volcanism, elevation, rainfall, biome, or other easy to detect parameters.

I don't get this thing.

[GreatWyrmGold]

We need some help from Toady. I suppose that's not a surprise, but still.

[Urist Da Vinci]

We need some help from Toady. I suppose that's not a surprise, but still.

I generated a world in 0.31.18, since that was the last version where the site finder told you what layer stones were present, and compared layer stones to the volcanism map (no dfhack required). In the versions after 0.31.18 the mineral occurrence was changed, but the layer stone selection appears to use the same algorithm. It looks like low volcanism doesn't automatically imply sedimentary stone, and that volcanism only matters for volcano placement and igneous extrusive rock. The 80-20 rule for igneous extrusive rock also holds in that version.

Yes, to get any better idea of how this works, we'd have to get the word of Toady (or Quietust's disassembly).