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

So, suppose the world is transparent enough for people to be able to understand, in character, the connection between “killing” (etc) and “power” (like… the world of SAO season 3?). And, similarly, that “power” is predictable enough that spells are clearly able to be ranked in power categories, and between that and spell slots, exact power “level”, and the importance of Prime Requisites (ie, Intelligence gives Wizards more spells) is known. As are things like whether or not you can Empower Owl’s Grace (or whatever) and company, and probably even *why*.

In such a world, about the lowest level you’d expect to see is level 10 or so, the level where hunting fairly safe game in fairly safe nearby areas can no longer allow one to gain XP.

That’s… not at all the expectations 3e was built under. Nor is it what the OP was looking for.

So what can we change to try to bridge this gap?

Adding in some kind of nebulous “challenge” requirement could maybe get us something closer to the expected level distribution, only… I don’t think we’d get quite what the OP described.

Maybe if only special people (like the PCs?) actually leveled with the listed level system? I'm not a fan of this method, making PCs and NPCs inherently different, but it's something of an option to create the hoard of low-level NPCs.

Actually… so long as we have the “pick your poison” setup that the OP implied, I doubt we’d ever end up anywhere near the expected level distribution, as people could always pick low-threat challenges (ie, low CR challenges) once they have a few levels under their belt.

More “actually”, I feel that the “pick your poison” aesthetic of the OP all but necessitates the concept of a very Gamist world, where different zones spawn different extremely predictable threats, and it’s not really possible to deplete an area of its spawns.

In this setup, (having our knowing someone with) Knowledge: Local seems tied in value with having a good start location.

So what might a good start location look like?

I’m thinking deep in the middle of the Elven Empire, where everything is either “animals” or “friendly” is a good place to safely get to level 10 or so. Some helpful bonuses might include someone high level to go hunting with (say, your dad, who is trying to help you overcome that psychological difficulty harming animals), someone with a trick to create free undead (Tainted Sorcerer, Blood Money, Merilith Simulacrum, whatever) to release from their control for you to safely practice on (zombies are slow and mindless), someone with Knowledge: Local who knows where good spots are,, and someone who can Resurrect you, just in case.

Quote:

---

Originally Posted by Samael Morgenst

Realistically speaking, most people would not know much of the world in a setting that predates modern information system.

---

Agreed, but... the 2000-year-old, NI level Wizard who trained me probably knows a lot. And they'd be kinda dumb not to give their apprentice useful information if said apprentice has a goal of, how do you say, "power leveling" per the OP, no?

Quote:

---

Originally Posted by Samael Morgenst

But I would not appreciate completely unjustified metagame knowledge. Expecially from a character like an Int 4 half-orc Barbarian.

---

That's fair. :smallbiggrin:

Quote:

---

Originally Posted by Samael Morgenst

Well, killing 40 between foxes, raccoons and ferrets (animals with similar CR) during the course of a year in a setting similar to medieval europe or precolonization americas is not going to cause extinctions.

But the issue of sustainability is one of the reasons why wild boars (and probably, dire boars too) are my prey of choice.

1. they are infestant, tough and reproduce quickly (here in Italy we have a problem with boars in many cities)
2. they damage crops and are dangerous to livestock, so most feudal age nobles allowed peasants to hunt them - or at least did not persecute boar hunters as hard as they did with deer poachers
3. they're edible and provide a lot of useful hide, meat, bone and fat.

---

Fair enough, a single individual is unlikely to completely eradicate the Dodo Bird population in their excitement to Power Level ("Safe Level"? Whatever it is they're doing.). I guess I was more concerned with the concept of the entire society attempting to Power Level. In a more Simulationist world, the poor Bandit population just won't survive the town's daily excursions for Darkcraft XP components. :smallamused:

Quote:

---

Originally Posted by Maat Mons

Spellcaster training is probably expensive, so it makes sense that not everyone can actually do that.

---

I guess it depends on whether the Master or the Apprentice is the one footing the bill as to whether they consider it "expensive"...

Re: character knowledge:

You are all too concerned with whether a character knows specific game details and not nearly concerned enough about all the things that are genuinely unknown even to you, such as enemy placement and terrain.

