Print

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

[GoblinCookie]

Basically speaking I do not like the present caverns much.  I do not see anything interesting about three generic spaces below the whole world that are filled with meaningless randomized detail where there is no sense of direction or actual meaningful development, it is just a place to get lost in and nothing more.  With fully developed kobold sites coming up imminently I feel we need to discuss the whole concept of developing the caverns into something akin to the surface world.

The basic idea here is to overhaul the present way the caverns are generated from the present where the whole world has three generic caverns and wherever you go you will find them to a system where caverns are generated based upon the caves that are randomly placed in the initial map.  During the initial generation of the surface world there are no caverns at all but then onto the surface world are placed caves exactly along the lines of the way that they are at the moment.  Then a new map is generated for the first layer of the caverns, with the map initially being nothing but rock but the other ends of the caverns on the surface are mapped in.  Around these caves are formed cavern biomes with a random size and shape, with their own unique cavern flora and fauna just as with the surface biomes, taken from the raws for the first layer of the caverns.  Sometimes more than one cavern will bump into each-other, especially if their caves on the surface are close together and then it will be possible to travel from one cavern biome to the other.  Then within, around or between the cavern biomes we create underground lake biomes with subterranean water creatures appropriate for the 1st cavern level. 

Then we add in underground passages into the cavern level just as we added in the caves to the surface and we create a 2nd cavern level from the 1st cavern level in the same manner we created the 1st cavern level before creating the 3rd cavern level in the same manner.  In the end we now have a 4 level world map that operates according to the exact same mechanics, we can add in the ability of groups to travel between the levels and we can model a whole world underground, with pretty much everything that happens on the .  We can have sites built and civilizations expand actually in the caverns, caves are no longer sites on the map in which kobolds live, instead the kobolds have sites in the caverns which function like all the other sides and the kobold thieves travel through the cave entrance which is no longer an actual site as such to steal from sites on the surface.  They could also potentially be joined in the caverns by the sites of other races, probably dwarves and goblins and steal from those too or they could head downwards into an even deeper cavern. 

The tricky part is goblin and dwarf fortresses which exist partly in the caverns.  This can however be solved by making those sites exist at all four levels at once, on the surface and each of the three cavern layers.  They take up space in those levels preventing other sites from colonizing there and they also act in all four levels as well, able to colonize and send armies or other parties to any cavern levels that they intersect.  Sometimes however they will be built on solid rock in all the levels are so will not be able to access any caverns at all, in the same manner if the player colonizes such a place he will find that there are no caverns at all and he can dig all the way down to the magma sea without encountering anything.

[Bumber]

AFAIK, caverns are going to be affected by sphere (nature, wealth, fire, etc.) biomes, once implemented. Supposedly even the HFS will be replaced.

[GoblinCookie]

AFAIK, caverns are going to be affected by sphere (nature, wealth, fire, etc.) biomes, once implemented. Supposedly even the HFS will be replaced.

As I understand it the sphere biomes are intended to work like good/evil biomes at the moment.  Since individual caverns now work just like surface biomes if this suggestion was implemented right away we would have evil and good caverns.  So under the upcoming sphere based biomes the caverns would use the same system as surface biomes. 

[SixOfSpades]

Sometimes however they will be built on solid rock in all the levels are so will not be able to access any caverns at all, in the same manner if the player colonizes such a place he will find that there are no caverns at all and he can dig all the way down to the magma sea without encountering anything.

I do dislike the way caverns are so ubiquitous and predictable, but a fortress entirely without them would seem rather boring. I crunched some numbers: Suppose, during worldgen, each Local tile had a 5% of having some Cavern 1, 5% of having any Cavern 2, & 5% for Cavern 3. This would mean that the average 4x4 tile fortress would have a 56% chance of having an open Cavern 1 on at least one of its tiles--with additional 56% chances for Caverns 2 and 3. This would mean the average fortress would still be able to access the fun and excitement of the caverns (particularly if they're looking for them), but finding them would still be a surprise.

Exploring this in a bit more detail . . .
The chances of a given tile having a given cavern layer should not be 5% across the board, but should vary with the types of layer stones present: Softer stones (especially potential aquifer stones) at the appropriate z-levels should be far more likely to harbor cavern spaces than harder stones like granite.
If two adjoining tiles have the same cavern level, then the relevant caverns should connect.
If a single tile has multiple cavern layers, there should be a 25% chance that at least 2 of the layers connect vertically--and not just through chasms: there should be narrow fissures, more common than chasms, naturally occurring at the borders between biomes. (Placing them at fault lines would be better, but until/unless tectonic plates are implemented, biomes will have to do. Fissures are most likely to appear at the border between biomes with very different surface elevations, like where a swamp meets a mountain.) Fissures can (and probably should) occur independently of caverns.
Caverns should only have water if they either a) are connected to the surface in some way, or b) are at least partially roofed with a stone porous enough to allow groundwater to seep in.
The deeper the cavern, the more water (and possibly magma) it is likely to have, perhaps even being completely flooded.
Caverns that do not have a good supply of water cannot have plants / fungi. Caverns that do not have a good supply of plants / fungi cannot have animals. In both cases, organisms should all have their own preferred range of temperatures, humidity, ambient light, savagery, oxygenation, etc., rather than the strict "Cavern 1, 2, 3" system currently used. Cavern flora should also have preferred stone types, as in Plant A might be most commonly found in caves of igneous rock, while Fungus B is usually surrounded by sedimentary stones, etc.
Forgotten Beasts that are spawned in sterile caverns (or any cavern that doesn't have what the FB eats) should create their own tunnels, connecting their original lair with the surface and/or a suitable cavern food source.
And finally, certain settings (like the % chance for a given Local tile to have a Cavern 1, Cavern 2, or Cavern 3, or the modifier by which the hardness of layer stone affects the odds of having a cavern there) should be adjustable in advanced worldgen.


