What would be the safest, not-cheesy way to level grind in a sand-box like setting?

Quote:

---

Originally Posted by Samael Morgenst

Deities appear to be mostly indifferent in this world, but they curse with either vampirism or lychantropy those who topple statues in their temples.

Also, deities of death sometimes gift their most fervent followers with slabs containing the secrets of life and death (necromancy , in short).

---

"Eat LA+8 Template, Evildoer! Just try and level now! Bwahahaha!" I guess I could see it. Is that a Dwarf Fortress thing?

Still, I think I like my gameplay of the race's actions determining their deity's spheres (meaning that there's more considerations than, "we winrar good?", more consequences for those actions) better than the idea of the Halfling deity being disconnected from her people because they went a different route than she's hard-coded spec'd for.

This brings up something I haven't explicitly stated yet: how Breed Lycanthropes can be a Created race. Can D&D Deities actually cause people to become Breed Lycanthropes? If so, then, yeah, I guess that makes more sense than having them as a starting race, since I had planned to not consider any others like Vampires. So much for my Wererats.

Quote:

---

Originally Posted by Vahnavoi

@Quertus: speculating about what a character in the setting would know isn't all that useful

---

Agreed, which is why I propose engineering a society where that's more of a known than an unknown.

Quote:

---

Originally Posted by Vahnavoi

all a player gets of any of that when starting theur own play is whatever Knowledge skills the character has, what's printed in the game manual and what's visible on the game map when choosing starting location. Things like who your neighbours are or where and how far safe areas extend have to typically be explored on one's own.

---

Or Society can just hand them a map. Cooperation and all the having a cluefulness of an evil-overlord-mandated 5-year-old advisor does a Society good.

Quote:

---

Originally Posted by Vahnavoi

Yes, this can include asking from higher-level NPCs, if those are available, but they too have to be found. Though with some luck, Gather Information and Knowledge (Local) will point a starting character to the right direction.

---

"Say, Master, ..." Funny how valuable being a Trained class is for finding those high-level NPCs, and doing so before the game even starts no less.

And this reminds me of another point: There's population statistics for distribution of NPC levels. If there any reason not to concentrate your population into planar metropolis sized chunks? If you're running Gamist logic for NPCs. If you're running Simulationist logic for NPCs, well, pretty much all races can get to ECL 9 pretty trivially. I'm guessing the latter isn't desirable, so we're going with the more Gamist logic of % of population.

Not that you need the high end of those charts to Mentor all the NPCs, or to chaperone them through the "safe" wilds to level up; that can easily be handled by some mid-tier "Graduates".

Quote:

---

Originally Posted by Vahnavoi

Anyways, back to safe leveling: if non-lethal combat between friendlies is allowed to grant XP, then simply holding non-lethal unarmed combat and wrestling matches beats most other options in simplicity, speed and availability of resources. The chief downside is the opportunity cost of some fraction of a community's fighting-capable people spending hours unconscious. Notable, the best time to attack a community is when their warriors are spent from training.

---

I imagine (given how quickly one could gain levels from this method, if it grants XP) that very few people need to be involved in "leveling", and that other duties, like gathering food, crafting goods, and reproduction services, will leave Society more vulnerable on average.

Quote:

---

Originally Posted by Vahnavoi

As for T=0 survival, I did a lot of number crunching on this once. Humans, with their extra skill point and feat, are one of the best early specializers. A human community can significantly reduce number of people required for Survival and, equally importantly, significantly increase productivity of goods and services via Profession and Craft skills. Races with Wisdom penalties face extra difficulty - notably, this means some shorter-lived races (such as orcs) that have an edge in how fast they can get first classed characters, can't sustain as big populations from the start.

---

Yeah, Wisdom penalties are bad in this scenario. But Humans aren't optimal the way you describe - any race can have members good at Survival - some far better than Humans. What they can't do is have as much redundancy of skills, be as good at recovering from suffering losses. Elves lose their Suppliers, and they're in trouble (well, except Grey Elves, who also get +1 skill point (from Intelligence)); Humans lose theirs, and someone else took it as a fallback skill, no biggie. Also, many other races (especially smaller races, and those with wisdom bonuses) greatly outperform humans, and are thus able to dedicate much more of their manpower to other tasks.

Also, we have some very different T=0 starting assumptions. My assumption is, human babies left in the wild die. Thus, my assumption is, all T=0 starting beings are "Adults". Which is nice if you're a Dragon. :smallwink:

If we're all starting as babies, IMO all the humaniods die off, and Hatchling Dragons rule the roost. Doesn't seem very interesting to me.

I suppose... *IF* you can unlock classes within the first year or two, then the shorter-lived races might have some advantage in that their young will come "of age" sooner, to be the first race with "good builds". Still, the Elf Commoner 1 or Elf Humaniod 1 will happily be an Elf Humaniod 1 / Wizard 19, so I don't think your young Orc Barbarian 20 is all that advantaged.

So, can you explain the exact scenario under which you see "short lifespan" as a big advantage for "getting classes"?

Quote:

---

Originally Posted by Vahnavoi

How to model characters before they reach 1st level in their chosen class is iffy for many humanoids, as 0th level is not a defined concept in 3rd edition. The choices I would pick between are 1 HD humanoid and 1st level commoner, which is then retrained to the correct class when a character meets starting age qualifications.

---

Oh, right, retraining. So it's Elf Wizard 20 vs young Orc Barbarian 20. So really not seeing how short-lived races have a class advantage. (A numerical advantage? Definitely. Enough so that I'm concerned Elves aren't actually tenable long-term. But I'm not seeing any reason the shorter-lived races would have a class advantage.)

Quote:

---

Originally Posted by Maat Mons

If classes aren’t a thing at simulation start, then I think Elves are the zerg-rushers. They start with longbow and shortbow proficiency before class levels or feats. Technically, they also get longsword and rapier proficiency, but good luck putting that to use before unlocking metalworking. Yeah, they take a Constitution penalty, but if enemies have to charge through arrow fire for a few hundred feet just to get close enough to attack and then are still stuck attacking at a -4 penalty because they haven’t gotten themselves any weapon proficiencies yet, it’ll be hard to capitalize on your lower HP. I guess a race with Darkvision could attack you on an overcast, moonless night? Eh, I guess you’ll just have to attack them before they get the chance.

There are also Elves without a Constitution penalty, but I think Jungle Elves might be the best rusher, even though they keep the Con penalty. You’d be stuck with a Shortbow, so -1 average damage compared to a longbow, but you’d have a handaxe. Ideally, your forces would kill any enemies before they got close, but since things don’t always go to plan, proficiency with a melee weapon is good to have as a backup. Jungle Halflings get all the same proficiencies, but I think we can all agree that halflings are lame.

---

And here I was, concerned that Elves were unplayable. Thanks for restoring my faith in their viability (even if they don't play quite the way I was envisioning).

Quote:

---

Originally Posted by Maat Mons

Under that setup, Aquatic Elves seem primed to dominate the water. They’d outnumber Merfolk by 4 to 1, Locatha by 9 to 1, and Sahuagin by 16 to 1. Actually, wait, there are still Aventi, Shoal Halflings, Shalarin, River Spirit Folk, Sea Spirit Folk, a bunch of Unearthed Arcana races, and anyone with the Amphibious template. Hmm… well unlike most of them, Aquatic Elves can venture onto land to collect wood to make spears. So, I guess it’s Aventi, Shoal Halflings, Spirit Folk, and templated creatures that you still need to worry about. Still less of a battle royale than the land.

---

OK, I'll admit, I was ignoring "water" as being equivalent to "lava" in this scenario. :smallamused:

I'd ask about weresharks, but Lycanthropes may be disqualified as starting races.

Quote:

---

"Eat LA+8 Template, Evildoer! Just try and level now! Bwahahaha!" I guess I could see it. Is that a Dwarf Fortress thing?

---

Yeessssssss....

@Quertus: I already explained why shorter lived races have an advantage. Do check the starting ages for classes again: shorter lived races typically mature faster, meaning they will achieve 1st level sooner. Which, again, means that if fast leveling tricks are in play at all, you will have a bunch of high level orc barbarians running around when the first elven wizards get out of school.

Thinking about it a bit more, Elves wouldn’t necessarily have to rush, even though they’d be good at it. Even in “present day” D&D, 90% of people are Commoners, regardless of which type of humanoid you’re talking about. I can’t imagine neolithic D&D would have a more robust public education system. Elven Commoners are capable enough to fight in defense of their homes, due to racial weapon proficiencies. If an attacking force is using only trained combatants, that’s at most 10% of their adult population being pitted against 100% of the adult population of the Elves. This would, I think, make up for the Elves’ lower population.

Of course, the attacking race could also send their Commoners into battle. But I think an Elven Commoner is a more able combatant than a Commoner of another race, which can still partially mitigate a numerical disadvantage, albeit not at the 10-to-1 rate Elves would enjoy if the attacking race left their Commoners at home. Comparing Commoner versus Commoner, it really depends on how heavily each race’s Commoners spec into combat. I think the default assumption is that Commoners don’t spec into combat at all, which puts the Elves in the best relative position.

