Print

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

[Footkerchief]

So I have over 3 million created, 80k exported wealth, 200+ dwarves and zero nobles. Can you please add a periodic check or something, maybe it could also make sure doctors aren't completely ignoring to diagnoses patient that have been resting in hospital for years.

Here's the bug report for not getting a baron.  It would be helpful to upload your save to DFFD and post the link on that report.

Is your non-diagnosed patient actually resting, or just lying there with "No Job"?  Either way, you should go ahead and upload a save to DFFD (unless the save for the baron problem also demonstrates this one).

[Mel_Vixen]

Of course, if you want to have a large population concentrated in a single area, it's almost impossible to do so without farming or at least a seriously coordinated herding effort (like a cattle drive to keep from overgrazing).  If anything, I'd expect farmers in some kind of underground "muck farm" made of decomposing parts and mud where slaves cultivate a sort of very base, low-quality fungus version of animal fodder that is mass-produced to feed livestock for slaughter if you have a meat-heavy goblin civ menu.  Sounds appropriately goblin-y, no?

Hmmmm and maybe a demon brings some stuff into the goblin culture like errr a noninteligent brute as cow stand-in or a demonic plant .... . How the demons and FBs as "influences from the outside" (and naturally different leaders like kings and priests) influence the overall behaviour of a civ on the worldmap would be interresting to know. Could they change how and where sites are created/expanded/abandoned and alike?

[Armok]

In many games mounts are basically motorbikes with legs, will mounts in DF, especially badly trained or with a low riding skill, be treated more realistically as separate agents that you just give commands? This is especially important with intelligent ones, which should probably be handled more like your companions than like a warhorse.

[NW_Kohaku]

Alright, hope I have enough credit for one more question on this round...

First one is regarding sites - are there plans to move the way sites are generated into the raws, and if so, are there any plans on when it will happen?

I don't have any specific plans or timelines at this point.  Things are obviously still in flux, and I don't know what I need to or can put out there.

That's a bit more sprawl than I expected, though I guess part of that is because presumably the farming villages occupy the same portion of region squares as towns. Maybe if villages could be larger than that? Anyway, it should be interesting to see how this works out for other races. Maybe the sprawl could be controlled in the entity raws as well?

I mentioned in the log that the little 3x3 gray squares are just the area for the village buildings.  Crops and pastures etc. will be spread out from that, forming possibly an 8x8 or a bit larger, I think.  The village buildings aren't precisely for the 3x3 either, although it might start that way.  If cottages end up spread down a single road a little farther than 3 pixels, but in a narrow strip, that would be fine too.  Large towns will also outgrow their 3x3s.

It looks like humans have gotten the most of the recent village development. How are the other races going to handle things? Are we going to see goblin villages and the like? And how are they going to be different from what you're doing with the humans?

The humans have received the attention at this point, but before the next release I'd like to get the dwarf and goblin models up.  I don't want to commit yet, until something is in, but it should be in a log soon.  They will be quite different from each other.  Kobolds will remain mostly unchanged and elves can't receive proper attention until they have their proper trees.

I guess my question in response to this would be, what is your reaction to this thread's suggestion about how we could have raw-modifiable "town pieces" that get procedurally sewn into a "hallway" whose shape and layout have variable guidelines? 

If we have a flexible enough code for what constitutes "hallways", and code for recognizing what needs towns have, and tokens for telling the system what rooms should be built to sate those needs, we could be able to raw-define cities for virtually any mod-created civilization we want, as well as create dwarven fortresses that would match fairly well with what players could create.

[Beardless]

The introduction of sprawl around the fort could change the rental situation drastically, but the overall price setting would apply there, assuming rooms in your fortress aren't so rare and the outside population so vast that they aren't all the exceptions you mentioned.  It's hard to say how it'll turn out.

I may be reading too much into this, but are you saying that dwarfs may emigrate to the sprawl if fortress housing prices are too high? More broadly, what are your plans for emigration during fortress mode in general? Currently, dwarfs that migrate to your fortress in search of whatever it is that makes them migrate stay forever, regardless of how horrific conditions may become. I look forward to having to work at keeping that population. Migrant labor is an intriguing concept as well. (I imagine this could easily appear as emergent behavior once the motivations for migration are programmed in.)

[Cruxador]

Currently, creatures pass on physical characteristics to their children. At least theoretically, this is respected in worldgen and over generations of breeding yields nations of dwarves who appear ethnically similar to one and other, but ethnically distinct from those of other nations. How will the entity sprawl interact with this genetic data-transfer?