Additionally, given the amazing utility of dwarven beard powers, why can't Miners detect that they are about to breach a cavern? I propose that, along with "wet" stone and "warm" stone, we get "echoing" stone, applied to any unmined stone wall (or floor) that borders a walkable or Open Space tile that has not yet been revealed.

[GoblinCookie]

Exploring this in a bit more detail . . .
The chances of a given tile having a given cavern layer should not be 5% across the board, but should vary with the types of layer stones present: Softer stones (especially potential aquifer stones) at the appropriate z-levels should be far more likely to harbor cavern spaces than harder stones like granite.

There is no set chance as such.  The caverns are grown around the passages that connect them to the surface or the cavern above, there is no chance that there will be a cavern of a given type in a given biome.  You can scroll down to see what caverns (if any) are below you and technically your site exists at all the map levels that there are, even if there are no caverns at all below you just solid rock; you can still choose not to scroll down if you do not want to know.  Another exiting potential of this idea is that it allows for us to easily embark in the caverns as opposed to the surface and then maybe dig our way up the surface instead.  No major differences since the caverns are now modeled as a 2nd world map attached to the to first world map (the surface) and are combined mechnically into a single map only upon onloading. 

It ties into the other plane ideas as well, since another plane map is simply generated around a portal in the same way that caverns are generated around a cave or passage above them.

If two adjoining tiles have the same cavern level, then the relevant caverns should connect.
If a single tile has multiple cavern layers, there should be a 25% chance that at least 2 of the layers connect vertically--and not just through chasms: there should be narrow fissures, more common than chasms, naturally occurring at the borders between biomes. (Placing them at fault lines would be better, but until/unless tectonic plates are implemented, biomes will have to do. Fissures are most likely to appear at the border between biomes with very different surface elevations, like where a swamp meets a mountain.) Fissures can (and probably should) occur independently of caverns.
Caverns should only have water if they either a) are connected to the surface in some way, or b) are at least partially roofed with a stone porous enough to allow groundwater to seep in.
The deeper the cavern, the more water (and possibly magma) it is likely to have, perhaps even being completely flooded.
Caverns that do not have a good supply of water cannot have plants / fungi. Caverns that do not have a good supply of plants / fungi cannot have animals. In both cases, organisms should all have their own preferred range of temperatures, humidity, ambient light, savagery, oxygenation, etc., rather than the strict "Cavern 1, 2, 3" system currently used. Cavern flora should also have preferred stone types, as in Plant A might be most commonly found in caves of igneous rock, while Fungus B is usually surrounded by sedimentary stones, etc.
Forgotten Beasts that are spawned in sterile caverns (or any cavern that doesn't have what the FB eats) should create their own tunnels, connecting their original lair with the surface and/or a suitable cavern food source.
And finally, certain settings (like the % chance for a given Local tile to have a Cavern 1, Cavern 2, or Cavern 3, or the modifier by which the hardness of layer stone affects the odds of having a cavern there) should be adjustable in advanced worldgen.

Additionally, given the amazing utility of dwarven beard powers, why can't Miners detect that they are about to breach a cavern? I propose that, along with "wet" stone and "warm" stone, we get "echoing" stone, applied to any unmined stone wall (or floor) that borders a walkable or Open Space tile that has not yet been revealed.

The great thing here is that it is easy to check the values of what is above since the caves or passages to the level above are generated first.  The cavern biome can easily be generated differently in a swamp rather than a desert because the caves the cavern is generated from are on a given terrain biome already.  The same applies to the rocks, the cave passes through rocks and those rocks can be used to generate the cavern as well.  I do not think I would bother spawning FBs in sterile caverns though, I would simply make FBs require that the cavern be able to support large animals in the first place. 

Echoing stone is a good idea definitely since accidentally digging into a cavern is a pretty major hazard in the game, though it would make the game easier and we need some challenging until invaders get the ability to dig.   

[SixOfSpades]

The chances of a given tile having a given cavern layer should not be 5% across the board, but should vary with the types of layer stones present: Softer stones (especially potential aquifer stones) at the appropriate z-levels should be far more likely to harbor cavern spaces than harder stones like granite.

There is no set chance as such.  The caverns are grown around the passages that connect them to the surface or the cavern above, there is no chance that there will be a cavern of a given type in a given biome.

Oh, I know, I was just posing a worldgen method alternative to yours, where caverns "grow" organically on their own and favor softer rocks, rather than sprout from randomly-placed caves like inverted negative trees.

But now that I think about it . . . from an anthropological / archaeological standpoint, basing dwarven civilizations around caves is clearly their most probable origin, so why should caverns be any different? Early amphibians were successful because they could move from water to land and back as circumstances dictated, catching prey and fleeing from predators as the need / opportunity arose. Similarly, early dwarves would most likely have been successful because (unlike most other races) they could move from the caverns to the surface and back, motivated by similar pressures. The (first) Mountainhome of every single dwarven civ would almost have to be located within spitting distance of a cave leading to a fertile cavern. If Armok is creating a world, and he wants dwarves to be in it, he would create a cavern system that gives dwarves at least a fighting chance, as well as a strong sense of home. So, indeed, why not inverted negative trees? But I would still argue for a large percentage of caverns that are not connected to these trees--both to make the trees less obvious and to make encountering random caverns more of a surprise.