On to other things: A poster up thread mentioned Incursion roguelike. I played the last working version to completion several times. Some thoughts:

Incursion starting characters have way more hitpoints than standard. It also uses "armor as damage reduction" rules (significantly more damage reduction than tabletop) and 3.0 rules for magic weapon requirements to pierce damage reduction. This makes armored characters, friend and foe alike, much sturdier, and several enemies undefeatable without right magical weaponry.

This means early character level squatting is easiest for characters that don't benefit from it much. A fighter can, with careful play, survive extremely long strings of dungeon level 1 and 2 encounters to grind XP. This won't net them equipment necessary to beat later foes. You can have all the hitpoints and all the feats, yet still find yourself losing by attrition due to being unable to scratch your enemies. Mages face a similar difficulty, since they need new spellbooks for spell access. The best classes to squat with are Clerics and Druids, since they don't depend on equipment as much.

The best place to squat is character level 2, dungeon level 2. This is due to presence of the ecumenical temple, a place where you can pray and sacrifice to majority of the game's gods. It's possible to lure monsters to the temple and sacrifice them at multiple altars, currying favor and worshipper benefits from multiple compatible gods. This allows for power growth independent of character level, allowing a character to better stay alive without decreasing XP gains. One god, in particular, can be offended to summon overleveled but underequipped monsters, which can then be killed and sacrificed for massive XP and piety gains. This a good example of a scenario specific strategy: you can't derive this from basic game mechanics, every new character has to go through the trouble of getting to dungeon level 2 and finding the temple first. Get blocked by unfavorable terrain generation or a tough monster on the way? Better forget about.

Additionally, this and similar strategies in other roguelikes and world simulators rarely last to the late game. At best, they shortcut you past the first half of level progression. Then you try to leave the low level area and find the power increase isn't a substitute for good tactics. Extrapolated to tabletop, cheesing your way to level 10 is an invitation for people who got to level 10 the hard way to kick your under-equipped and underplayed behind. Fittingly, my first win in Incursion was a constitution-based single class, single deity Fighter, before I'd figured out multiclassing or the above pantheon-cheesing tactic. It took me less time to beat a roguelike without cheap tricks, than it took me to find the cheap tricks.

Quote:

---

Originally Posted by Crake

XP is gained for overcoming challenges, not killing things.

Just like in the real world, the best way to level up in dnd, is set yourself a realistic, but challenging goal, and achieve it. Then set the next one, and achieve that, and so on. That's literally all it takes, but most people are satisfied with the status quo, which is why most people are level 1 commoners.

---

Depends on DM . Killing things , resolving situation or impress DM works for me players .

When discussing the world general populace and why everyone isn't a wizard or level 10+, what no-one here seems to have considered yet is motivation. If your father is a wool merchant and you've been helping him run the business since you were knee-high, why would you do anything else?

Elon Musk, Mark Zuckerberg, Jeff Bezos, Bill Gates: four individuals in the tech industry who have made billions. Are they smart? You bet. Are they the smartest people on the planet? Not even close. Clever though they undoubtedly are, I don't think many people would rate their intelligence higher than, say, NASA engineers or quantum scientists working at CERN. But some people want to make money, and devote a lot of time to doing it. And if they are smart (and not a little lucky) they can become wealthy - sometimes, astoundingly-so. But many people, in fact I would suggest the majority, are only interested in earning enough to satisfy their wants and needs. Once you have enough to be happy, why exert yourself needlessly?

Back to 3.5 world and now you're a farmer/storekeeper/town guard. You work during the day, have a couple of mugs of ale in The Slaughtered Lamb inn in the evening and then go home to your family. You're making a decent living and no-one is trying to cut off your head. 99% of your days involve no challenge and that's the way you like it. I believe (and create/run my gaming world accordingly) that that explains why normal folk rarely get about level 2 or 3: there's no incentive to ever do so.

PCs on the other hand are generally greedy, power-hungry sociopaths. And so they work (grind) their way as far as they possibly can...