If the attacking race fully specializes their Commoners for battle, they’re either spending a feat for a weapon proficiency or each Commoner has to pick between a sling, a longspear, or a regular spear. Crossbows would be off the table until the Middle Ages, at which point we’re well past any starting issues. That means each unit is either ranged or melee, not both. The elven units, on the other hand, can all switch rolls as needed. They can have 100% of their forces shooting while the enemy is distant, and 100% of their units stabbing when the enemy is close. Other Commoner troops would have to be split between the two rolls at some ratio or another. Plus, the Elves are still much better at ranged combat, because slings are shorter-range weapons.

If the attacking force gives their Commoners Martial Weapon Proficiency (longbow), things are more even. They can each have a good ranged weapon and a good melee weapon. They still need to pick between a reach weapon and a non-reach weapon for melee, so each unit is specialized to be either front row or second row. Jungle Elves, in particular can have a shortbow and a handaxe from their racial proficiencies, plus a longspear as their weapon of choice from Commoner. This gives them the full trifecta. Alternately, the attacking race could give their Commoner Simple Weapon Proficiency to also have the trifecta, but then they’re back to using slings, which gives Elves the advantage at long range. Plus, the Elven Commoners still have a feat to play with, while the attacking Commoners don’t unless they’re human. I’m not sure which feat gives the best advantage here. My first instinct was Wild Cohort, but they couldn’t reliably hit the Handle Animal DC to direct their pets.

Quote:

---

Originally Posted by Vahnavoi

I already explained why shorter lived races have an advantage. Do check the starting ages for classes again: shorter lived races typically mature faster, meaning they will achieve 1st level sooner. Which, again, means that if fast leveling tricks are in play at all, you will have a bunch of high level orc barbarians running around when the first elven wizards get out of school.

---

Not untrue, however…

According to Races of the Wild (page 13) a 15-year-old Elf is equally as physically mature as a 15-year-old Human, and somewhat more mentally mature. From there, the Human fully completes their physical growth by age 20, and the Elf by age 25. Since a 15-year old human gets full adult Human stats, I don’t see why the same wouldn’t be true of Elves. The only thing I can see holding Elves back is that their fertility rate is 1/5th that of Humans. Admittedly, this is going to be a big issue starting around 15 years after simulation start and continuing for however long it takes before everyone’s populations have reached the maximum their respective territories can sustain.

Quote:

---

Originally Posted by Vahnavoi

@Quertus: I already explained why shorter lived races have an advantage. Do check the starting ages for classes again: shorter lived races typically mature faster, meaning they will achieve 1st level sooner. Which, again, means that if fast leveling tricks are in play at all, you will have a bunch of high level orc barbarians running around when the first elven wizards get out of school.

---

Now, I’m hand-waving a few things for brevity, but… 4 months after game start (EDIT: that’s “4 months after initial generation”, not thousands of years later after the player starts playing), you could have a batch of level 9 characters under one interpretation. That’s shorter than most pregnancies, even if you set them to “reproduction services” immediately, and they’re successful. So you’ll have mid if not high level characters running around before you’ve got babies running / crawling / lying around.

So, again, I’m not seeing how lower starting age of shorter lived races really matters here.

Now, sure, that’s Generations 1. Generation 2? Sure they’ll get a head start on the shorter-lived races. But that seems trivial next to the number of epic level wrestlers your technique might have put out on both sides by then.

The big advantage they have is just in sheer reproductive rate - how quickly they transition from “baby” to “breeding”, and how successful those attempts are… and how big the “litter” is, I suppose.

Sure, age to maturity matters - it determines how long they’re a resource sponge before transitioning into a productive member of society, getting levels, what have you. But that new 1st level character just isn’t significant compared to their epic level elders.

Now, to give some Arbitrary Internet Numbers, you could argue that, when multiple generations go to war, level [50, 30, 10] beats level [50, 10]. I’d file that under “reproductive rate” rather than “mature faster”, but we could be saying the same thing.

However, if we call that T=X? Then at T=2X, after the “short lived” part of “short lived races” comes into play? Then we’re looking at the same level [50, 30, 10] vs level [90, 50, 10]. And that’s what we’re looking at from now to eternity. And that’s assuming the longer-lived races only live twice as long.

There’s a short window of level advantage from the new crop for secondary fighters; otherwise, Level advantage goes to the longer-lived races. It’s only an advantage of numbers that the short-lived races can hope to rely on after that. Quantity vs Quality.

However, that’s all Simulationist logic. If, OTOH, we go with Gamist logic of 3e population tables, more reproductive races gain population faster, therefore build cities and planar metropolises faster, therefore level faster. Quantity = Quality.

So it depends on which kind of silly the underlying system is, as to what level NPCs will be.

But this thread was, I believe, about leveling a single PC*. And, here, “hunting / trapping tasty animals well inside the safely patrolled areas with high-level NPCs (and a map)” and “coed naked wrestling” seem valid answers, if similarly dependent upon the underlying system (of how random encounter table generation works, and getting XP from friends, respectively).

* where the question about how races would fare is simply a question of, “what races can I predict won’t have gone extinct before the game actually starts?”, or “what would the society of a surviving race reasonably look like in this scenario?”, to replace those “unknown unknowns” (or whatever) you were talking about with “known knowns” (or whatever).

@Quertus: I'm talking of the first generation. The age adjustment for starting as class is added to threshold for reaching adulthood. If at T=0, everyone is an untrained young adult, it makes a difference whether you add +1d4 or +4d6 to that number.

Quote:

---

Originally Posted by Vahnavoi

@Quertus: I'm talking of the first generation. The age adjustment for starting as class is added to threshold for reaching adulthood. If at T=0, everyone is an untrained young adult, it makes a difference whether you add +1d4 or +4d6 to that number.

---

OK, let me try this: You are saying that *if* everyone starts as a "young adult" (mostly undefined except for dragons, but mostly defined by the next bit), *and* everyone starts off without class levels (because they're at the "before training" age), *and* class levels don't need to be unlocked, *due to the fact that* shorter lived races add smaller random numbers to their "young adult" age to get their "trained age", *then* in this situation, the shorter lived races will gain their 1st level first *and therefore* gain their higher levels first *and therefore* have an advantage. Is that right?

If so, I start with the response of, "That was never a scenario I had considered". That is, I considered "levels not unlocked, humanoid HD" and "levels unlocked, 1st class level" as the potential T=0 start conditions (well, and "age = 0, humanoids just die", due to my senile half-remembered memories of the last time this concept came up).

I guess I'd say, off hand gut reaction, it feels inelegant to force the "humanoid HD" to determine skills / skill points for just the first few years of the simulation, rather than just go with class levels straight off the bat. But other than the fact that I'd never want to write the code that way (and certainly *wouldn't* write it that way if "NPC Levels are determined by Population", so add "*and* NPC Levels are determined by Simulationist 'earning XP' rather than Gamist 'population tables'" to the opening statement of what I think you're saying), I don't see anything inherently *wrong* with imagining that as a possible simulation vector, and if everything worked that way, maybe the shorter lived races would have the early level advantage.

Personally, I think the level advantage will be much more significant for the long-term races living longer and getting to accumulate more levels, or for the short-lived races having more population and therefore winning the level advantage in "level determined by population tables" scenario. Shrug.

Also, for the "small and has a Wisdom bonus" races like Anthroporphic Bats being able to raise significantly larger families or otherwise support more production than other races per Resource Gatherer (and, if a mechanical limit is implemented, able to support more of their race off the same resource pool).

Those are the advantages I'm looking at.

Jermlaine are Tiny, which means they require only 1/4th as much food as a medium creature. They also have a +6 racial bonus to Wisdom. So, a 1st-level Jermlaine Commoner with Skill Focus (Survival) could feed themself and 23 others with their daily foraging.

According to the internet, reptiles need to consume only 1/8th as much food as mammals of the same size. I haven’t been able to find anything in the D&D rules to model this. Does anyone know if it’s been addressed anywhere?

If anyone is curious, the internet says birds consume 40% more food than mammals of the same size. Since Anthropomorphic Bats have been brought up, I’ll note that I suspect bats might be closer to birds than to other mammals in terms of metabolism, but I haven’t checked to verify this.

@Quertus: I noticed you didn't consider it, that's why I reminded you, since it is a corollary to the argument from starting ages. As far as elegance goes, "everyone starts with racial HD only" and "everyone starts as a commoner" before finishing class training are both more elegant for populations at T=0 than "everyone starts as 1st level in some class". Reason being, with everyone starting from scratch, it is possible to use the starting age by class rules to determine the rate at which various classes appear in the world. "Everyone starts as 1st level in some class" requires additional assumptions to cover why those classes appear in quantities that they do, similar to DMG's demographic tables.