Another exiting potential of this idea is that it allows for us to easily embark in the caverns as opposed to the surface and then maybe dig our way up the surface instead.

Which is, IMO, a possibility that the game has been lacking for far too long (especially since the game still won't let you embark with aboveground seeds). My idea of the "classic" embark would be that you are only allowed to embark underground anywhere within the network of caverns that are at least partially within (or no more than 1 world tile away from) your chosen civilization's home territory. Your wagon stops just inside the edge of the map, at the cavern entrance that is the most direct route back to the Mountainhome. Future traffic to/from the Mountainhome would be via this same cavern entrance (unless you block it off, that is).

I do not think I would bother spawning FBs in sterile caverns though, I would simply make FBs require that the cavern be able to support large animals in the first place.

Fair enough, but allowing FBs to dig from a sterile cavern to a (reasonably close) fertile one promotes cavern interconnectivity, which strikes me as a good thing. Besides, we shouldn't underestimate the flavor value of your Adventurer standing in front of a tunnel that was quite obviously burrowed right through solid rock, by a very large creature . . . with very large claws. G-gulp.

One side effect of that, however, would be that if FBs are allowed to dig during wordgen, they would logically be able to dig during the rest of the game, too, which may or may not be desirable.

[GoblinCookie]

Oh, I know, I was just posing a worldgen method alternative to yours, where caverns "grow" organically on their own and favor softer rocks, rather than sprout from randomly-placed caves like inverted negative trees.

But now that I think about it . . . from an anthropological / archaeological standpoint, basing dwarven civilizations around caves is clearly their most probable origin, so why should caverns be any different? Early amphibians were successful because they could move from water to land and back as circumstances dictated, catching prey and fleeing from predators as the need / opportunity arose. Similarly, early dwarves would most likely have been successful because (unlike most other races) they could move from the caverns to the surface and back, motivated by similar pressures. The (first) Mountainhome of every single dwarven civ would almost have to be located within spitting distance of a cave leading to a fertile cavern. If Armok is creating a world, and he wants dwarves to be in it, he would create a cavern system that gives dwarves at least a fighting chance, as well as a strong sense of home. So, indeed, why not inverted negative trees? But I would still argue for a large percentage of caverns that are not connected to these trees--both to make the trees less obvious and to make encountering random caverns more of a surprise.

Yes, this touches on one of other problems with the development of dwarf civilization I think we may have discussed earlier. They originate from a fortress which straddles the surface and the cavern worlds not from a hillocks on the surface but it takes an advanced civilisation to create such a fortress so how did the dwarves develop a civilisation capable of producing such a thing? The answer here is that the first fortress was not constructed by the dwarves at all, it was a cave between the surface and cavern world the dwarves moved into initially.  They lived in the manner you described exploiting both worlds and then in order to accommodate a larger population in the cave they made expansions to the initial cave creating the first fortress.  They developed the metallurgy to do so because they had ores from the cavern and wood/oxygen/smoke disposal from the surface. 

I do not think that the initial dwarf fortress should be created in the vicinity of a cave, I think the initial dwarf fortress should *be* a cave and that the initial dwarf fortress should replace the cave.  This means they will likely have access to to some underground plants and animals to start with and will be able to set up sites in the underground even if there are no suitable terrain for them to build surface sites in the vicinity.  Further fortresses are not built on the caverns but may be build on the surface OR underground, with them in all cases have access to all levels but the core depot area at a different level. 

I do not agree with should have separate caverns not connected to the surface, that is because any such caverns would have to be lifeless since they have no access to nutrients.  A better way to do it would be to add underground rivers within the initial tree system that then flow downwards into caverns on a deeper level and sinkholes ending surface rivers.  We then model the flow of those rivers through the rock and create caverns around them as their own trees, caverns which can only be accessed by swimming through water so will normally be cut off from surface races that are not [AMPHIBIOUS].  These caverns could feasibly have life in them because the inflow of water from the other caverns or the surface brings nutrients downwards. 

Which is, IMO, a possibility that the game has been lacking for far too long (especially since the game still won't let you embark with aboveground seeds). My idea of the "classic" embark would be that you are only allowed to embark underground anywhere within the network of caverns that are at least partially within (or no more than 1 world tile away from) your chosen civilization's home territory. Your wagon stops just inside the edge of the map, at the cavern entrance that is the most direct route back to the Mountainhome. Future traffic to/from the Mountainhome would be via this same cavern entrance (unless you block it off, that is).

With the suggestion and the introduction of the caravan arc the game would track the caravan returning to your civilisations home sites in the same manner that it would on the surface.  No need for a specific mechanism to restrain you from embarking in particular caverns, we simply set it up that you cannot embark in a place that your civilisation's caravans cannot feasibly reach.  This would apply on the surface as well, so that you cannot embark on an island or another continent, except on the coast if your civilisation has access to a port. 

Fair enough, but allowing FBs to dig from a sterile cavern to a (reasonably close) fertile one promotes cavern interconnectivity, which strikes me as a good thing. Besides, we shouldn't underestimate the flavor value of your Adventurer standing in front of a tunnel that was quite obviously burrowed right through solid rock, by a very large creature . . . with very large claws. G-gulp.

One side effect of that, however, would be that if FBs are allowed to dig during wordgen, they would logically be able to dig during the rest of the game, too, which may or may not be desirable.

