Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 17 18 [19] 20 21 ... 27

Author Topic: DF v0.43.03+ Worldgen Cookbook Thread!  (Read 86301 times)

PatrikLundell

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #270 on: June 27, 2017, 03:04:58 pm »

Fleeting Frames was faster than I, so there's a substantial overlap...

The PS_ parameters are so called PSVs (Pre Set Values) and are part of vanilla DF (the advanced world gen World Painter produce them, for instance).

Both geomanipulator and regionmanipulator are written against the current DF version. They may work with earlier versions if the DFHack mapping of the layout they use is the same.

Changing geography changing parameter can change the geography to eradicate features completely (including making world where no dorfs can exist). Thus, the amount of change to site abundance is dependent on which parameters you change and how.

Toady recently said the next release will overshoot the one year mark of the previous one slightly (July 5:th). This probably means the next release will appear in the latter half of July or in August (no promises, though). However, if you are dependent upon tools such as DFHack or DT on the LNP (containing the others), and possibly also tile sets, it will take some time for those to catch up, probably a month or two (although DT is a bit iffy, since there is currently nobody who maintains it officially).
Logged

grenedle

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #271 on: June 28, 2017, 12:09:50 pm »

If you change the world's geography, you may change everything that comes after it - pre-set values constrain this, but even the smallest changes I've seen has moved rivers at least 5 local map tiles, with usual effect being erasing their existence entirely in favour of different ones. However, it doesn't seem to always change things - I've increased elevation in one place without it affecting rivers in another place sometimes.

Changing parameter can change the geography to eradicate features completely (including making world where no dorfs can exist). Thus, the amount of change to site abundance is dependent on which parameters you change and how.

So the Fields of Hope won't exist if I generate world and modify the parameters a bit. Unfortunate, but not unexpected. If I don't use the Pre-Set Values, could the DestinedButterfly#730 parameters generate a world with similar locations, or is it still up to chance? At this point, I'm not looking for a specific/perfect single location (although certainly I may want one in the future); I wanted to have a world that has lots of locations with my requirements so that I have options on where I could make a fort.

Added a parameter "megas" check into the above script without which it excludes megabeasts and procedurals, then confirmed adding ogres to gobbos when used before their placement during world gen.

Does this change mean that semi-/megabeasts will show up announced as normal? I don't think I'm ready for megabeasts to show up like normal animals.

Toady recently said the next release will overshoot the one year mark of the previous one slightly (July 5:th). This probably means the next release will appear in the latter half of July or in August (no promises, though). However, if you are dependent upon tools such as DFHack or DT on the LNP (containing the others), and possibly also tile sets, it will take some time for those to catch up, probably a month or two (although DT is a bit iffy, since there is currently nobody who maintains it officially).

I think that I'll probably continue using the present current version until the new LNP is created.
Logged

feelotraveller

  • Bay Watcher
  • (y-sqrt{|x|})^2+x^2=1
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #272 on: June 29, 2017, 05:26:53 am »

For many small biomes you could have a look at what I did here http://www.bay12forums.com/smf/index.php?topic=158646.msg7324834#msg7324834.  Then adapt to your needs.  Beware that this will have basically no lakes/seas (will need water edges as recommended a couple of posts following, or decreasing min elevation below 100 depending on preference).  It also generates pretty deep embarks generally (levels_above... settings). You might also play with the temperature min/max as I was finessing that for precisely what I liked. 

A great resource if you haven't found it already is http://dwarffortresswiki.org/index.php/DF2014:Advanced_world_generation.

It is trivially easy to increase megabeast/titan/semi-mb numbers.  Similarly with civiliations. (Though kobolds and caves go together, don't leave them out in the cold, particularly with extra mb's.) Mineral scarcity=100 gives you more ores than you will ever use... a lot/most embarks will have iron.  It's the towers that will be the tricky part that I can't help with but generally long histories and lots of civ's for conflicts are recommended.  Maybe increasing the evil square count as well, I dunno?  There are a couple of threads that will pop by searching the forum that will be useful, including recently.