I haven't decided quite how to do it yet, mainly because I was hoping to mingle things a bit at the overlaps.  At the most simple, it would just take the full range of the initial historical pairs and realize villagers from that range when they are generated (that would make citizens a bit more diverse than they are now, since there wouldn't be any selection, but it would still be a subset).  It could use the surviving hist figs instead if it wanted to have less variety.  Ideally, it might track a bit more information.  We'll see when I start generating villagers as units on the local map.

I would think that the optimal solution here would be to generate a set of frequencies for the traits for each nation, at the end of worldgen, and then draw directly on those to generate sprawl people.

[NW_Kohaku]

Currently, creatures pass on physical characteristics to their children. At least theoretically, this is respected in worldgen and over generations of breeding yields nations of dwarves who appear ethnically similar to one and other, but ethnically distinct from those of other nations. How will the entity sprawl interact with this genetic data-transfer?

I haven't decided quite how to do it yet, mainly because I was hoping to mingle things a bit at the overlaps.  At the most simple, it would just take the full range of the initial historical pairs and realize villagers from that range when they are generated (that would make citizens a bit more diverse than they are now, since there wouldn't be any selection, but it would still be a subset).  It could use the surviving hist figs instead if it wanted to have less variety.  Ideally, it might track a bit more information.  We'll see when I start generating villagers as units on the local map.

I would think that the optimal solution here would be to generate a set of frequencies for the traits for each nation, at the end of worldgen, and then draw directly on those to generate sprawl people.

Or the sprawl could be created out of Mountainhome emmigrants, meaning they have distant relatives back in the Mountainhome, although if they emmigrated at the first generation, they may have unique traits to that nation, so that one particular village is the only place you find "flax-colored" hair because it was where the only two flax-colored haired dwarves in the whole Mountainhome went at the start of worldgen, and they interbred only with that one particular isolated village's population without really ever immigrating back to the Mountainhome, and only slightly with some of the other villages.  A small hamlet set up by some of those decendents, however, might also be one of the strongholds of flaxen-hair-dom.

[Cruxador]

Currently, creatures pass on physical characteristics to their children. At least theoretically, this is respected in worldgen and over generations of breeding yields nations of dwarves who appear ethnically similar to one and other, but ethnically distinct from those of other nations. How will the entity sprawl interact with this genetic data-transfer?

I haven't decided quite how to do it yet, mainly because I was hoping to mingle things a bit at the overlaps.  At the most simple, it would just take the full range of the initial historical pairs and realize villagers from that range when they are generated (that would make citizens a bit more diverse than they are now, since there wouldn't be any selection, but it would still be a subset).  It could use the surviving hist figs instead if it wanted to have less variety.  Ideally, it might track a bit more information.  We'll see when I start generating villagers as units on the local map.

I would think that the optimal solution here would be to generate a set of frequencies for the traits for each nation, at the end of worldgen, and then draw directly on those to generate sprawl people.

Or the sprawl could be created out of Mountainhome emmigrants, meaning they have distant relatives back in the Mountainhome, although if they emmigrated at the first generation, they may have unique traits to that nation, so that one particular village is the only place you find "flax-colored" hair because it was where the only two flax-colored haired dwarves in the whole Mountainhome went at the start of worldgen, and they interbred only with that one particular isolated village's population without really ever immigrating back to the Mountainhome, and only slightly with some of the other villages.  A small hamlet set up by some of those decendents, however, might also be one of the strongholds of flaxen-hair-dom.

The reason this requires special consideration is because members of entity sprawl do not explicitly exist until encountered. Thus your suggestion is essentially impossible. It would, I suppose, be possible for historical figures to be removed from the general population with the explanation that they emigrated, and then their traits could be given to the members of a certain village. But this would be a lot of extra work and infrastructure, and the benefit would be small at best, while at worst it could end up with villages of near clones, and no similarity between members of different villages, even if they're members of the same national group and should be members of the same ethnic group.

So yeah, I don't believe that's a viable solution, nor would I consider it to be a good idea were it viable.

[Neonivek]

First I would like to thank Toady for answering my questions (or question, I havn't finished it) anyhow.

Edit addition: Ohh, It seems I do have unanswered questions... though seeing as I don't remember them very well I guess I don't mind.

An item made of superior materials than the body materials would generally be better in a non-magical setting, given equal training time.  In a magical setting, which would include martial arts that behave like magic, anything goes, as powerful as magic can get, and we'd want to try to respect that fact all around, so that if it is easy to learn, people wouldn't bother with weapons so much in general, if they had access to the knowledge.  In thinking about the generation of magic systems, we're trying to remain mindful of easy learning/industrialization of magic.  It's not a bad thing if you want it, but it is extra work to support that kind of setting

Ohh no that wasn't what I was trying to say at all... Ill try again.