FB making tunnels between caverns is one of those ideas that I would support if the devs had infinite resources. 

[SixOfSpades]

They [dwarves] originate from a fortress which straddles the surface and the cavern worlds not from a hillocks on the surface but it takes an advanced civilisation to create such a fortress so how did the dwarves develop a civilisation capable of producing such a thing?

Actually, the ancient city of Jericho provides an answer there. A whopping 9,800 years ago, they built a 12-foot high stone wall to protect the town. They didn't even have pottery yet, but there they were, walking (and presumably fighting from) atop curtain walls that sported stone towers with internal staircases. Now imagine what they might have built if they were dwarves instead of humans.

I do not think that the initial dwarf fortress should be created in the vicinity of a cave, I think the initial dwarf fortress should *be* a cave and that the initial dwarf fortress should replace the cave.

That's arguably the most valid and likely scenario, yes, but there's also room for variations. Sure, maybe the dwarves decide to fortify the cave leading to the cavern rather than the cavern itself, for maximum security, while the cavern is mainly just used for farming. But maybe the dwarves place their fortress at the far end of the cavern instead, in the safest corner, to act more like a citadel of refuge than a gatehouse. Or maybe the dwarves were forced out of the cavern entirely, and set up a 'temporary' surface fort that gradually became permanent . . . when the dwarves finally retook the cavern, they decided to let the surface fortress remain, to impress surface-dwellers with their tenacity. That's why I only said the Mountainhome should be near a cave--I don't care if it's above or below ground, or built around the cave itself, only that it's nearby.

I do not agree with should have separate caverns not connected to the surface, that is because any such caverns would have to be lifeless since they have no access to nutrients.

Whoa, hold up there--while I'm all for realism, applying real-world biology to the caverns would instantly eradicate over 99% of cavern life, and yes that includes ALL caverns, including those open to the surface. Or did you think that a biosphere rich enough to support olm men waging a war to drive the rutherers out from the goblin-cap forest could subsist on nothing more than bat guano, some river detritus, and absolutely no sunlight? No, if the caverns teach us anything, it's that they most definitely do NOT need surface materials to survive. Apparently, all they need is the rock itself, and water to act as a solvent & growth medium.

As we all know, when you're digging your surface fort, all exposed dirt and mud in your fort is just that, dirt and mud--until you breach the caverns, at which point the cavern spores start mixing with your air and taking root in your fort, and soon you've got floor fungus & the like growing everywhere it can. Serious question: Does this occur even if the cavern you've breached is a bone-dry and barren one? If so, then that means a space that has been completely sealed off from all energy and organisms for a very long time (perhaps even forever) is still, somehow, teeming with life, and all it needs is water.

As for caverns (fertile or not) existing without any connection to the surface or other caverns, I for one firmly support them, for the reasons of realism, diversity, and unpredictability. Earth's crust is full of weird little pockets that we have no idea exist until we accidentally drill right into them. Including a wide variety of caverns makes the game more interesting, so by all means have some caverns dry, some wet, and some flooded. Have some connect to a surface cave, and have some exist all on their own, completely invisible until you find them--because that's just how mining works.

. . . No need for a specific mechanism to restrain you from embarking in particular caverns, we simply set it up that you cannot embark in a place that your civilisation's caravans cannot feasibly reach.

Oh, I wasn't proposing that that should be the limit for all kinds of embarks, I was just holding up what would make for a "classic" embark. If you start out in a cavern whose only surface exit(s) is occupied by your own Mountainhome, you are safe from any surface invaders coming down to cause trouble, so all you have to worry about is cavern beasties--which should be minor, since this is, after all, your civ's own home ground. Once your military is trained up, then you can dig for the sky and start building a surface fort to impress the other races and attract their trade--and add to your own civ's surface territory.

[GoblinCookie]

Actually, the ancient city of Jericho provides an answer there. A whopping 9,800 years ago, they built a 12-foot high stone wall to protect the town. They didn't even have pottery yet, but there they were, walking (and presumably fighting from) atop curtain walls that sported stone towers with internal staircases. Now imagine what they might have built if they were dwarves instead of humans.

Building high walls is more a function of muscle power than having really advanced technology, a functional dwarf fortress requires a fair deal of technology, especially if we make mining realistically hard rather than maintaining the present ease of carving through the crust. In any case we are not talking about Jericho, we are talking about the situation before that, basically we are talking about how dwarves have to be more advanced to build the first fortress than any of the others need to be in order to build their first sites.  Either there was previous sites along the lines of the other civilizations which were then cannibalized by the fortress, or the fortress was originally a cave. 

That's arguably the most valid and likely scenario, yes, but there's also room for variations. Sure, maybe the dwarves decide to fortify the cave leading to the cavern rather than the cavern itself, for maximum security, while the cavern is mainly just used for farming. But maybe the dwarves place their fortress at the far end of the cavern instead, in the safest corner, to act more like a citadel of refuge than a gatehouse. Or maybe the dwarves were forced out of the cavern entirely, and set up a 'temporary' surface fort that gradually became permanent . . . when the dwarves finally retook the cavern, they decided to let the surface fortress remain, to impress surface-dwellers with their tenacity. That's why I only said the Mountainhome should be near a cave--I don't care if it's above or below ground, or built around the cave itself, only that it's nearby.