Related, the split you draw between "simulationist" and "gamist" is nonsense. In general, it is not a good idea to use Forge terms in context the Forge largely failed to cover. "Everyone levels by XP" is not more or less simulation-like, nor more or less game-like, than "population spreads are determined by tables". They aren't even mutually exclusive. We're instead talking of different ways to set up a simulation, specifically contrasting ways to set up the initial conditions for it. All methods, on a fundamental level, involve arbitrary authorial decisions regarding the initial populations, and "everyone levels by XP" is an additional rule on top that may or may not be in place in any version.

In any case, being biologically long-lived or even immortal poses no real benefit for leveling or numbers of higher-level characters. The first reason for that is that every race that advances by class is capable of reaching high levels in their lifespan - the second reason is that the only real limit is the existence of sufficiently high CR encounters. Other classed characters are an obvious source of higher CR encounters. The third reason is that there will still be all normal reasons for hostilities: not everyone wants to watch you get to arbitrary levels, and a long lifespan on its own doesn't make a character more resistant to unnatural death. It's more likely to work the other way around: achieving noticeable level advantage allows reaching full natural lifespan.

When high level characters can cast Genesis to make new land for living space and create items to supply endless food, the amount of time between being born and casting Genesis becomes relevant for determining their population's exponential growth rate. If the time is significantly shorter for shorter-lived races, then their exponential takeoff will inevitably pull ahead.

Quote:

---

Originally Posted by Tohron

When high level characters can cast Genesis to make new land for living space and create items to supply endless food, the amount of time between being born and casting Genesis becomes relevant for determining their population's exponential growth rate. If the time is significantly shorter for shorter-lived races, then their exponential takeoff will inevitably pull ahead.

---

Um... no? Say there's just 2 of one race, "Adam" and "Eve". Adam's build can Survival to feed 2 people, Eve 1. Even at level 999,999,999,999. They can support at most 1 child, no matter how many Genesis planes they create. (EDIT: And that's only if they spend all their time making Survival checks to support the little Resource Sponge!)

Point is, there's a few more variables involved.

Quote:

---

Originally Posted by Vahnavoi

@Quertus: I noticed you didn't consider it, that's why I reminded you, since it is a corollary to the argument from starting ages.

---

I'll grant it's a 3rd option beyond "Starting Age" and "Dead Babies".

Quote:

---

Originally Posted by Vahnavoi

As far as elegance goes, "everyone starts with racial HD only" and "everyone starts as a commoner" before finishing class training are both more elegant for populations at T=0 than "everyone starts as 1st level in some class". Reason being, with everyone starting from scratch, it is possible to use the starting age by class rules to determine the rate at which various classes appear in the world. "Everyone starts as 1st level in some class" requires additional assumptions to cover why those classes appear in quantities that they do, similar to DMG's demographic tables.

---

How to explain "elegance"?

Humaniods

Have two states: "Resource Sponge" and "Productive Member".
Resource Sponges follow these rules: ... and transition to Productive Members with 1 Class Level based on rules of Race, Age, AvailableClasses, Population, and Strategy.
Productive Members follow these rules: ...
At game Start, create X New Productive Member Humaniods with 1 class level as though they just transitioned from Productive Members
On TimePasses, check for Resource Sponge Humanoids with CurrentAge >= ProjectedStartingClassAge; transition them to Productive Members.
On TimePasses, check for Productive Member with CurrentAge >= MaxAge...

vs

Have three states: "Resource Sponge", "Racial HD", and "Productive Member".
Resource Sponges follow these rules: ... and transition to Productive Members with 1 Class Level based on rules of Race, Age, AvailableClasses, Population, and Strategy.
Racial HD follow these rules: ... and transition to Productive Members with 1 Class Level based on rules of Race, Age, AvailableClasses, Population, and Strategy.
Productive Members follow these rules: ...
At game Start, create X New Racial HD Humaniods with these rules for starting skills: ...
On TimePasses, check for Resource Sponge Humanoids with CurrentAge >= ProjectedStartingClassAge; transition them to Productive Members.
On TimePasses, check for Racial HD Humanoids with CurrentAge >= ProjectedStartingClassAge; transition them to Productive Members.
On TimePasses, check for Productive Member with CurrentAge >= MaxAge...
On TimePasses, check for Racial HD Humaniods with CurrentAge >= MaxAge...

2 data objects vs 3, 2 sets of rules vs 4, 2 TimePasses check vs 4 3, and a lot of redundancy in the code. OK, fine, there is 1 line of arguably inelegant "cheat code" necessary to make the 1st one work, but it's paralleled with a similar line in the other set.

I would never write the latter when the former was a possibility.

With complex multiple inheritance, you could remove the code redundancy, but then the object definitions would be rather inelegant.

Still, if you wanted no classes whatsoever until classes were unlocked, you could have those 3 states, but the Resource Sponge objects would always follow the same check of looking for available classes to determine whether to flow into Racial HD or a Productive Member, no extra inelegance introduced by starting the T=0 Humaniods off with Racial HD (except for the crossed-out line I accidentally added above).

So, all in all, just starting with class levels is the most elegant solution.

Quote:

---

Originally Posted by Vahnavoi

Related, the split you draw between "simulationist" and "gamist" is nonsense. In general, it is not a good idea to use Forge terms in context the Forge largely failed to cover. "Everyone levels by XP" is not more or less simulation-like, nor more or less game-like, than "population spreads are determined by tables". They aren't even mutually exclusive. We're instead talking of different ways to set up a simulation, specifically contrasting ways to set up the initial conditions for it. All methods, on a fundamental level, involve arbitrary authorial decisions regarding the initial populations, and "everyone levels by XP" is an additional rule on top that may or may not be in place in any version.

---

I mean, if you want your (AFB, Internet numbers) 10,000th Humaniod to spring from the womb as a 20th level character, because that's what the Population Table says you get at Humaniod #10,000, and you want to call that Simulating how things work rather than an oversimplification (ie, a Gamist abstraction), you do you. However the Forge may have done these terms dirty, I find them to have meaning, and I call the population tables Gamist, and tracking individual growth Simulationist (even if XP is, itself, arguably a Gamist abstraction).

Quote:

---

Originally Posted by Vahnavoi

In any case, being biologically long-lived or even immortal poses no real benefit for leveling or numbers of higher-level characters. The first reason for that is that every race that advances by class is capable of reaching high levels in their lifespan - the second reason is that the only real limit is the existence of sufficiently high CR encounters. Other classed characters are an obvious source of higher CR encounters. The third reason is that there will still be all normal reasons for hostilities: not everyone wants to watch you get to arbitrary levels, and a long lifespan on its own doesn't make a character more resistant to unnatural death. It's more likely to work the other way around: achieving noticeable level advantage allows reaching full natural lifespan.

---

Eh, no. Let's step through these.

Quote:

---

Originally Posted by Vahnavoi

The first reason for that is that every race that advances by class is capable of reaching high levels in their lifespan

---

Maybe.

Let's say the only method of leveling, period, is through random encounters, and the random encounter tables end at CR 1. Then NOBODY will be able to get beyond 9th level, regardless of their lifespan.

Let's say the only method of leveling, period, is through your wrestling scheme (I like calling your XP scheme "coed naked wrestling" and mine of killing animals in traps "baby seal clubbing", btw), or my "Hone your Craft" theory. In such a sysstem of earning XP, infinite levels are theoretically possible, as your (to use your idea) wrestling partner(s) hopefully win about half the time, and hopefully level up with you.

In practice... lots of annoying fiddly details, but imagine Adam and Eve, one (starting or) pulling ahead, therefore winning more often... but the other earns more XP for their wins... OK, it's probably fine, especially with a larger population.

Because of the way the XP tables are set up, rate of level gain while wrestling equal-level opponents is, all things being equal, static. But static at what rate? If it takes 1 year per level, vs if it takes 20 years per level? That's a huge difference wrt what level races can reach based on their lifespan.

And, at 1 year per level, a 1,000 year old elf is looking pretty scary if you ask me.

Let's say the only method of leveling, period, is through killing other "active participant" races. While this is arguably the most interesting "1 way" scenario, it's also the one least in keeping with the thread premise. But in this scenario, enslaving other races and farming them for XP is the dominant strategy, IMO.

Let's say the only method of leveling, period, is through the DMG demographic tables. Then "age" is irrelevant; it's reproductive stats (including how quick a reproduction rate the Productive Members can support with their Survival skill) is the most important set of stats in the game.

That last one is the only "only" that's likely to see play (and it really doesn't favor the Elves); otherwise, there's likely a combination of factors that can add to XP. But how those factors are calibrated (and little things, like whether Society is smart enough to use Maps) will determine how quickly levels can be gained.

Quote:

---

Originally Posted by Vahnavoi

The third reason is that there will still be all normal reasons for hostilities: not everyone wants to watch you get to arbitrary levels,

---

Anyone "watching" will likely be "allied" or "hostile" already (although I suppose the system / AI / whatever could allow more nuance than that). With a map size and exploration rate anything like this world, I imagine most races will be unaware of most races for thousands of years, minimum.

Quote:

---

Originally Posted by Vahnavoi

and a long lifespan on its own doesn't make a character more resistant to unnatural death. It's more likely to work the other way around: achieving noticeable level advantage allows reaching full natural lifespan.