Toady how are you going to balance out techniques with weapon wielders that use a person's physical body (and other less then full strength attacks)? In real life a body attack doesn't deal anywhere close to as much damage as a weapon strike but there are often opportunities where it is advantageous.
-My comment on other games is that when you use a bodily attack while weilding a weapon they tend to be oddly powerful. For example in Soul Calibur your kicks scale to your weapon

[NW_Kohaku]

Or the sprawl could be created out of Mountainhome emmigrants, meaning they have distant relatives back in the Mountainhome, although if they emmigrated at the first generation, they may have unique traits to that nation, so that one particular village is the only place you find "flax-colored" hair because it was where the only two flax-colored haired dwarves in the whole Mountainhome went at the start of worldgen, and they interbred only with that one particular isolated village's population without really ever immigrating back to the Mountainhome, and only slightly with some of the other villages.  A small hamlet set up by some of those decendents, however, might also be one of the strongholds of flaxen-hair-dom.

The reason this requires special consideration is because members of entity sprawl do not explicitly exist until encountered. Thus your suggestion is essentially impossible. It would, I suppose, be possible for historical figures to be removed from the general population with the explanation that they emigrated, and then their traits could be given to the members of a certain village. But this would be a lot of extra work and infrastructure, and the benefit would be small at best, while at worst it could end up with villages of near clones, and no similarity between members of different villages, even if they're members of the same national group and should be members of the same ethnic group.

So yeah, I don't believe that's a viable solution, nor would I consider it to be a good idea were it viable.

Actually, having isolated areas have very similar genetic traits with one another due to interbreeding of the same 2-3 families is a very realistic mechanic, and is simply the extreme end of the very "genetic similarity" effect this entire mechanic is supposed to create, so I have trouble seeing why you'd suddenly be opposed to it. 

(And the less isolated a group is, the less it will have problems of a lack of genetic diversity.)

As for implimentation, this could be achieved without creating a huge backlog of individual historical figures by having a much less detailed "Historic immigration/emmigration flow" record.  Rather than tracking individual dwarves (and their genes), you could have what amount to average rates of integrating populations, which create an abstract range of possible bloodlines for all the citizens of one particular area.  The actual data on the bloodline only becomes hard data when a member of that bloodline is encountered.  Until that point, villages are essentially just Shrodenger's Cat's Descendents, composed of proportionate amounts of possible bloodlines until you actually look at them. (At which point, the game rolls dice to see who, exactly, they belong to.)

If, for example, four families immigrated at Year 1 to set up an isolated mountain village, then there are only two families to draw genetic information from, and they will almost certainly interbreed repeatedly.  If a citizen of another group immigrates in halfway through worldgen, and intermarries with that isolated village, and he was from a group that had 12 families largely stuck together, then, as long as none of those 12 families have yet been forced to be defined, he can have theoretical properties of any of those other twelve families as well (only some of which would have to be defined if you met this person, as there is only one exemplar of this entire bloodline unless you meet that original village), then, depending on how much further down worldgen you go, you could have this one thread of a bloodline also be added into the original four families, or you could have this bloodline only partially mixed in with the bloodlines of the original four if this occured soon before the cutoff date on worldgen.

[Cruxador]

Or the sprawl could be created out of Mountainhome emmigrants, meaning they have distant relatives back in the Mountainhome, although if they emmigrated at the first generation, they may have unique traits to that nation, so that one particular village is the only place you find "flax-colored" hair because it was where the only two flax-colored haired dwarves in the whole Mountainhome went at the start of worldgen, and they interbred only with that one particular isolated village's population without really ever immigrating back to the Mountainhome, and only slightly with some of the other villages.  A small hamlet set up by some of those decendents, however, might also be one of the strongholds of flaxen-hair-dom.

The reason this requires special consideration is because members of entity sprawl do not explicitly exist until encountered. Thus your suggestion is essentially impossible. It would, I suppose, be possible for historical figures to be removed from the general population with the explanation that they emigrated, and then their traits could be given to the members of a certain village. But this would be a lot of extra work and infrastructure, and the benefit would be small at best, while at worst it could end up with villages of near clones, and no similarity between members of different villages, even if they're members of the same national group and should be members of the same ethnic group.

So yeah, I don't believe that's a viable solution, nor would I consider it to be a good idea were it viable.

Actually, having isolated areas have very similar genetic traits with one another due to interbreeding of the same 2-3 families is a very realistic mechanic, and is simply the extreme end of the very "genetic similarity" effect this entire mechanic is supposed to create, so I have trouble seeing why you'd suddenly be opposed to it.