One big question is “Lawful” vs “Chaotic”: does society provide good to amazing starting gear, or is the individual left to their own devices? Despite traditionally being labeled “chaotic”, elves are actually about second only to Necropolitan societies wrt how much they’d reasonably invest in their young. So, much like in many modern video games, starting characters can likely expect to be kitted out with spare bling from high level adventurers.

For the purposes of this thread, I feel it’s assumed that the character in question has to “make their own way”, and that’s a fine constraint; I just felt the need to point out how this is both unrealistic and suboptimal.

Quote:

---

Originally Posted by Vahnavoi

You are all too concerned with whether a character knows specific game details and not nearly concerned enough about all the things that are genuinely unknown even to you, such as enemy placement and terrain.

---

All? Nah, I’d say I’m far more concerned with “enemy placement” than “character knowledge”. Or, to flip that, one of my… premises? Thesis? Whatever… is that “starting location” is one of the most critical elements here.

That said, I’m concerned with whether “the area” / enemy placement / available threats is known or unknown / the extent to which it is knowable.

This is part of why I contended that an optimal *ahem* a very good starting point might well be in the middle of the elven empire, where one can (relatively) safely level up on local wildlife, with no threat of anything else causing problems / suddenly slaughtering the character.

Quote:

---

Originally Posted by Daisy

When discussing the world general populace and why everyone isn't a wizard or level 10+, what no-one here seems to have considered yet is motivation. If your father is a wool merchant and you've been helping him run the business since you were knee-high, why would you do anything else?.

PCs on the other hand are generally greedy, power-hungry sociopaths. And so they work (grind) their way as far as they possibly can...

---

That’s part of the problem: one can spend decades studying weaving techniques, or an afternoon killing stuff to gain the same skills. If it’s known how the world works, then many of your lazy townsfolk will likely gain their skills through power leveling.

OTOH, if it’s not known how the world works, a Wizard might (should), in character, be attempting to hone their craft by scribing magic scrolls.

Quote:

---

More “actually”, I feel that the “pick your poison” aesthetic of the OP all but necessitates the concept of a very Gamist world, where different zones spawn different extremely predictable threats, and it’s not really possible to deplete an area of its spawns.

---

It's a semi-realistic sandbox world. Different biomes are inhabited by different fauna, some are more dangerous than others but there are exceptions.

Also, I guess we should assume that some exceedingly large, immortal and destructive monsters do not exist or are somewhat confined, else life would not be present.

Quote:

---

When discussing the world general populace and why everyone isn't a wizard or level 10+, what no-one here seems to have considered yet is motivation. If your father is a wool merchant and you've been helping him run the business since you were knee-high, why would you do anything else?

Elon Musk, Mark Zuckerberg, Jeff Bezos, Bill Gates: four individuals in the tech industry who have made billions. Are they smart? You bet. Are they the smartest people on the planet? Not even close. Clever though they undoubtedly are, I don't think many people would rate their intelligence higher than, say, NASA engineers or quantum scientists working at CERN. But some people want to make money, and devote a lot of time to doing it. And if they are smart (and not a little lucky) they can become wealthy - sometimes, astoundingly-so. But many people, in fact I would suggest the majority, are only interested in earning enough to satisfy their wants and needs. Once you have enough to be happy, why exert yourself needlessly?

Back to 3.5 world and now you're a farmer/storekeeper/town guard. You work during the day, have a couple of mugs of ale in The Slaughtered Lamb inn in the evening and then go home to your family. You're making a decent living and no-one is trying to cut off your head. 99% of your days involve no challenge and that's the way you like it. I believe (and create/run my gaming world accordingly) that that explains why normal folk rarely get about level 2 or 3: there's no incentive to ever do so.

PCs on the other hand are generally greedy, power-hungry sociopaths. And so they work (grind) their way as far as they possibly can...

---

I mostly agree.
But I also think that for most people misery and lack of opportunities play a role too, plausibly. In any era.

Quote:

---

Originally Posted by Quertus

That’s part of the problem: one can spend decades studying weaving techniques, or an afternoon killing stuff to gain the same skills. If it’s known how the world works, then many of your lazy townsfolk will likely gain their skills through power leveling.