The dwarves cannot set up a fortress the far end of the cavern because they do not yet have the metallurgy to make picks nor any knowledge of mining.  To build a dwarf fortress requires such knowledge inherently which means that the dwarves must have started off living in a cave because it is less of a stretch to go from a cave site to a fortress than it is to build a fortress ex nihilo.  Once you set up in a cave then this location is very desirable but the space is limited.  You have therefore every possible motivation to develop mining skills and technology because it allows you to carve out rooms in the rock around the cave.  Once you start doing this you basically have a fortress along the lines of what we see in adventure mode and at the same time you have developed the ability to make more fortresses entirely artificially whether from the surface or from the surface. 

Whoa, hold up there--while I'm all for realism, applying real-world biology to the caverns would instantly eradicate over 99% of cavern life, and yes that includes ALL caverns, including those open to the surface. Or did you think that a biosphere rich enough to support olm men waging a war to drive the rutherers out from the goblin-cap forest could subsist on nothing more than bat guano, some river detritus, and absolutely no sunlight? No, if the caverns teach us anything, it's that they most definitely do NOT need surface materials to survive. Apparently, all they need is the rock itself, and water to act as a solvent & growth medium.

As we all know, when you're digging your surface fort, all exposed dirt and mud in your fort is just that, dirt and mud--until you breach the caverns, at which point the cavern spores start mixing with your air and taking root in your fort, and soon you've got floor fungus & the like growing everywhere it can. Serious question: Does this occur even if the cavern you've breached is a bone-dry and barren one? If so, then that means a space that has been completely sealed off from all energy and organisms for a very long time (perhaps even forever) is still, somehow, teeming with life, and all it needs is water.

As for caverns (fertile or not) existing without any connection to the surface or other caverns, I for one firmly support them, for the reasons of realism, diversity, and unpredictability. Earth's crust is full of weird little pockets that we have no idea exist until we accidentally drill right into them. Including a wide variety of caverns makes the game more interesting, so by all means have some caverns dry, some wet, and some flooded. Have some connect to a surface cave, and have some exist all on their own, completely invisible until you find them--because that's just how mining works.

I think you rather underestimate the biomass of cavern life that actually exists on earth SixofSpades.  Yes the whole situation where the whole world is full of caverns that are automatically crawling with generic lifeforms that are the same the world over is what I intend to replace with a more meaningful situation where cavern geography is variable, much as the surface is and the main determinant for this is the amount of access to nutrients coming in from the surface, the status quo is basically cavern nutrient max state.  There will be plenty of barren rock caverns about the place, so to add a whole load of completely isolated caverns to the cavern map is really not worth the effort of generating since nobody will be able to ever built sites there and there will be no life there.  They could still be holes in the rock but they should be generated as the rocks are generated not as part of the cavern map; filling them with nasty gas or whatever could actually even give them a purpose. 

Oh, I wasn't proposing that that should be the limit for all kinds of embarks, I was just holding up what would make for a "classic" embark. If you start out in a cavern whose only surface exit(s) is occupied by your own Mountainhome, you are safe from any surface invaders coming down to cause trouble, so all you have to worry about is cavern beasties--which should be minor, since this is, after all, your civ's own home ground. Once your military is trained up, then you can dig for the sky and start building a surface fort to impress the other races and attract their trade--and add to your own civ's surface territory.

That is basically the idea and the AI can do that too.   

[SixOfSpades]

Building high walls is more a function of muscle power than having really advanced technology

Actually, the heavy lifting is by far the least important part of it. Any fool can stack stones on top of each other, but just try stacking them into a 12-foot tower with an internal staircase that's still standing hundreds thousands of years later. That takes intense knowledge of the mass, strength, and frictional properties of various kinds of stone--precisely the things dwarves have in spades. And what's this about advanced technology? I put forth Jericho as an example of how a plausible fortress could exist at very low tech levels.

In any case we are not talking about Jericho, we are talking about the situation before that, basically we are talking about how dwarves have to be more advanced to build the first fortress than any of the others need to be in order to build their first sites.

I can't see any justification for this at all. Apart from the elves' supposed nature affinity (which would make livestock domestication a snap), I don't see a single advantage that the other races would have over dwarves, as far as earliest-civilization fortress development is concerned.

The dwarves cannot set up a fortress the far end of the cavern because they do not yet have the metallurgy to make picks nor any knowledge of mining.

"The far end of a cavern" is, by definition, the farthest you can go without using a pick. But as far as metallurgy is concerned, dwarves would have a head start there as well--who else has more opportunity to see what happens when certain types of stone are exposed to magma?

To build a dwarf fortress requires such knowledge inherently which means that the dwarves must have started off living in a cave because it is less of a stretch to go from a cave site to a fortress than it is to build a fortress ex nihilo.

Mostly correct--but by that reasoning, all the races must have started out in caves, not just the dwarves.

Once you set up in a cave then this location is very desirable but the space is limited.  You have therefore every possible motivation to develop mining skills and technology because it allows you to carve out rooms in the rock around the cave.

While I'd be the last person to argue that dwarves wouldn't want to develop mining & technology, wouldn't it still be easier to just explore the land outside the cavern? Granted, dwarves get cave adapted & all that, but that's not enough to outweigh all those yummy plants & animals on the surface.

I think you rather underestimate the biomass of cavern life that actually exists on earth SixofSpades.

I doubt it. Yes, most caves have a reasonable amount of biodiversity close to the cave mouth. Yes, some caves do have a real influx of nutrients from the surface, like (as you mentioned) those fed from a river sinkhole. But for by far the greatest part of their geographical extent, most caverns are almost completely lifeless. Yeah, there's a few worms, maybe a spider or two. That's a biomass density comparable to, say, an iceberg.