I remember seeing some reddit threads (at an earlier date) with some misleading information/assumptions.  Particularly what springs to mind is increasing the subregion count to 20000.  The game caps this at 5000 and increasing it beyond this limit has no effect.  But best to confirm this in practice for yourself as you play with the other variations.

For quickly testing your parameter sets for geography keep the history to the minimum of 2 years to make things go much quicker.  Tweak regenerate, tweak regenerate, until you are pretty consistently getting geographies you like then go for the real ones with longer histories that might get played.  If you find a map with a geography you particularly like it is possible to regenerate it with the main seed but a different history seed for somewhat different (locations can be limited) civilization distribution.  I assume with enough repetitions you might get towers occasionally in desired spots? Best of luck. ;)
Logged

PatrikLundell

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #273 on: June 29, 2017, 06:10:30 am »

To elaborate on feelotraveller's point about short histories: If you find some thing that looks like a really promising start, you can copy the complete seed set and change the history length to the desired one. Similarly when geography is good, but not the rest: copy only the main seed to get the same geography and keep randomizing  the rest. When you're getting closer to what you want you can also increase the history a bit (the early parts don't take too long) and go for longer ones with the complete set for the promising leads.
Logged

Fleeting Frames

  • Bay Watcher
  • Spooky cart at distance
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #274 on: June 29, 2017, 08:09:06 am »

So the Fields of Hope won't exist if I generate world and modify the parameters a bit. Unfortunate, but not unexpected. If I don't use the Pre-Set Values, could the DestinedButterfly#730 parameters generate a world with similar locations, or is it still up to chance? At this point, I'm not looking for a specific/perfect single location (although certainly I may want one in the future); I wanted to have a world that has lots of locations with my requirements so that I have options on where I could make a fort.
Depends on what parameters, but probably it will in some form in #730. I'd almost never call it for certain; I think I've had pretty much unchanged world generation for almost every parameter that doesn't modify all tiles. Here is screenshot of the worksheet I use to keep track of it (had to compress some less important lines to fit on screen).

It will be quite similar, but not quite the same. For sticking with same embark spot/geography and just changing history and underground, I've written over 4k lines(like this - here I'm incrementing pre-set savagery in the path of goblin pits so that history may change and I would get more towers and bigger town. Annoying, but don't know any other way to keep start of history the same (so that the nomadic necro would be born) while changing late one, which is incredibly specific constraint).

One thing not listed in that worksheet - I arrived at the specific setup of layer materials by messing with embark-nearby volcanism (even saw some without aquifers due bottom 2 layers being entirely types clay, but those had fewer sedimentary layers). Cavern parameters are my favourite ones to modify for small to moderate history changes that apply at beginning of world.

But I get the feeling you want broader, biome-level changes. This post has two sceenshots for what level of changes to expect in a pre-set value world. Generally, expect regional tiles to say the same, as the local ones change. A non-pre-set world would have less constraints, so broader changes. But, uh, you probably won't get as many towers - I've seen only five different nomadic necromancers in variants of this world.