---

I'd argue that Society, and creating Safe Zones (my idea of hunting all dangerous creatures until the Random Encounter Table makes sense - just "Humans" in a Human town, just Elves, Allies, and Animals in an Elven Woods, etc) do a lot to make a character more resistant to unnatural death. Also, if we're following RAW, outlawing "Businesses", as they invoke rolls on the "Natural Disasters" table. :smallamused:

But, yes, obtaining levels, which happens somewhere between "quickly" and "instantly" (depending upon the underlying system) upon reaching the appropriate age, definitely provides a nice buffer against many things that cause unnatural death.

Quote:

---

Originally Posted by Quertus

Um... no? Say there's just 2 of one race, "Adam" and "Eve". Adam's build can Survival to feed 2 people, Eve 1. Even at level 999,999,999,999.

---

Why wouldn't the level 1 trillion minus 1 character have a higher Survival modifier? They have lots of skill ranks to throw around, and a very high skill cap. Every +2 to Survival is another person feed with the same time expenditure.

Quote:

---

Originally Posted by Maat Mons

Why wouldn't the level 1 trillion minus 1 character have a higher Survival modifier? They have lots of skill ranks to throw around, and a very high skill cap. Every +2 to Survival is another person feed with the same time expenditure.

---

Because that wasn't the build? Maybe the build only gets 2 skill points per level, and has to keep Bluff and Sense Motive maxed, in order to be competitive in the Mirror Match.

Maybe we've set up the game such that not just Classes, but even Skills need to be unlocked, and they just never unlocked Survival.

The point was, there's other considerations than just resources, any one of which can turn out to be the Limiting Factor.

My question was more, whether by RAW, resources were ever a Limiting Factor preventing our Humanoids and other whatchamacallit (Participating Races?) from clustering in Planar Metropolis-sized chunks, either for purposes of goods available for purchase, or in the scenario in which class and level are determined by (local) population.

Quote:

---

Originally Posted by Quertus

If we’re starting from 0, with no infrastructure, the Simulation may need to run for thousands of years before domesticated animals become a thing. Do note that “domesticating animals” is the leading cause of disease, which may make this a dubious strategy for some races.

It also makes it interesting that, just as you get dogs from wolves, you might generate things like “domesticated hydras”, “domesticated Dragons”, or even “domesticated elves”.

---

Quote:

---

Originally Posted by Vahnavoi

Odd domesticated animals are a possibility, ingenuity of the algorithm maker allowing, but might be reasonable expectation of D&D. There's a lot of fantastic fauna (giant lizard, dinosaurs, magical beasts etc.) that are part of the core rules and have seen use in handcrafted settings, so yeah. You may get jungle elf barbarians herding triceratopses or something.

---

Oh, I was taking the ability to use Handle Animal to do things like get Mounts or War Mounts per the rules as a given.

What I actually meant by this comment was more, "Wolves get Domesticated into Dogs. What happens if, in Fantasy Land, other species get Domesticated? What new species would Domesticated Hydras, Domesticated Dinosaurs, or Domesticated Dragons produce? Which I've realized is a lot of coding (and data entry, and art, and...) for very little gain. Cool concept for potential world building purposes, not much use here IMO.

OTOH, raising a herd of 3-horns, for their meat, for their hide, and to animate their skeletons in defense of Society? That actually sounds pretty cool.

@Quertus: your argument for elegance involves another abstracted category of characters that is not derived from anything in the d20 rules. Your version for two data objects only looks simpler because the assumptions behind who belongs in that category are not enumerated.

Similar relationship exists between simulation of individual objects versus picking groups from a table. The latter solution is actually computationally simpler, as it involves simulating less objects over time. But it is not more elegant because of all the additional assumptions that go into making those tables, even if those assumptions won't be visible in the finished code.

This also the flaw in your "gamist" versus "simulationist" argument. Your decrying of tables as more "gamist" is based on assuming silly reasoning behind those tables. That's baseless, the reasoning can be anything. Similar argument applies to individual simulation: that only produces more "simulationist" results than tables if the assumptions about those individuals aim for realism. Experience points are an abstract game scoring mechanic, they are fundamentally non-realist. The only thing you're simulating by applying d20 experience point rules is the d20 game system, but this isn't more accurate to source material than applying rules of population demographic tables. It's just an arbitrary decision of which parts of the source material to emphasize.

Back to elegance: consider the following progression tracks for characters:

Dependent (child) ---> Dependent (Adult) ---> Classed individual

Dependent (child) --- > Adult-in-training (either racial HD or 1st level commoner) --- > Classed individual

In both formats, the Dependent (child) category has to be specially defined for the simulation, as the base d20 system isn't concerned with child characters. But there is no reason to specially define Dependent (Adult), since we already have options in d20 rules already describe adult characters without character class. Doing things the former way may be computationally simpler, but it also loses details and requires additional assumptions about capabilities of Dependent (Adult) category as well as additional rules somewhere to explain where commoners and characters with just racial HD come from. Remember, at some point, the simulation has to lead to a playable game world, so all dependent characters have to be transformed into viable entities described by d20 ruleset at some point. The alternative is that they just disappear when actual play starts, never being anything more than background numbers.

Regarding old age:

Quote:

---

Originally Posted by Quertus

Let's say the only method of leveling, period, is through random encounters, and the random encounter tables end at CR 1. Then NOBODY will be able to get beyond 9th level, regardless of their lifespan.

---

Yes, that's a special solution that makes longevity even less relevant, but the assumptions made by core 3.5 D&D aren't anything like this, so we can't use it to draw useful conclusions about the simulation.

Quote:

---

Originally Posted by Quertus

Let's say the only method of leveling, period, is through your wrestling scheme (I like calling your XP scheme "coed naked wrestling" and mine of killing animals in traps "baby seal clubbing", btw), or my "Hone your Craft" theory. In such a sysstem of earning XP, infinite levels are theoretically possible, as your (to use your idea) wrestling partner(s) hopefully win about half the time, and hopefully level up with you.

---

It's rather relevant to the argument you're commenting on that these are NOT the only ways: at any point, anyone hostile to you can also gain XP by terminating you. This means that continuing to level up in perpetuity requires achieving an insurmountable advantage over every hostile entity.

Quote:

---

Originally Posted by Quertus

Because of the way the XP tables are set up, rate of level gain while wrestling equal-level opponents is, all things being equal, static. But static at what rate? If it takes 1 year per level, vs if it takes 20 years per level? That's a huge difference wrt what level races can reach based on their lifespan.

---

It's rather relevant to the argument you're commenting on that they way XP tables are set up, the rate at which XP is gained is so quick that longer lived races gain no real advantage. Which leads us to...

Quote:

---

Originally Posted by Quertus

And, at 1 year per level, a 1,000 year old elf is looking pretty scary if you ask me.

---

That elf had no real advantage reaching 9th level when they were a young adult, compared to a young adult human or a young adult orc. Them making it to 1,000 years old requires they reached such a great advantage earlier in life that no-one of any other race who started leveling at the same time could stop them.

Could the simulation maker adjust leveling rates so that extreme lifespans are required to reach high level? Yes. But core 3.5 doesn't really do this. Indeed, given its starting age rules, it does the opposite. An elf at 110 has not learned more than a human at 16.

Quote:

---

Originally Posted by Quertus

Let's say the only method of leveling, period, is through the DMG demographic tables. Then "age" is irrelevant; it's reproductive stats (including how quick a reproduction rate the Productive Members can support with their Survival skill) is the most important set of stats in the game.

---

Untrue. You forgot to account for death rate: aging rules set the limit for natural lifespans too. This obviously impacts population growth curves, with larger effect the longer the simulation runs: when the first generation elves start to drop from old age, the several human generations will already be dust. Though, again, for this to become visible, the conditions have to be such that characters can actually live long enough to die naturally.

The most significant practical limit is that an actual simulation on an actual computer isn't going to run forever. The player has to stop the simulation and begin play after some practical real number of cycles.

Related:

Quote:

---

Originally Posted by Quertus

Anyone "watching" will likely be "allied" or "hostile" already (although I suppose the system / AI / whatever could allow more nuance than that). With a map size and exploration rate anything like this world, I imagine most races will be unaware of most races for thousands of years, minimum.

---

Practical constraints mean a simulation is not going to run for that long in anything resembling realistic scale. The effective distances in time and space will be shorter. A 2,000 years a long time to simulate using Dwarf Fortress proper, especially for a large world. It's much more reasonable to expect the simulation to cover smaller terrain, such as a country or an island, over (low) hundreds of years, than it is to expect full planet-size simulation over thousands of years. Alternatively, as already pointed by bringing up Civilization 6 for contrast, if you have a nominal planet for nominal thousands of years you can expect the simulation to not be very detailed at the early end and the actual number of game turns using full game rules to be much less than the amount of years.

---

@Tohron: the time and place within the simulation for the first 17th or so level character capable of casting Genesis to appear, is fundamentally unpredictable. More practically, Genesis has to be cast on the Ethereal plane. For simulating a game world, characters and civilizations who go this route to grow can rightly be considered to have ****ed off to go play another game entirely. The bulk of their expansion would, literally, not touch the prime subject of the simulation, and would require different set of rules alltogether to model.