Yes the whole situation where the whole world is full of caverns that are automatically crawling with generic lifeforms that are the same the world over is what I intend to replace with a more meaningful situation where cavern geography is variable, much as the surface is and the main determinant for this is the amount of access to nutrients coming in from the surface, the status quo is basically cavern nutrient max state.

Good. I strongly support cavern diversity--although while I agree that caverns (especially the shallowest ones) should be sensitive to things like surface climate, and any influx of water & nutrients, the cavern's biology should not be determined by these factors. "Realistic" is good, but "Interesting" is better.

Consider this absolutely ludicrous but still intellectually chewy possibility: Rocks are partially organic. Whenever stone comes in contact with water, a small amount of spores are washed off--and different types of rocks produce different types of spores. Blood thorns might be "descended" from obsidian, for example, while plump helmets come from diorite. These spores can take root and grow in any subterranean mud or appropriate soil, but they are more likely to do so when in close proximity to their own parent stone. (Trees are particularly unusual because, as they grow, their oldest wood turns into stone--cut down a large tree and be left with a small stalagmite.) Such a game mechanic would make cavern flora depend on the local mineral strata--and then the cavern fauna would depend on the flora, making for some quick & easy cavern diversity, without even having to artificially stratify organisms by cavern layer.

They could still be holes in the rock but they should be generated as the rocks are generated not as part of the cavern map; filling them with nasty gas or whatever could actually even give them a purpose.

+1 for random gas pockets as well as other surprises, like giant crystals or fossilized FBs or archaeological ruins.

[GoblinCookie]

Actually, the heavy lifting is by far the least important part of it. Any fool can stack stones on top of each other, but just try stacking them into a 12-foot tower with an internal staircase that's still standing hundreds thousands of years later. That takes intense knowledge of the mass, strength, and frictional properties of various kinds of stone--precisely the things dwarves have in spades. And what's this about advanced technology? I put forth Jericho as an example of how a plausible fortress could exist at very low tech levels.

You stack up the stones and then you invent an internal staircase because it is more convenient that climbing all the way to top using a ladder.  You start off with a town without a wall and then you decide you need a wall.  The comparison with Jericho is not apt because there is nothing other than living in a cave in DF terms that would drive a civilization to start building their first dwarf fortress, since otherwise they would simply build a walled settlement along the lines of Jericho instead of building a dwarf fortress. 

I can't see any justification for this at all. Apart from the elves' supposed nature affinity (which would make livestock domestication a snap), I don't see a single advantage that the other races would have over dwarves, as far as earliest-civilization fortress development is concerned.

Justification for what?  The world has people all over the place that started off presumably at a similar level of technology and most of those failed to make much progress above the 'starting point' while only a very few of them got beyond the medieval era. 

"The far end of a cavern" is, by definition, the farthest you can go without using a pick. But as far as metallurgy is concerned, dwarves would have a head start there as well--who else has more opportunity to see what happens when certain types of stone are exposed to magma?

Kobolds and cave pops have the same advantage as well; where are their fortresses?

Mostly correct--but by that reasoning, all the races must have started out in caves, not just the dwarves.

The other races started out either deep inside the caverns or on the surface, having access only to one or the other.  The dwarves live in the caves and they mine the caves into their first dwarf fortresses because they need to make more space than the caves themselves naturally provide. 

While I'd be the last person to argue that dwarves wouldn't want to develop mining & technology, wouldn't it still be easier to just explore the land outside the cavern? Granted, dwarves get cave adapted & all that, but that's not enough to outweigh all those yummy plants & animals on the surface.

The group wants to stay together and is driven to do so by security/economic motivations.  It is a bit like asking why instead of building the walls of Jericho you mention didn't everyone simply scatter to the four winds. 

I doubt it. Yes, most caves have a reasonable amount of biodiversity close to the cave mouth. Yes, some caves do have a real influx of nutrients from the surface, like (as you mentioned) those fed from a river sinkhole. But for by far the greatest part of their geographical extent, most caverns are almost completely lifeless. Yeah, there's a few worms, maybe a spider or two. That's a biomass density comparable to, say, an iceberg.

Yes, the caves far away from any cave entrance or water supply fed for the surface should be lifeless.  This is very much a gameplay balance necessity since otherwise the goblins and dwarves get a very strong advantage over humans and elves; so both realism and gameplay balance are in alignment.

Good. I strongly support cavern diversity--although while I agree that caverns (especially the shallowest ones) should be sensitive to things like surface climate, and any influx of water & nutrients, the cavern's biology should not be determined by these factors. "Realistic" is good, but "Interesting" is better.

Consider this absolutely ludicrous but still intellectually chewy possibility: Rocks are partially organic. Whenever stone comes in contact with water, a small amount of spores are washed off--and different types of rocks produce different types of spores. Blood thorns might be "descended" from obsidian, for example, while plump helmets come from diorite. These spores can take root and grow in any subterranean mud or appropriate soil, but they are more likely to do so when in close proximity to their own parent stone. (Trees are particularly unusual because, as they grow, their oldest wood turns into stone--cut down a large tree and be left with a small stalagmite.) Such a game mechanic would make cavern flora depend on the local mineral strata--and then the cavern fauna would depend on the flora, making for some quick & easy cavern diversity, without even having to artificially stratify organisms by cavern layer.

The rocks are not organic, they are alive.  Organic means something else, which is carbon-based life in particular. 

Organisms would be controlled by biomes like surface animals but the caverns would be divided up according to their access to nutrients, energy, water and oxygen. 