As for a world with multiple embark spots - there currently aren't existing tools to add, move and remove regions successfully after world is generated, so you need to consider what spots you want as you generate.
Does this change mean that semi-/megabeasts will show up announced as normal? I don't think I'm ready for megabeasts to show up like normal animals.
Just the existing Hydra and bronze colossus. Tbh, you can add them later in-play during fort mode with region-pops or just before embarking with PatrikLundell's biomediversity, but they won't be changing history then as when added mid-gen.
(Also, I'm not sure if it actually messes with things or the crash was unrelated dfhack crash - path fail messages are pretty common, mostly happening with your own dwarves.)
I think that I'll probably continue using the present current version until the new LNP is created.
Likewise, or on 43.03 for a bit. Not sure the next version will be save-compatible, but 730 doesn't have any kobolds.
It is trivially easy to increase megabeast/titan/semi-mb numbers.  Similarly with civiliations. (Though koboldsand caves go together, don't leave them out in the cold, particularly with extra mb's.) Mineral scarcity=100 gives you more ores than you will ever use... a lot/most embarks will have iron.  It's the towers that will be the tricky part that I can't help with but generally long histories and lots of civ's for conflicts are recommended.  Maybe increasing the evil square count as well, I dunno?  There are a couple of threads that will pop by searching the forum that will be useful, including recently.
Towers are done by mortals bestowed slabs by dwarven and human death gods (just did a quick testgen to confirm that elven and goblin gods won't work for animal people) after some violent events in their life. Then they may go make tower right away, or after a battlefield in an empty land spot where they pick the corpses (which is how you may get multi-tower necros). Any particular death god is not terribly eager to bestow them, usually limiting their numbers to what can be counted on a single hand, so the easiest way to increase their number is increasing the number of dwarf and human civs, or carefully prodding a nomadic necro to hustle - though mind the secret number in that it limits the first.

Neither of those are tricky part - in fact, megabeasts support the existence of towers by causing death and violence. The tricky part is getting that, while maintaining strong, major civilizations who will send you sieges from a town/dark fortress, not some dinky little pit. Lots of megas tends to kill off civs, lots of civs tends to result in large population centers surrounded by small villages, being that existing centers block new ones post-civ placement near them.
« Last Edit: June 29, 2017, 08:51:21 am by Fleeting Frames »
Logged

grenedle

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #275 on: June 29, 2017, 07:37:20 pm »

For many small biomes you could have a look at what I did here http://www.bay12forums.com/smf/index.php?topic=158646.msg7324834#msg7324834.  Then adapt to your needs.  Beware that this will have basically no lakes/seas (will need water edges as recommended a couple of posts following, or decreasing min elevation below 100 depending on preference).  It also generates pretty deep embarks generally (levels_above... settings). You might also play with the temperature min/max as I was finessing that for precisely what I liked. 

A great resource if you haven't found it already is http://dwarffortresswiki.org/index.php/DF2014:Advanced_world_generation.

It is trivially easy to increase megabeast/titan/semi-mb numbers.  Similarly with civiliations. (Though kobolds and caves go together, don't leave them out in the cold, particularly with extra mb's.) Mineral scarcity=100 gives you more ores than you will ever use... a lot/most embarks will have iron.  It's the towers that will be the tricky part that I can't help with but generally long histories and lots of civ's for conflicts are recommended.  Maybe increasing the evil square count as well, I dunno?  There are a couple of threads that will pop by searching the forum that will be useful, including recently.

I remember seeing some reddit threads (at an earlier date) with some misleading information/assumptions.  Particularly what springs to mind is increasing the subregion count to 20000.  The game caps this at 5000 and increasing it beyond this limit has no effect.  But best to confirm this in practice for yourself as you play with the other variations.

For quickly testing your parameter sets for geography keep the history to the minimum of 2 years to make things go much quicker.  Tweak regenerate, tweak regenerate, until you are pretty consistently getting geographies you like then go for the real ones with longer histories that might get played.  If you find a map with a geography you particularly like it is possible to regenerate it with the main seed but a different history seed for somewhat different (locations can be limited) civilization distribution.  I assume with enough repetitions you might get towers occasionally in desired spots? Best of luck. ;)

I've been testing out your Quilt parameters and they look good, as far as I can tell. I don't mind about the lack of seas or lakes really. They've been more of a nuisance than anything, unless there was some sort of benefit that I'm unaware of.