An array of brad ethnicities is not the same as making the members of each village ethnically unique. Some geographic variation within national entities might not be a bad thing, but it should certainly not happen in this fashion. And it also shouldn't happen immediately.

As for implimentation, this could be achieved without creating a huge backlog of individual historical figures by having a much less detailed "Historic immigration/emmigration flow" record.  Rather than tracking individual dwarves (and their genes), you could have what amount to average rates of integrating populations, which create an abstract range of possible bloodlines for all the citizens of one particular area.  The actual data on the bloodline only becomes hard data when a member of that bloodline is encountered.  Until that point, villages are essentially just Shrodenger's Cat's Descendents, composed of proportionate amounts of possible bloodlines until you actually look at them. (At which point, the game rolls dice to see who, exactly, they belong to.)

If, for example, four families immigrated at Year 1 to set up an isolated mountain village, then there are only two families to draw genetic information from, and they will almost certainly interbreed repeatedly.  If a citizen of another group immigrates in halfway through worldgen, and intermarries with that isolated village, and he was from a group that had 12 families largely stuck together, then, as long as none of those 12 families have yet been forced to be defined, he can have theoretical properties of any of those other twelve families as well (only some of which would have to be defined if you met this person, as there is only one exemplar of this entire bloodline unless you meet that original village), then, depending on how much further down worldgen you go, you could have this one thread of a bloodline also be added into the original four families, or you could have this bloodline only partially mixed in with the bloodlines of the original four if this occured soon before the cutoff date on worldgen.

The main problem with this implementation, at least for the immediate future, is that it is now a big new system that needs to be worked out and have all its own problems and bugs worked out. The other problem that I immediately see is that the game will have to essentially piece together bloodlines on the fly, and with this it's going from both ends of the family tree. That will either result in all sorts of weirdness, or it will necessitate so much data storage that the purpose of entity sprawls in the first place is negated.

[NW_Kohaku]

First I would like to thank Toady for answering my questions (or question, I havn't finished it) anyhow.

Edit addition: Ohh, It seems I do have unanswered questions... though seeing as I don't remember them very well I guess I don't mind.

An item made of superior materials than the body materials would generally be better in a non-magical setting, given equal training time.  In a magical setting, which would include martial arts that behave like magic, anything goes, as powerful as magic can get, and we'd want to try to respect that fact all around, so that if it is easy to learn, people wouldn't bother with weapons so much in general, if they had access to the knowledge.  In thinking about the generation of magic systems, we're trying to remain mindful of easy learning/industrialization of magic.  It's not a bad thing if you want it, but it is extra work to support that kind of setting

Ohh no that wasn't what I was trying to say at all... Ill try again.

Toady how are you going to balance out techniques with weapon wielders that use a person's physical body (and other less then full strength attacks)? In real life a body attack doesn't deal anywhere close to as much damage as a weapon strike but there are often opportunities where it is advantageous.
-My comment on other games is that when you use a bodily attack while weilding a weapon they tend to be oddly powerful. For example in Soul Calibur your kicks scale to your weapon

If you're talking about what I think you're talking about, you might want to describe your various kung-fu fighting styles, or at least link to the Dual Wielding thread.  Currently, all attacks with a weapon are assumed to actually be with the weapon.

(Of course, I see no reason why carrying a steel sword would actually make you kick harder, since DF actually keeps track of individual armor parts and body parts.  If anything, a steel BOOT would make you kick harder.)

[Neonivek]

If you're talking about what I think you're talking about, you might want to describe your various kung-fu fighting styles, or at least link to the Dual Wielding thread.  Currently, all attacks with a weapon are assumed to actually be with the weapon

I am not. I think you entirely missed what I was trying to say/ask.

This isn't even Kung-Fu, you could see it being done with almost any weapon under the sun and was done so in every location on earth.

Of course, I see no reason why carrying a steel sword would actually make you kick harder

It doesn't. It is done in videogames not for realism but for gameplay. Well... Ok. There are SOME ways to get stronger kicks using weapons in your hand but those arn't what I am refering to.

[Askot Bokbondeler]

i guess he's talking about something like feinting an attack with your sword, and while the enemy is busy parrying it you kick it's knee, unbalancing him and getting an opportunity for a coup de grace

[Neonivek]

i guess he's talking about something like feinting an attack with your sword, and while the enemy is busy parrying it you kick it's knee, unbalancing him and getting an opportunity for a coup de grace

Yeah that whole collection. There is nothing in the game that would make the kick in anyway advisable as it is currently. So I am wondering if there ever will be.