[SixOfSpades]

Jericho, internal staircase, first fortresses, discovering metallurgy

I hope you don't think it's rude, but I'd like to drop the early history angle, at least in this thread--we're arguing too much and it's almost entirely beside the point anyway. We agree that each dwarven civ should be founded in/near a cavern mouth, that's the only relevant bit.

. . . But for by far the greatest part of their geographical extent, most caverns are almost completely lifeless.

Yes, the caves far away from any cave entrance or water supply fed for the surface should be lifeless.

Well, there's a very wide gulf between are and should be, but at least one thing is clear: If DF caverns end up looking like Earth caverns, they're gonna be boring as hell. It should be quite possible for your planned 4-layer worldgen to include 4 layers of rivers, 3 of which exist for the sole purpose of irrigating the caverns, but to me that feels even more strained and artificial than the "cavern magic" that makes them verdant now.

We're agreed that the Embark screen should show ALL pertinent data for the site under consideration--or at least, all pertinent data that would be known to your chosen parent civ. Let's hammer out a few more things.
There should be a greater variety of underground flora, including some that grow underwater (some surface aqueous flora wouldn't be amiss, either). This variety should allow for at least a few extremophile organisms that flourish in places like hot springs. It would also allow for more edible plants--if we decide that different plants/fungi do prefer to grow near certain types of stone, creating more edible flora will mean a greater variety of strata can be inhabited.
There should be at least a couple of simple fungi that can grow both above AND below ground.
I think that as far as worldgen is concerned, caverns should track their [TEMPERATURE], [HUMIDITY], [O2LEVEL], and [CO2LEVEL]. The temperature and humidity would influence the "climate" and thus help determine flora, which would then influence the O2/CO2 balance--caverns with a large plant population would generate more oxygen and allow for the placement of more animals, whereas caverns carpeted mostly in fungi would produce a far less hospitable atmosphere. Caverns low on both O2 and CO2 would be presumed to be filled with some other gas(es), such as methane, helium, or sulfurous fumes--FBs (particularly gas-based FBs) could also influence this.
Whether or not there should be "cavern magic" that makes caverns a rich and biologically diverse ecosystem is something that we (along with a lot more people besides the 2 of us) are just going to have to hash out.

[GoblinCookie]

Well, there's a very wide gulf between are and should be, but at least one thing is clear: If DF caverns end up looking like Earth caverns, they're gonna be boring as hell. It should be quite possible for your planned 4-layer worldgen to include 4 layers of rivers, 3 of which exist for the sole purpose of irrigating the caverns, but to me that feels even more strained and artificial than the "cavern magic" that makes them verdant now.

We're agreed that the Embark screen should show ALL pertinent data for the site under consideration--or at least, all pertinent data that would be known to your chosen parent civ. Let's hammer out a few more things.
There should be a greater variety of underground flora, including some that grow underwater (some surface aqueous flora wouldn't be amiss, either). This variety should allow for at least a few extremophile organisms that flourish in places like hot springs. It would also allow for more edible plants--if we decide that different plants/fungi do prefer to grow near certain types of stone, creating more edible flora will mean a greater variety of strata can be inhabited.
There should be at least a couple of simple fungi that can grow both above AND below ground.
I think that as far as worldgen is concerned, caverns should track their [TEMPERATURE], [HUMIDITY], [O2LEVEL], and [CO2LEVEL]. The temperature and humidity would influence the "climate" and thus help determine flora, which would then influence the O2/CO2 balance--caverns with a large plant population would generate more oxygen and allow for the placement of more animals, whereas caverns carpeted mostly in fungi would produce a far less hospitable atmosphere. Caverns low on both O2 and CO2 would be presumed to be filled with some other gas(es), such as methane, helium, or sulfurous fumes--FBs (particularly gas-based FBs) could also influence this.
Whether or not there should be "cavern magic" that makes caverns a rich and biologically diverse ecosystem is something that we (along with a lot more people besides the 2 of us) are just going to have to hash out.

We can only have 'cavern magic' in a world that has a high level of magic and in that case we simply model the magical sources of stuff the same way we model stuff coming through cave entrances/rivers.  I do not think having a whole load of underground rivers to provide nutrients/oxygen to the caverns is in any way artificial, it is simply the realistic mechanism by which you would end up with lots of caverns that are more 'interesting' than real-life caverns.  The rivers go underground and feed cavern life exists in real-life, we just have this go on on a larger scale than on earth in order to get more cavern life in some places (we also have larger caverns).  We do however actually want a lot of boring earth-like caverns and not too many interesting ones because it keep things balanced as the cave dwelling pops should be balanced against the surface ones. 

Temperature does not have to be modeled separately since realistically (not at the moment) the deeper into the earth you go the hotter it should get.  Cavern plants should not produce oxygen, instead all oxygen should come from the surface since there is no photosynthesis in the dark.  C02 does not really have to be modeled at all since it is not toxic and can build up to extreme levels without causing any harm (as far as I know) provided there is oxygen to breathe.  Humidity is a function of temperature+water really, so deep caverns with plenty of water would automatically be humid without us having to track it separately.  Deeper hotter caverns should have less water in them that shallower colder caverns all else being equal. 