I learned just before you posted about how the Mineral Scarcity numbers work. Before now, I've been generating worlds with MS set to 25000, under the belief that it would be a lot. I felt pretty dumb after figuring that out, especially because I've been complaining about lack of iron ores :P

I tried to upsize the Quilt parameters to 129x129 (with only minor changes to make the parameters similar to the regular Medium worlds) but the worldgen keeps getting rejections because of too many subregions. I've been able to allow that type of rejection and continue worldgen-ing normally, but I was wondering if there was a way to stop the rejections from happening

Also (regardless of world size) when I try to continue worldgen past 2 years (always up to 100 years for test purposes), practically every civilization gets obliterated by megabeasts by year 30. Is this something you had a problem with? Using the Medium version I made (with megabeasts/titans reduced by half), there are still a lot of worlds with very few civilizations. When I upped the number of civs to 60, more survived, but most had >400 individuals each, except for gobbos (who always seem to flourish unless a megabeast gets them). The parameters I originally posted can get civs with 2000-5000 individuals even with only 20-30 civs and 70+ megabeasts. I'm not sure what's causing the difference, but admittedly, I still don't really understand how all the parameters fit together.
Logged

Fleeting Frames

  • Bay Watcher
  • Spooky cart at distance
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #276 on: June 29, 2017, 08:15:56 pm »

Subregion rejection: set it in world_gen.txt to 31000 - some of the limits in in-game gui are soft and this is one of them. For sharing, you want a world without rejections, due worlds with them not generating same way.

There's quite a few differences between quilt and MED MANY SMALL BIOMES, but without generating here's some that jump out just from their respective texts, in order:
MED has larger elevation variance and minimum elevation, but lower other variances; it's ranges are also 25% as prominent and highly focused on making the world more forested as far as biomes go (1000:50 for both rainfall and drainage). All races can eventually build on a neutral forest, though it may take more time...
Quilt has north and south pole, so it's more like 33x65 to MED's 129x129 when it comes to biomes, and makes its biomes further smaller.
Quilt has 240 uninvited guests, MED has 84. Also, megabeasts are blocked off by oceans in MED, plus they generally have a range.
Quilt is 30% evil/good, MED is 11%. Messes with civs on average.
MED has 150 erosion cycles, Quilt has 300, further exacerbating the erosion differences.

And...oh.
Quilt has 18 caves and 132 maximum sites in world (i.e. dozen sites per civ); MED has 300 caves for megabeasts to hide in and 2000 site cap.

PS: Mind that while those 2000-5000 individuals all are potential source for visitors, invasions will only come from 1 site.

grenedle

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #277 on: June 29, 2017, 11:26:24 pm »

Subregion rejection: set it in world_gen.txt to 31000 - some of the limits in in-game gui are soft and this is one of them. For sharing, you want a world without rejections, due worlds with them not generating same way.

There's quite a few differences between quilt and MED MANY SMALL BIOMES, but without generating here's some that jump out just from their respective texts, in order:
MED has larger elevation variance and minimum elevation, but lower other variances; it's ranges are also 25% as prominent and highly focused on making the world more forested as far as biomes go (1000:50 for both rainfall and drainage). All races can eventually build on a neutral forest, though it may take more time...
Quilt has north and south pole, so it's more like 33x65 to MED's 129x129 when it comes to biomes, and makes its biomes further smaller.
Quilt has 240 uninvited guests, MED has 84. Also, megabeasts are blocked off by oceans in MED, plus they generally have a range.
Quilt is 30% evil/good, MED is 11%. Messes with civs on average.
MED has 150 erosion cycles, Quilt has 300, further exacerbating the erosion differences.

And...oh.
Quilt has 18 caves and 132 maximum sites in world (i.e. dozen sites per civ); MED has 300 caves for megabeasts to hide in and 2000 site cap.

PS: Mind that while those 2000-5000 individuals all are potential source for visitors, invasions will only come from 1 site.

Is the subregion rejection the same as [subregion_max:#] in world_gen.txt?