Admittedly, casting Genesis to add space is a significant way off for a starting civilization (they could use Stone Shaped tunnels earlier on), but for the food situation, it just takes one level 5 Artificer/Cleric/Archivist/Healer to make an at-will Create Food & Water item/trap, which can then supply food for 15 * 600 * 24 = 216,000 people per day. So you don't need that many crafters to support a very large population, and level 5 is very achievable using the CR < 1 creature XP farming methods outlined previously.

@Tohron: where are you getting 600 as a multiplier? A day has 24 x 60 = 1440 minutes. If it takes 10 minutes between trap resets, one 5th CL trap can produce food to at most to 144 x 15 = 2160 people. Yes, that's a lot for 7,500 gp , 600 XP and seven days of work. It becomes less impressive when you look at the number of animals you could have for the same costs and factor in that the trap doesn't produce any other goods or materials beyond simple sustenance. Additionally: if you can get any of the farming techniques to work that net you all the required animals to keep leveling, your food problems have already been covered. You don't need Create Food and Water for that.

I was using 1/round for activations. An unlimited-use command word activation item could presumably do this, and would cost 1.8 times as much as the trap (the time between auto-resetting trap activations is never defined in the rules, to my knowledge, so that's a bit of a grey area). One of the benefits of using food-generating items is space: without needing room for farming or livestock, you can house more people as your population grows exponentially (and you also don't need to defend the farms/livestock). The generated food can even be used to feed the monsters for XP farming.

Precisely because there is no other defined rule for resetting traps, it makes sense to default to casting time of the spell, which in this case is 10 minutes.

As for unlimited use command word item, it's possible to achieve that, yes, if you can ensure someone is constantly activating the item throughout the day. This only becomes practical once you genuinely have huge number of mouths to feed and several workers working in shifts to keep activating the item and distributing the food. The logistic requirements mean that such an item is unlikely to offer practical benefits over the trap or other limited use sources of the spell when it first becomes available. Leveling up is much faster than population growth, so it's possible to get 5th level characters when settlements still have population numbers in the low hundreds. You just don't need this much food this quick... which means a significant chunk of the item's potential and hence the cost of its making is wasted in the short term. If a neighbouring hostile settlement used only resources proportionate to their actual population for food production, and the rest to offense, they now have an advantage in short term confrontations and can come in take the it from your cold, dead hands.

Quote:

---

Originally Posted by Vahnavoi

@Quertus: your argument for elegance involves another abstracted category of characters that is not derived from anything in the d20 rules. Your version for two data objects only looks simpler because the assumptions behind who belongs in that category are not enumerated.

Similar relationship exists between simulation of individual objects versus picking groups from a table. The latter solution is actually computationally simpler, as it involves simulating less objects over time. But it is not more elegant because of all the additional assumptions that go into making those tables, even if those assumptions won't be visible in the finished code.

This also the flaw in your "gamist" versus "simulationist" argument. Your decrying of tables as more "gamist" is based on assuming silly reasoning behind those tables. That's baseless, the reasoning can be anything. Similar argument applies to individual simulation: that only produces more "simulationist" results than tables if the assumptions about those individuals aim for realism. Experience points are an abstract game scoring mechanic, they are fundamentally non-realist. The only thing you're simulating by applying d20 experience point rules is the d20 game system, but this isn't more accurate to source material than applying rules of population demographic tables. It's just an arbitrary decision of which parts of the source material to emphasize.

Back to elegance: consider the following progression tracks for characters:

Dependent (child) ---> Dependent (Adult) ---> Classed individual

Dependent (child) --- > Adult-in-training (either racial HD or 1st level commoner) --- > Classed individual

In both formats, the Dependent (child) category has to be specially defined for the simulation, as the base d20 system isn't concerned with child characters. But there is no reason to specially define Dependent (Adult), since we already have options in d20 rules already describe adult characters without character class. Doing things the former way may be computationally simpler, but it also loses details and requires additional assumptions about capabilities of Dependent (Adult) category as well as additional rules somewhere to explain where commoners and characters with just racial HD come from. Remember, at some point, the simulation has to lead to a playable game world, so all dependent characters have to be transformed into viable entities described by d20 ruleset at some point. The alternative is that they just disappear when actual play starts, never being anything more than background numbers.

---

Again, I don't know Dwarf Fortress, but imagine for a moment, in Hypothetical!DwarfFortress...

I'm playing a Dwarf. I sharpen my Axe, don my Hat of Disguise and/or Cloak of Invisibility, and sneak in to the Human Nursery. I slaughter the couple caregivers (however many it takes to transfer food from those Repeating Food Traps or whatever into the mouths of the babies) and bathe in the blood of thousands of dead babies. Then I move on to the toddlers, small kids, large kids, and teens. Each of these 5 categories has their own placement, and their own graphics; thus, the underlying code has to have 5 separate data objects for them (or 1 versatile data object sorted into 5 buckets, with 5 different graphics (or 5 different *sets* of graphics, if they don't all look the same to me)).

And that's fine for the game, if it wants to treat them differently that way, and make them distinct, slayable objects rather than have them "disappear into the background" (or whatever).

Me, I only care about being able to do the math on food production vs food requirements, determining the level (etc) of the level-able population, and such concerns, not on properly assigning the "Female Half-Orc Apprentice Wizard" graphic to this particular data object. And, for these purposes, one data model is clearly more elegant than the other.

Ugh, that's another vector for targeted strikes (also known as "war crimes"): gender. Not sure whether it's more or less effective than targeting Resource Gatherers (ie, those who have skill at Survival).

-----

Given that the OP was talking about "dogs", I feel it's safe to assume that this simulation really ought to be set up such that it's expected that it will run for thousands of years before the player joins, to give time for Wolves to be Domesticated into Dogs. At which point, much of your commentary wrt early game expectations seem incongruous, like we're playing very different games. I'm trying to envision a T=0 world setup that will lead to multiple races still existing thousands of years later when Dogs walk the earth, and asking myself which races with which Societies have the best chance of existing at that point. Any setup that doesn't produce playable results after thousands of years of simulation, I'm saying the fault is in the setup, and therefore something needs to be changed.

So, sure, I'm expecting some churn, expecting some races to have died out when the simulation kicks out into active play at T=X thousand years later, with there maybe being dogs for the OP to train. But I'd say any initial assumptions, any settings for the initial setup and underlying systems (NPC leveling, population numbers, tech tree, whatever) that don't involve at least the majority of the initial races still being "in the game" as playable choices probably produces a suboptimal play experience, and should be reconsidered.

-----

I honestly believe that the easiest thing... not the easiest thing to code, but the easiest thing for the poor computer to run cycles on is (modified versions of) the DMG Population Tables. And having NPC levels be based on Population really skews what's good. And, yes, I agree, lifespan does factor into that, but when you're looking at a Fibonacci Sequence to simulate Human generations, or faster growing numbers (especially for high wisdom / small size races, like Anthropomorphic Bats), only keeping the last 3 numbers out of 2,000 - 3,000 - 5,000 - 8,000 - 13,000 - 21,000 - 34,000? It really doesn't feel like you're losing much when you're comparing 68k vs 86k. Especially compared to the Level [90,50,10] vs level [50,30,10] differences from the other example.

-----

Oh, and XP are certainly a Gamist abstraction in most games. In 3e, however, where they are spent to cast spells and craft items, figure into payment for spellcasting services, and actively figure into characters' planning and strategies, they pretty much have to be an actual, existent thing in the 3e world, something that's truly part of the Simulation, rather than a mere Gamist abstraction. That's why I worded my statement about them the way I did, because 3e expectations all but mandate it being one of the exceptions to them being a Gamist abstraction.

Quote:

---

Originally Posted by Quertus

Me, I only care about being able to do the math on food production vs food requirements, determining the level (etc) of the level-able population, and such concerns, not on properly assigning the "Female Half-Orc Apprentice Wizard" graphic to this particular data object. And, for these purposes, one data model is clearly more elegant than the other.

---

It's simpler, yes. It's also a classic case of Spherical Cows in a Vacuum - by lumping everyone without a class into same category to make a problem more tractable, you end up losing details that would be relevant to the end result.

Dwarf Fortress proper, as a game, puts level of detail way ahead of simplicity. Running a massive zero player strategy game on the background to generate a playable world is very processing intensive and clunky. Proposing a Dwarf Fortress- style simulation hence, very like, includes coding you'd deem inelegant.

-----

Quote:

---

Originally Posted by Quertus

Given that the OP was talking about "dogs", I feel it's safe to assume that this simulation really ought to be set up such that it's expected that it will run for thousands of years before the player joins, to give time for Wolves to be Domesticated into Dogs. At which point, much of your commentary wrt early game expectations seem incongruous, like we're playing very different games. I'm trying to envision a T=0 world setup that will lead to multiple races still existing thousands of years later when Dogs walk the earth, and asking myself which races with which Societies have the best chance of existing at that point. Any setup that doesn't produce playable results after thousands of years of simulation, I'm saying the fault is in the setup, and therefore something needs to be changed.