I think that oxygen should be superimposed onto the cavern biomes determined by the amount of water and nutrients, with high altitude areas on the surface also enjoying a reduced oxygen level equivalent to some caverns.  I would go for dotted lines of a certain level, blue line equals plenty of oxygen, yellow indicates scarce oxygen, orange very scarce oxygen and red equals trace to no oxygen.  Cavern creatures (including dwarves) universally have the [LOW_OXYGEN_ADAPT] token that allows them to live without penalty in yellow oxygen areas.  Everything can survive with penalties to their stamina in the oxygen level one step below them, so dwarves can survive with penalties in orange areas but humans will suffocate but humans get penalties but can live in yellow areas.  Everything that does not have [NO_BREATHE] will suffocate in the red areas but things like undead can hole up in those places taking advantage of how nobody can follow them there but they can leave the areas to attack.

The reason for modelling oxgen separately in this manner is we want oxygen to be able to freely change and rapidly as the game circumstances do; for instance digging into a cavern would provide air that would then cause a bubble of oxygen to appear in the cavern, making areas of the cavern accessible and alive when previously they were airless and dead.  Similarly there should be a ton of consequences for certain actions, cavern dwellers should *not* be happy with a site that seals off their only source of oxygen whether a river or a cave entrance or something magical.  Populations, including civilized ones should reduce the available oxygen in a cavern but industry should have the greatest negative effect.  It should be possible to commit genocide effectively in a cavern if you decide to fire up your furnaces inside a cavern but the intelligent cavern creatures whoever they are should not simply die off passively. 

Hence metal should be scarce in caverns, with cavern dwellers preferring to mine ore and sell it to surface sites rather than trying to smelt it themselves.  A greedy cavern dweller that decides to smelt all that ore themselves into metal items should find themselves set upon by the others and possibly involved in a genocidal war.  The result of this however would possible be the mutual destruction of both sides as they both rush to use up the air supply to forge weapons.  In the end both sides suffocate, or perhaps by digging your fortress into the cavern you inadvertently save their lives by providing a new source of air. 

[Qyubey]

I like this. I like this a lot; has good implications for Fortress mode, and BETTER implications for Adventurer Mode. Think about it: currently, adventurers can't explore caverns at all because they're all locked off from the surface. Unless you build a fortress specifically to allow cavern access, you're stuck. Organically building tunnels -- maybe using some of the same code that currently creatures rivers and streams -- allows you to adventure down, deep into the dark depths of the world.

As for laying entities on the map, I say just record Z map data for every site. Just like how we allot X and Y tiles during embark, allow us to assign Z tiles as well, with the default being all 4 Z tiles (surface, first cav, 2nd, 3rd, magma sea). This also leaves potential open for more z levels in the sky; floating islands, mountains, and high altitude buildings like heavenly temples amid the clouds.

A cool aspect about real caves? There's multiple kinds:
https://en.m.wikipedia.org/wiki/Cave

• Soluble Caves: We already record geological rock data, so just setting these caves in tiles that have limestone, dolmite, chalk, etc works. Or maybe adjacent to rivers.
• Primary Caves: Easily done, they occur around volcanoes on the first layer, and any layer with magma on the layers below.
• Sea Caves: Forms on ocean tiles, simple. The tricky part with these would be avoiding them flooding entire cave complexes they connect to, so I recommend they be physically separate from the ocean.
• Corrasional cave or erosional cave: Same as souble caves, but in hard rocks like granite. Also occur around water. Less prominent than soluble caves.
• Glacier cave: Forms on glacier biomes. Pretty simple.
• Fracture cave: Occur in mixed layers of soluble and insoluble rock. Might work for creating deeper layers. Recommend they occur near mountains to imply geological formation along fault lines.
• Talus cave: Change code to generate tons of boulders below cliffs, and holes between them which could punch down into cavern layers.
• Anchialine cave: Same as Sea cave, but containing water pools, both fresh and salt.

This creates a variety for where caves can spawn depending on the tile they're in. You're unlikely to see any caves in dense forests or open plains, but they're often found around open water sources, like rivers and oceans, and beside mountains, for magma caverns and fracture caves. Map tiles would record 'Cave Mouth' in their tiles, like how they do with aquifiers and specific metals.

After forming the entrance, it can also vary the content according to real caves:

• Branchwork caves resemble surface dendritic stream patterns.
• Angular network caves form high, narrow, straight passages that persist in widespread closed loops.
• Anastomotic caves largely resemble surface braided streams with their passages separating and then meeting further down drainage. They usually form along one bed or structure, and only rarely cross into upper or lower beds.
• Spongework caves form a pattern that is three-dimensional and random, resembling a sponge.
• Ramiform caves form as irregular large rooms, galleries, and passages.
• Pit caves (vertical caves, potholes, or simply "pits") consist of a vertical shaft rather than a horizontal cave passage. They may or may not be associated with one of the above structural patterns.

Verbatim from wiki. These structures would allow caverns to physically look distinct as well; whether they form large open Ramiform 'rooms', or thin, winding branchwork tunnels. Given our fantasy nature, I could also see fantasy tunnels being carved -- great open tunnels formed by gigantic worms, or patchwork mines dug out by dwarves, goblins, and other subterranean folk containing old equipment and rotted support pillars.

[Gargomaxthalus]

http://www.bay12forums.com/smf/index.php?topic=76007.0  http://www.bay12forums.com/smf/index.php?topic=76578

  Giving that Toady is currently working on parts of the magic system and that ecology needs to be taken into consideration, these threads could be of great use in this discussion. You also need to keep in mind that there is going to be a special caste of Dwarves living in the caverns that need to be accounted for (sorry if they were mentioned already). It's also very important to keep in mind that creation  myths could end up influencing the structure of high magic worlds, so we'll actually need a number of these systems, not just one.