Do you have any suggestions for how to modify Quilt without making it a copy of MED MANY SMALL BIOMES? (When I generate worlds to actually play in, I always set there to be only a North pole).

re:MED's uninvited guests, I haven't been using the displayed in the parameters. I usually set them to megabeasts=70-100, semimegabeasts=200-300, and titans=50+/-10. Even so, I can still generate the amounts of civs (with similar numbers of individuals) that I mentioned. Although you did point out the number of caves. Are UGs predisposed to prefer living in caves rather than wandering about if possible?

re:Caves and sites, I can see how to find the number of caves in a world, but how do you find the site cap?


edit:added rest of QUILT MED parameters. Must have copy/pasted incorrectly before
« Last Edit: June 29, 2017, 11:47:17 pm by grenedle »
Logged

Fleeting Frames

  • Bay Watcher
  • Spooky cart at distance
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #278 on: June 29, 2017, 11:40:59 pm »

1) Yes.

2) Greater or no site cap, more caves. Haven't tried even looking, tho. Can't say which is 'stronger'.

3) They like to have places to live, albeit this tends to be sadly a temporary arrangement.

4) The relevant tag is [SITE_CAP:132] (c+p from quilt). For some reason, what you have in spoiler doesn't have that tag.

feelotraveller

  • Bay Watcher
  • (y-sqrt{|x|})^2+x^2=1
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #279 on: June 30, 2017, 03:44:47 am »

Do you have any suggestions for how to modify Quilt without making it a copy of MED MANY SMALL BIOMES? (When I generate worlds to actually play in, I always set there to be only a North pole).

I linked to Quilt for the 'many small biomes' aspect.  Been a while since I did this (actually tweaked it a bit more looking at my own copy...).  I was making the mistake of thinking medium map = middle size, a pox on the Great Toad for that  :P (not really).  For a medium map I suggest upping the mesh size to 5 (the first number in RAIN_FREQUENCY and SAVAGERY_FREQUENCY).  This should help with maintaining some smaller areas.  The good and evil square counts also need to be increased fourfold to maintain map proportions.  The leaning towards good was my preference but look at them (x4 for medium map) to be 'soft' suggestions for min/max counts.  Reducing them too much will take away from biome diversity, increasing them too much from civ/tower sites.  No real downside to lacking seas/lakes unless you want to embark on them!

For the civilization side: the SITE_CAP includes both civilization sites and beast sites and definitely needs raising, I would start tweaking from the figures used by Med Many Small Biomes.  Similarly for civilization, cave and beast numbers.  And of course TOTAL_CIV_POPULATION.  These are not geographical in the sense of influencing biome size/distribution.  There will be a trade-off here - more sites = less beast attacks but more small towns, so find the right number for the amount of wipeouts you can put up with (sidenote: wipeouts are far more common early in history).  More caves should offset this to some extent.  I would go ridiculously high with the civ_pop to start with (a downside being that legends data might get a bit cluttered - can be offset by switching cull_historical_figures to 1, if needed).

The towers stuff is trickier: once again a no-brainer to set the secret_number to some high number.  Having more neutral forests would allow more sites for human occupation (I think only human civ's? do the necromancer thing).*  (Have a look here as to the what makes a biome a forest http://dwarffortresswiki.org/index.php/Biome.)  More areas of neutral savagery can be somewhat easily influenced by upping the fourth number (i.e., 40-60 savagery) in the savagery_frequency tag... but then you get less areas of high savagery.  This reduces biome diversity to some degree, so it's your call on the tradeoff.  The erosion cycle wears down mountains towards plains (and ups river counts if I remember correct) - you can watch this take place during worldgen - adjust for what suits.

Fleeting Frames is correct about the subregion_max count.  A quote from the advanced world generation page of the wiki :
Quote
It is also interesting to note that the maximum subregions is 5000 which is more than the total number of squares for a pocket or small map. However, for a medium or large map (16641 or 66049 squares) it quickly becomes a mere fraction of the total number of possible subregions. In fact it would be quite easy on a large map to end up with far too many subregions and get endless rejections of this type.

Once you have a basic worldgen sorted I would do something that I learned from PatrikLundell: generate world with short history until a desired geography happens; then save that seed (=main seed only) and regenerate with full length history until you get the distribution of civ's (and towers) that you want.