---

It really isn't a safe assumption, for reasons already listed. To approach this from another angle: domesticating an animal is just a Handle Animal check in 3.5 D&D ruleset. The simulation can include both an initial population of canids and initial population of humanoids at T=0, and it will take only take as much time for domestic canids to appear as it takes for the two groups to meet and some humanoid succeeding a Handle Animal check. The problem with your argument is that you flip-flop between assumed realistic simulation of the domestication process and simulation of D&D rules, when the target of discussion is only the latter. Again, an alternative comparison can be made to Civilization 6: yes, dogs have to be domesticated, but this is a simple-to-achieve thing that pretty much every civilization in every game manages by turn 60 out 500. Even if nominally, the time from turn 1 to turn 60 runs from 4,000 BC to 1,000 BC, those are just numbers, with basically no relationship to the amount of actions taken by game characters.

Quote:

---

Originally Posted by Quertus

So, sure, I'm expecting some churn, expecting some races to have died out when the simulation kicks out into active play at T=X thousand years later, with there maybe being dogs for the OP to train. But I'd say any initial assumptions, any settings for the initial setup and underlying systems (NPC leveling, population numbers, tech tree, whatever) that don't involve at least the majority of the initial races still being "in the game" as playable choices probably produces a suboptimal play experience, and should be reconsidered.

---

Which races exist as playable option to a player once the player enters the simulation, doesn't necessarily have anything to do with what has survived world generation. If the simulation is based on 3.5 D&D core rules, it's safe to assume the player will have character generation choices based on 3.5 D&D core rules. It does not follow every character described by the core rules will survive or be present in the world. That is, if the player chooses to be an elf, it is possible they are the ONLY remaining elf. Talking about "suboptimal play experiences" I consider to be completely absurd. It is ordinary for procedurally generated worlds to only contain subset of possible civilizations in them. This extremely obvious in strategy games both old and new. The appeal of that is that you can then use the same procedural generation rules to create and play in relevantly different worlds.

In Dwarf Fortress proper, the number of civilized races is much smaller (5) than the amount of civilizations world generation can handle (160 for Large worlds). Still, since doubles are possible, smaller worlds might not contain a civilization of every possible race (for example, a Smaller world with Low number of civilizations has only 6, so two doubles would mean one race is missing). 3.5 D&D has a significantly larger pool of races to choose from even if we only include core player picks and their common opponents, for example, humans, dwarves, halflings, gnomes, elves, orcs, goblins, kobolds, lizardfolk, trolls and giants. That's 11, which would necessitate a Smaller world with High civilization count, or a Small world with Medium civilization count, to have a civilization of every race even at the start of the game.

The more races you want to be present when play starts, the larger the world has to be, and play has to begin at an earlier era, meaning everyone will be less developed.

-----

Quote:

---

Originally Posted by Quertus

I honestly believe that the easiest thing... not the easiest thing to code, but the easiest thing for the poor computer to run cycles on is (modified versions of) the DMG Population Tables.

---

It is both easier to code and run iterations on DMG population tables, since they were meant to be used by hand, by a human, on pen & paper. If I wanted to automate this, I could do this in Excell, in a day. Dwarf Fortress-style simulation, by contrast, is much more intricate, involving running aforementioned massive zero player strategy game for an extended period and logging the results. Dwarf Fortress-style simulation is borderline impossible to do by hand, or would take a prohibitively large amount of time, and on maximum settings it can be unreasonably slow even on a contemporary commercial computer.

-----

Quote:

---

Originally Posted by Quertus

Oh, and XP are certainly a Gamist abstraction in most games. In 3e, however, where they are spent to cast spells and craft items, figure into payment for spellcasting services, and actively figure into characters' planning and strategies, they pretty much have to be an actual, existent thing in the 3e world, something that's truly part of the Simulation, rather than a mere Gamist abstraction. That's why I worded my statement about them the way I did, because 3e expectations all but mandate it being one of the exceptions to them being a Gamist abstraction.

---

That's all humbug. Experience points existing in a game world does not make them less abstract or less game-like. They don't simulate anything beyond themselves: a person, in the world, explaining experience points to another doesn't have a better choice than "they're a point score you get for doing things on this arbitrary list of things, which you can then expend on things on this other arbitrary list of things". Their existence in the world is a result of simulating game rules, and a person in the world would reasonably conclude "the world works like a game".

Or, to put it simply: nothing is expected from being game-like in a simulation, when what you are simulating is a game.

@Vahnavoi - I'm not sure that our comments even have enough in common to qualify as "talking past one another". But I'll poke at a few places where we might be able to get on the same page again.

Quote:

---

Originally Posted by Vahnavoi

It's simpler, yes. It's also a classic case of Spherical Cows in a Vacuum - by lumping everyone without a class into same category to make a problem more tractable, you end up losing details that would be relevant to the end result.

Dwarf Fortress proper, as a game, puts level of detail way ahead of simplicity. Running a massive zero player strategy game on the background to generate a playable world is very processing intensive and clunky. Proposing a Dwarf Fortress- style simulation hence, very like, includes coding you'd deem inelegant.

---

Can you give any specific use case(s) where you believe relevant details are lost?

Quote:

---

Originally Posted by Vahnavoi

It really isn't a safe assumption, for reasons already listed. To approach this from another angle: domesticating an animal is just a Handle Animal check in 3.5 D&D ruleset. The simulation can include both an initial population of canids and initial population of humanoids at T=0, and it will take only take as much time for domestic canids to appear as it takes for the two groups to meet and some humanoid succeeding a Handle Animal check. The problem with your argument is that you flip-flop between assumed realistic simulation of the domestication process and simulation of D&D rules, when the target of discussion is only the latter. Again, an alternative comparison can be made to Civilization 6: yes, dogs have to be domesticated, but this is a simple-to-achieve thing that pretty much every civilization in every game manages by turn 60 out 500. Even if nominally, the time from turn 1 to turn 60 runs from 4,000 BC to 1,000 BC, those are just numbers, with basically no relationship to the amount of actions taken by game characters.

---

Some races are derived. These include half-elves, half-ogres, half-dragon whatevers - all with pretty obvious cause and effect there, right? As well as infected lycanthropes and vampires... and even breed lycanthropes in Dwarf Fortress, apparently (sadness). And my intention was for none of those derived races (even the highly effective Quasilycanthropes) to be present at T=0.

But here's the thing: "Dogs" are a mutant, derived race, too - the base race is "wolf". If one makes the appropriate Handle Animal roll, they get a trained wolf. Getting a "dog" requires Necromancy spells lost in y2k when the great Wizards of Old went away.

So, barring 2e Necromancy spells, getting "Dogs" from "Wolves" the old fashioned way, likely takes on the order of thousands of years. So, in order for the plan listed in the OP of "raising and training war dogs" to be not just technically possible but sufficiently probable, we've got to have the simulation running for thousands of years before play starts.

Which means the rest of the variables, like world size, population density, level rate, etc, need to be created such that they support this timeframe.

Quote:

---

Originally Posted by Vahnavoi

doubles

---

I had actually considered dumping, say, 100 tribes of 1000 members of each race (so 100 tribes of elves, for 100,000 elves total, 100,000 Humans, 100,000 Dwarves, etc) as simulation start as one way to maximize the odds that a given race had survived to the start of gameplay.

Quote:

---

Originally Posted by Vahnavoi

The more races you want to be present when play starts, the larger the world has to be, and play has to begin at an earlier era, meaning everyone will be less developed.

---

Um... that doesn't follow. I imagine, if I made a Dyson Sphere the size of Jupiter's orbit, and placed 20 races randomly in that sphere, they'd be able simulate longer into being more advanced than what you were picturing with 5 or 6 races. Line up all your variables to get to "advanced enough to have dogs" and "desirable odds of number of races remaining".

Quote:

---

Originally Posted by Vahnavoi

That's all humbug. Experience points existing in a game world does not make them less abstract or less game-like. They don't simulate anything beyond themselves: a person, in the world, explaining experience points to another doesn't have a better choice than "they're a point score you get for doing things on this arbitrary list of things, which you can then expend on things on this other arbitrary list of things". Their existence in the world is a result of simulating game rules, and a person in the world would reasonably conclude "the world works like a game".

Or, to put it simply: nothing is expected from being game-like in a simulation, when what you are simulating is a game.

---

They don't have to simulate anything beyond themselves, any more than the carpet at the tavern the PCs are staying at does, to be a real game object in their world that the PCs can (and do) consciously interact with. But I think we're lost in semantics at this point.

@Quertus: the overarching theme of my posts concerns this comment from the OP:

Quote:

---

Originally Posted by Samael Morgenst

The setting is a procedurally generated sandbox-like world (think about Dwarf Fortress) very similar or identical to core Dungeons and Dragons edition 3.5.

---

I'm focusing on "think about Dwarf Fortress" bit, using real attributes of the named game to gauge what this world could reasonably include.