Oh, and it might be cool to post the worldgen here so that we call al share.  :)


*If dwarfs do it too, then more neutral mountains will be wanted.
Logged

PatrikLundell

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #280 on: June 30, 2017, 05:38:05 am »

Dwarves and humans produce necros. Goblins and elves do not, since they do not die of old age (but I think goblin civs can produce necros, if those are of the right races).
Forest is the elven domain. If you want humans you need plains (hilly or not). Races can expand into territory the cannot start on, though.
Logged

Max™

  • Bay Watcher
  • [CULL:SQUARE]
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #281 on: June 30, 2017, 06:39:08 am »

Oh yeah, since you have the biomes/geography/savagery stuff for post worldgen I thought you might find this handy:
Code: (worldgenedit.lua) [Select]
--Edit certain worldgen parameters while it is running.
local gui = require 'gui'
local widgets = require 'gui.widgets'
local wv = dfhack.gui.getCurViewscreen()
local wd = df.global.world.world_data
local wg = df.global.world.worldgen.worldgen_parms
local stn
local stc
local pop
local eny

genedit=defclass(genedit,gui.Screen)
genedit.focus_path = 'worldgenedit'
function genedit:init()
if not df.viewscreen_new_regionst:is_instance(wv) then
self:dismiss()
qerror("Should be used during worldgen.")
end
self:addviews{
widgets.Label{
view_id="main",
frame = {t=18,l=0},
text={
{text="Quit: Enter/Esc        "},{id="prec", text="Site Num: "}, {id="curs", text=self:callback("getSites")},NEWLINE,
{text="Edit: Left/Right(Fast) "},{id="pres", text="Site Cap: "}, {id="caps", text=self:callback("getCaps")},NEWLINE,
{text="Edit: Up/Down(Fast/Z)  "},{id="ends", text="End Year: "}, {id="endy", text=self:callback("getEny")},NEWLINE,
{text="Edit: PgUp/Dn(Fast)    "},{id="pops", text="Pop Cap: "}, {id="popc", text=self:callback("getPops")},
}
}
}
end

function genedit:getSites()
stn = wd.next_site_id-1
return stn
end

function genedit:getCaps()
stc = wg.site_cap
return stc
end

function genedit:getPops()
pop = wg.total_civ_population
return pop
end

function genedit:getEny()
eny = wg.end_year
return eny
end

function genedit:onInput(keys)
if df.viewscreen_new_regionst:is_instance(wv) then
if keys.LEAVESCREEN or keys.SELECT then
        self:dismiss()
end
if keys.CURSOR_LEFT then
wg.site_cap = wg.site_cap-25
elseif keys.CURSOR_RIGHT then
wg.site_cap = wg.site_cap+25
elseif keys.CURSOR_LEFT_FAST then
wg.site_cap = wg.site_cap-100
elseif keys.CURSOR_RIGHT_FAST then
wg.site_cap = wg.site_cap+100
elseif keys.CURSOR_DOWN then
wg.end_year = wg.end_year-1
elseif keys.CURSOR_UP then
wg.end_year = wg.end_year+1
elseif keys.CURSOR_DOWN_FAST then
wg.end_year = wg.end_year-25
elseif keys.CURSOR_UP_FAST then
wg.end_year = wg.end_year+25
elseif keys.CURSOR_DOWN_Z then
wg.end_year = wg.end_year-1000
elseif keys.CURSOR_UP_Z then
wg.end_year = wg.end_year+1000
elseif keys.STANDARDSCROLL_PAGEDOWN then
wg.total_civ_population = wg.total_civ_population-250
elseif keys.STANDARDSCROLL_PAGEUP then
wg.total_civ_population = wg.total_civ_population+250
elseif keys.SECONDSCROLL_PAGEDOWN then
wg.total_civ_population = wg.total_civ_population-1000
elseif keys.SECONDSCROLL_PAGEUP then
wg.total_civ_population = wg.total_civ_population+1000
end
end
self.super.onInput(self,keys)
end

function genedit:onRenderBody(dc)
self._native.parent:render()
end

genedit():show()
I could add more fields to edit but the ones there are safe to screw with mid-worldgen, not sure about some of the others.
Spoiler (click to show/hide)
Logged

PatrikLundell

  • Bay Watcher
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #282 on: June 30, 2017, 09:05:48 am »

It's a nice feat technically, but I have trouble coming up with situations where you'd want to change the parameters mid world gen, rather than before, as I can't see you'll suddenly realize it at that stage. I guess using standard world gen options you might use this to modify parameters that rather than going for advanced world gen, but the number of users who know enough to do that while still not choosing advanced world gen would be rather small?
Logged

Fleeting Frames

  • Bay Watcher
  • Spooky cart at distance
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #283 on: June 30, 2017, 11:32:50 am »

Dwarves and humans produce necros. Goblins and elves do not, since they do not die of old age (but I think goblin civs can produce necros, if those are of the right races).
Forest is the elven domain. If you want humans you need plains (hilly or not). Races can expand into territory the cannot start on, though.
Not quite. Screenshot of quick 100 elven/goblin civs with 10 megas and 10 titans each, in 696:

(and, jeez, I don't even see the ǐ lot of times, just have straight µ appear)
What happens when you see goblin or elven towers is that the necro last visited that place before creating and becoming the captain of the zombie group (though creating the tower doesn't always immediately follow this). However, while animalmen necros are quite common, they'll still need to have slab granted by a human or dwarven god (or possibly learn it from someone, but while it should be possible I've yet to notice an apprentice learning from someone and then going off to create their own tower.)
Oh yeah, since you have the biomes/geography/savagery stuff for post worldgen I thought you might find this handy:
Interesting. Unlike PatrikLundell, I can think of at lest one situation where I might want to change parameters mid-world gen - to prevent the spread of goblin till x year, via modifying site cap, where savagery = 100 is too weak and gets pits built on. (other way around, past x year, is already covered by site cap, though you perhaps could do allow other sites while limiting theirs with precise timing.)

Though personally, for me the questions I'd want most mid-worldgen are first "who/how many lives in the selected site" and secondly, "is <name> person alive/what they're doing" when dealing with nomadic necros.
Having more neutral forests would allow more sites for human occupation (I think only human civ's? do the necromancer thing).*
If your goal is more humans or dwarves overall, you should go for hilly or flat neutral plains instead of forests. Yes, they can cut down forests, but hillocks and hamlets can be plopped nigh-instantly on low-savagery plains, and while dwarves need mountains to start with, just fortresses with no hillocks get wrecked by FBs; unable to recover with the adjacent support population.
More areas of neutral savagery can be somewhat easily influenced by upping the fourth number (i.e., 40-60 savagery) in the savagery_frequency tag... but then you get less areas of high savagery.  This reduces biome diversity to some degree, so it's your call on the tradeoff.
Hm? Unless you're working on improving a specific set of seeds, I don't think it should - regions don't get cut up by savagery in them, and untamed wilds will almost always have plain more species, i.e. be more diverse, than wilderness or calm.

It, however, would make the spread and placement of sites easier (non-dwarf fort sites can be placed on 50 savagery), so there's that.

Max™

  • Bay Watcher
  • [CULL:SQUARE]
    • View Profile
Re: DF v0.43.03+ Worldgen Cookbook Thread!
« Reply #284 on: June 30, 2017, 03:39:19 pm »

It's a nice feat technically, but I have trouble coming up with situations where you'd want to change the parameters mid world gen, rather than before, as I can't see you'll suddenly realize it at that stage. I guess using standard world gen options you might use this to modify parameters that rather than going for advanced world gen, but the number of users who know enough to do that while still not choosing advanced world gen would be rather small?
There are cases where you are trying to get a certain number of sites placed in certain areas and either reach it before the cap would be hit and want to let history keep going without more sites filling up everywhere, or need more sites to achieve it, want to end on a specific year, want more or less total population.
Logged
Pages: 1 ... 17 18 [19] 20 21 ... 27