You, initially unaware of Dwarf Fortress, have some intuitions and ideas of how the world would be generated that are at odds with realities of the program, and other comparable programs. I consider it relevant to point these out when they cause you to make unsafe assumptions that'd matter for player strategy.

Moving on:

Quote:

---

Originally Posted by Quertus

Can you give any specific use case(s) where you believe relevant details are lost?

---

Most notably, neither children nor commoners nor characters with racial HD are simply "resource sponges" under normal d20 rules. They can be assigned tasks and contribute as untrained labor, as crafters, as survivalists, as animal handlers, so on and so forth. In terms of strategy game, each such unit of early population can hence be given a number for how much they'd produce (in terms of food and goods), not just how much they consume, and this varies based on race.

For a simple example, a 1st level human commoner with average wisdom has extra 4 skill points and 1 extra feat over a 1st level elf commoner with average wisdom. This means the human commoner can take both Self-sufficient and Skill Focus (Survival) feats, as well as spend four points to get 2 ranks in Survival (cross class), to get a +7 modifier, total, to survival checks, while still having 8 skill points left over. The elf can only manage +5 (ranks and Skill Focus) and has only 4 skill points left over.

So, based on Survival rules, we can say that a human commoner assigned to early food production can sustain 1 extra population over an elf commoner assigned to early food production. We can also say the human has more potential to succeed if assigned to another task - since they have more skill points left to improve craft or profession.

These kind of differences add up in a ground-up simulation, during the years when no-body has 1st level in their PC class yet. They also accumulate for every subsequent generation, for the period when those characters are young adults in training. If, instead, everyone is the same kind of abstract "resource sponge" before reaching 1st level, you lose those differences. Effectively, all your wizard apprentices do no useful work. They do not even sweep floors or copy texts for their masters.

In comparison to existing strategy games, the difference is similar to adjusting starting tech levels. If you start a Civ 6 game at an era other than Ancient, the competitive balance shifts: some civilizations lose their competitive edge because the game starts at a point where their special features have already ceased to matter, while others gain a new advantage because they don't have to fight to survive to the point where their special qualities start to matter.

Quote:

---

Originally Posted by Quertus

Some races are derived.

---

Yes, but how they are derived and inherited doesn't have to follow realistic rules. Again, the problem is that you flip-flop between assuming realistic simulation of domestication and inheritance processes and simulation of game rules, but the simulation is only ever of the game rules. Dogs can exist by a fiat decision simply because they are part of 3.5 core, with the only difference between Dog (wild) and Dog (domestic) being whether someone passed a Handle Animal check. Or the number of generations between Wolf (tame) and Dog (domestic) can be significantly lower than reality.

The same applies to various half-creatures. Mechanically, in 3.5 D&D, majority of these are templates. You can slap them on offspring in one generation. Likewise, the time from Lycanthrope (infected) to Lycanthrope (born) is one generation, as the former template can be slapped on an existing creature. It's not safe to assume a world age of thousands of years based on existence of such creatures, because by the game rules, thousands of years are not required.

Quote:

---

Originally Posted by Quertus

I had actually considered dumping, say, 100 tribes of 1000 members of each race (so 100 tribes of elves, for 100,000 elves total, 100,000 Humans, 100,000 Dwarves, etc) as simulation start as one way to maximize the odds that a given race had survived to the start of gameplay.

---

These are unfeasible numbers for Dwarf Fortress proper. In terms of possible starting civilizations, the cap is 160 for Large worlds. You'd need a different type of simulator to deal with these numbers.

Quote:

---

Originally Posted by Quertus

Um... that doesn't follow. I imagine, if I made a Dyson Sphere the size of Jupiter's orbit, and placed 20 races randomly in that sphere, they'd be able simulate longer into being more advanced than what you were picturing with 5 or 6 races. Line up all your variables to get to "advanced enough to have dogs" and "desirable odds of number of races remaining".

---

There is, as of now, no simulator, certainly not Dwarf Fortress, that could simulate a world that big. Meanwhile, the actual reality of Dwarf Fortress is that number of civilizations tends to go down from the initial lineup as a years pass: weaker civilizations often starve or are terminated by competing neighbours or megabeasts. Furthermore, as repeatedly noted, world size places constraints on number of civilizations. From which follows the exact thing I already noted: the more races you want to be around when play starts, the larger the world has to be and the earlier the starting era. In this style of simulation, there is no way to guarantee everyone will always be around, you can only stop the world generation when the results look promising. Notable, Dwarf Fortress proper has special eras for when nearly all or all fantastic creatures have died off, for when every civilization has died off, and for when no civilization actually survived very long.

---

Regarding games and simulations:

Quote:

---

Originally Posted by Quertus

But I think we're lost in semantics at this point.

---

The entire point is that GNS terms have bad semantics - the distinction you're trying to make is useless. Your life will be improved by never using "gamism", "simulationism" and "narrativism" again - as opposed to "game", "simulation" and "narrative", which have robust definitions outside outdated RPG theories. :smalltongue:

Quote:

---

Originally Posted by Vahnavoi

I'm focusing on "think about Dwarf Fortress" bit, using real attributes of the named game to gauge what this world could reasonably include.

You, initially unaware of Dwarf Fortress, have some intuitions and ideas of how the world would be generated that are at odds with realities of the program, and other comparable programs. I consider it relevant to point these out when they cause you to make unsafe assumptions that'd matter for player strategy.

---

Perfectly reasonable to point out places where I might have misconceptions, especially when I’m a self-reported ignoramus.

That said, even if I knew Dwarf Fortress, I’d focus on “simulating 3e”, considering DF to just be a relevant example of doing so.

Quote:

---

Originally Posted by Vahnavoi

neither children nor commoners nor characters with racial HD are simply "resource sponges" under normal d20 rules. They can be assigned tasks and contribute as untrained labor,

---

Ah, so that’s a misunderstanding and a correction.

The misunderstanding is simple: we’re on the same page wrt everything but “children” (assuming you’re not treating 20 HD Dragons with no class levels as “untrained laborers” :smallwink:).

Yes, I was thinking of children exclusively as “resource sponges”, because that matches other games I’ve played, matches my Simulationist expectations for babies and small / lazy children, and met my Gamist desires not to have “reproduce as fast as possible” be the obvious correct dominant strategy for all races (tied into a Simulationist belief that it shouldn’t be the “always optimal” strategy).

But, sure, had I written such code, it would be trivial to change pre-class Humanoid Children from “resource sponge” status to… “Aid Another” actions, probably.

(EDIT: thinking about it… I think, if I were writing the Simulation, *working* children would end up Commoners; *learning* “resource sponge” children would get (or have a chance at) better classes. Which also ties into my belief in PCs / adventures being “children of nobles”, with the opportunity to study advanced techniques, rather than farm boys who happened to pick up a sword or spell book.)

Quote:

---

Originally Posted by Vahnavoi

So, based on Survival rules, we can say that a human commoner assigned to early food production can sustain 1 extra population over an elf commoner assigned to early food production. We can also say the human has more potential to succeed if assigned to another task - since they have more skill points left to improve craft or profession.

---

Absolutely, Humans are more adaptable, We’ve been on the same page here the whole time.

That they’re also better at specializing at arbitrary things (as opposed to Elven Weapon proficiencies, or Petals and Anthropomorphic Bats beating Humans at Per capita resource gathering) is something I assumed, but didn’t state, because I didn’t know what specific feats might allow such.

Quote:

---

Originally Posted by Vahnavoi

during the years when no-body has 1st level in their PC class yet

---

That’s an assumption I’m *not* making. That is, we could simulate this many ways, *many* of which involve classes being available at the beginning. So my answers are often complex, as they often are designed to address the various ways one might create such a simulation, or conditional, stating their limited scope up front.

Since it keeps coming up, does Dwarf Fortress have class unlocking mechanics?

Quote:

---

Originally Posted by Vahnavoi

Dogs can exist by a fiat decision

---

Sure, races and classes can exist by fiat decision. But if we go down the rabbit hole of allowing derived races at T=0, then I’ll argue Quasilycanthropes are a strong contender.

Quote:

---

Originally Posted by Vahnavoi

Yes, but how they are derived and inherited doesn't have to follow realistic rules. Again, the problem is that you flip-flop between assuming realistic simulation of domestication and inheritance processes and simulation of game rules, but the simulation is only ever of the game rules. Dogs can exist by a fiat decision

---

Flip-flip? Ah, not exactly. I follow a hierarchy, which one could oversimplify to: stated givens, 3e rules, reality.

Even Breed Lycanthropes, which I was willing to consider a “base” race, are considered derived in Dwarf Fortress.

If there were a rule, “dogs are a base race”, it would stick out and be incongruous with the “start at the beginning and evolve a civilization” feel of the question, *especially* if coupled with things like “no classes at game start” and “start naked”,

In the absence of such a rule, we look at how dogs are derived. No rule in our exceptions, no rule in 3e (except the conspicuous *absence* of similar rules from 2e), so we look to reality for our rules for creating dogs.

One consistent process that terminates at various levels, not arbitrary flip-flopping.

Quote:

---

Originally Posted by Vahnavoi

These are unfeasible numbers for Dwarf Fortress proper. In terms of possible starting civilizations, the cap is 160 for Large worlds. You'd need a different type of simulator to deal with these numbers.

---

I have no intention of limiting myself to the limitations of the Dwarf Fortress Simulation, merely to what could be simulated. Granted, the Dyson sphere wasn’t realistic to simulate, it was merely an example to exemplify how tweaking variables can change outcomes.

Quote:

---

Originally Posted by Vahnavoi

number of civilizations tends to go down from the initial lineup as a years pass: weaker civilizations often starve or are terminated by competing neighbours or megabeasts.

---

Wandering megabeasts? That sounds cool.

Quote:

---

Originally Posted by Vahnavoi

The entire point is that GNS terms have bad semantics - the distinction you're trying to make is useless. Your life will be improved by never using "gamism", "simulationism" and "narrativism" again - as opposed to "game", "simulation" and "narrative", which have robust definitions outside outdated RPG theories. :smalltongue:

---

I may need to research this further…

Quote:

---

Originally Posted by Quertus

The misunderstanding is simple: we’re on the same page wrt everything but “children” (assuming you’re not treating 20 HD Dragons with no class levels as “untrained laborers” :smallwink:).

---

Untrained labor is a specific rule governing earning money using Profession checks without any actual ranks in the skill. Yes, even 20 HD dragon will count as untrained labor if they have no actual professional training. Far more relevantly, dragons in 3.5 D&D tend to be LA:- races, so they aren't available as player options. In a Dwarf Fortress-type simulation, they are more like to be counted among megabeasts (more on those later).

Quote:

---

Originally Posted by Quertus

Yes, I was thinking of children exclusively as “resource sponges”, because that matches other games I’ve played, matches my Simulationist expectations for babies and small / lazy children, and met my Gamist desires not to have “reproduce as fast as possible” be the obvious correct dominant strategy for all races (tied into a Simulationist belief that it shouldn’t be the “always optimal” strategy).

---

Population is power no matter how you try to twist the base rules. The actual limits to "reproduce as fast as possible" as a strategy are food production, available space, neighboring hostiles, opportunity costs, etc.. As noted in the earlier tangent about Create Food & Water items, trying to naively maximize one thing isn't necessarily the smartest thing to do. Relevantly, some of the faster-reproducing races have Wisdom or Intelligence penalties, so they are more likely to bump into early limits of food production due to lacking Survival, Profession, Craft etc. skills.

Quote:

---

Originally Posted by Quertus

That they’re also better at specializing at arbitrary things (as opposed to Elven Weapon proficiencies, or Petals and Anthropomorphic Bats beating Humans at Per capita resource gathering) is something I assumed, but didn’t state, because I didn’t know what specific feats might allow such.

---

Quick checking d20 SRD is an easy way to check what options would reasonably be available to 1st level characters, as those options are 1) core, 2) what would be reasonable available to simulation maker as source material. See also: aforementioned Incursion roguelike, which does just that, taking material from 3.0 SRD under OGL.

Quote:

---

Originally Posted by Quertus

That’s an assumption I’m *not* making. That is, we could simulate this many ways, *many* of which involve classes being available at the beginning. So my answers are often complex, as they often are designed to address the various ways one might create such a simulation, or conditional, stating their limited scope up front.

---

I agree there's more than one way to set up the initial conditions, the specific subtopic was which kinds of details are lost doing things the way you suggest, versus the way I suggest. The group that is most relevant to that discussion is not children, it's young adult character who haven't achieved a class yet. Giving a human Wizard apprentice the abilities of a 1st level commoner makes them an economic factor for the 2 to 12 years they spend in training - they can effectively earn their own keep, even if it's only 1 silver piece per day. Sidenote: work done during training is also an explanation for where starting money for class comes from. The alternative is that wealth is created from nothing upon reaching class.

Quote:

---

Originally Posted by Quertus

Since it keeps coming up, does Dwarf Fortress have class unlocking mechanics?

---

If I recall right, Dwarf Fortress has an intricate combination of point buy and learning by doing. That is, if your Embark points weren't enough to make your dwarves sufficiently cool, you have to build them a shooting range or training weapons and then have them play around with those, hopefully without killing each other. It's a sufficiently different system that you cannot directly draw conclusions from it and apply them to a system such as the OP proposed, where character advancement is based on 3.5 D&D style experience point acquisition. This said, Dwarf Fortress does have analogues to some of the strategies discussed here, such as the toad farm, trap obstacle course and the village wrestling tournament. They just have different implications: the Dwarven Wrestling Champion is the best wrestler. This doesn't also somehow lead him to become the greatest wizard. Additionally, skills in Dwarf Fortress can become worse if not used for extended periods. So the Dwarven Wrestling Champion cannot just lock themself in a tower to study wizardry for years and expect to remain the best wrestler. So on and so forth.

Quote:

---

Originally Posted by Quertus

Sure, races and classes can exist by fiat decision. But if we go down the rabbit hole of allowing derived races at T=0, then I’ll argue Quasilycanthropes are a strong contender.

---

There's a pretty strong difference between assuming mundane animals as baseline versus late generation lycanthropes. Even in D&D, a background of mundane, non-templated animals is usually given, but the selection of monsters and especially templated monsters varies by setting. Undoubtedly, a D&D simulation includes some monsters that you'd consider "derived" from the start, but majority of these are LA: -. They aren't candidates for player characters nor candidates for initial civilizations. They are obstacles for those civilizations (again: see megabeasts).

Quote:

---

Originally Posted by Quertus

Flip-flip? Ah, not exactly. I follow a hierarchy, which one could oversimplify to: stated givens, 3e rules, reality.

[ . . . ]

If there were a rule, “dogs are a base race”, it would stick out and be incongruous with the “start at the beginning and evolve a civilization” feel of the question, *especially* if coupled with things like “no classes at game start” and “start naked”.

---

That's the same thing. There is no good reason to assume the simulation will at any point realistically model timespans involved in domestication and inheritance, and many good reasons based on limitations of simulations to assume otherwise. Majority of games of this genre cut corners such as this at the early end. In D&D 3.5 itself, literacy is the best example: a character not specifically marked as illiterate, knows how to read. There is no extensive learning process behind this, even an illiterate character can become literate with token cost of 2 skill points, and this applies to every language the character is able to speak.

So, while in reality, invention and learning of reading and writing are long and involved processes, in the simulation, we can expect the capacity to be there at T=0 and the limitation is building the requisite infrastructure to produce and store books. Same applies to majority of Craft, Profession and Performance skills: we can except "new" goods and professions to appear at the rate at which characters hit the relevant skill modifiers, not however long it took in reality.

You can call it "incongruous", but such incongruities are normal for the genre.

Quote:

---

Originally Posted by Quertus

I have no intention of limiting myself to the limitations of the Dwarf Fortress Simulation, merely to what could be simulated. Granted, the Dyson sphere wasn’t realistic to simulate, it was merely an example to exemplify how tweaking variables can change outcomes.

---

Of course tweaking variables changes the outcome, but once the variables are sufficiently far from what the OP posited, no useful information can be gleaned from them for purposes of answering the OP's question.

Quote:

---

Originally Posted by Quertus

Wandering megabeasts? That sounds cool.

---

Yes, Dwarf Fortress has various extremely powerful monsters that influence course of history. This pool includes classics such as Titans and Dragons, also present in D&D, as well as procedurally generated monstrosities - think randomized Arch Demons and the like. A world has limited population of these and the world's historical era will change depending on how many are left. For example, if one singularly powerful Dragon is present at world generation, you may quickly get either "Age of [Foo] the Dragon", acknowledging it as the defining great power for remaining civilizations, or "Age of Emptiness" if that same Dragon kills off early civilizations in the crib.

Quote:

---

Originally Posted by Quertus

I may need to research this further…

---

From the Wikipedia article concerning GNS theory:

"On December 2, 2005, [Ron] Edwards closed the forums on the Forge about GNS theory, saying that they had outlived their usefulness."

It was never a good theory. It was internet forum posts level, built on earlier internet forum posts level musings. Its successor, the Big Model theory of roleplaying games, did not fare much better. Here is what RPG museum has to say on the topic:

"The Big Model has been significantly criticised and is no longer widely used, even by many of the people who liked it when it was new and current. For example, Vincent Baker has said that, while it was a useful tool to diversify thought around what role-playing games could be (i.e. that there was more than a single type of RPG that could be played), the attempt to categorise all RPGs and all players (e.g. using the GNS creative agendas) do not hold up, and furthermore RPG design has moved on and left the Big Model behind."

The only lasting legacy these theories had was the use of certain phrases, such as the words "gamism", "simulationism" and "narrativism", by people who happened to be part of early 2000s internet hobby discussion. Majority of people who use those words, use them wrong, with no significant relationship to their special definitions in these theories. Outside of these theories, there isn't enough settled common usage for them to even net their own dictionary definitions.