---

On the other hand, the neighbor kid with lycanthrope parents was born with 26 hit dice and the kind of crazy numbers that come with it. Kind of disheartening when you contemplate how much work it would take you to get to where he started at.

Pure RAW: a creature with CR less than 1 counts as CR 1 for all encounter calculations except the amount of XP gained for defeating it. A lvl 8 PC can get XP from defeating a CR 1 creature, and thus from a CR 1/10th creature. Toads are CR 1/10th, breed very quickly, cannot escape you if you start attacking them, and cannot attack back. 1200 toads will get you to lvl 9. After that, you need to find some good CR 2s. It'll take 40 CR 2s to reach lvl 10, then 40 CR 3s to reach lvl 11, and so on. If you only pick fights "tough enough" to give you any XP at all, lvl 1 to lvl 20 is, at absolute maximum, 1640 fights away.

But that's the absolute maximum, fighting the weakest things that still technically advance you. Adventurers can generally punch a bit higher up than "the literal weakest creatures that technically grant you XP". The game expects you to fight something closer to 247 fights in order to get from 1 to 20. If you punch high enough out of your weight class, you could theoretically get there in just 19 fights, leveling up after each epic battle.

The biggest difficulty in answering your question is "non-cheesy". There is no set community definition for what counts as cheese, there's just 10000 people who all have their own opinion on what stuff is "perfectly fine", what stuff is "practical optimization" and what stuff is "cheesy tryhard nonsense". The only person who can answer what's too cheesy for this game is the DM running it. Maybe they think toad genocide is fine. Maybe they think energy drain/restoration loop stuff is fine. Maybe they think teleporting past random encounters counts as an "Easy If Handled Properly" encounter, since you used resources to deal with an obstacle that could've slowed you down on a time-sensitive mission. Maybe they think setting up a Fight Club where you can win fights without anyone dying is fine. Who knows?

Once you know where your DM's cheese line is, figuring out the safest options from those that remain on the table should be easy.

Quote:

---

Originally Posted by Samael Morgenst

Suppose you had to make your character grow in experience, levels and - possibly - wealth and power, while minimizing risks as much as possible.

---

Quote:

---

Originally Posted by Quertus

If I take the flaw… name… uh… the one that says you need to make a DC 15 Will save to harm animals… suddenly, it’s a lot harder and more challenging to kill animals. So, by Challenge logic, setting the goal of “kill animals” is a challenging goal for the character, and they’ve got to overcome the challenge of “I don’t like killing animals” in addition to the actual challenge involved in killing the animal, so they would earn *more* XP for killing animals than the average hunter - significantly more for low CR threats like bats 🦇 or foxes 🦊. And this is especially important if the world is more Simulationist than Gamist, if the local wildlife is finite rather than simply being random encounters and/or challenges with set DCs to locate specific types of wildlife (more XP!).

---

Quote:

---

Originally Posted by Quertus

I’m thinking deep in the middle of the Elven Empire, where everything is either “animals” or “friendly” is a good place to safely get to level 10 or so. Some helpful bonuses might include someone high level to go hunting with (say, your dad, who is trying to help you overcome that psychological difficulty harming animals), someone with a trick to create free undead (Tainted Sorcerer, Blood Money, Merilith Simulacrum, whatever) to release from their control for you to safely practice on (zombies are slow and mindless), someone with Knowledge: Local who knows where good spots are,, and someone who can Resurrect you, just in case.

---

Quote:

---

Originally Posted by Samael Morgenst

Personally, I would prefer that the character was at least present during the hunt / battle, even if not necessarily involved in first person.

Even better, using traps / summons / trained animals to weaken the kill, but take it down himself.

---

So, how's this, then:

Every morning, my Elf Wizard goes out with their father (a high-level Arcane Archer, probably) to check the traps. If they can make their DC 15 Will save, they put the animal out of its misery; if they can't, [choose one: their father kills it, they let it go, they feed it in order to keep it alive and trapped for another try tomorrow]. Maybe they earn extra XP attempting to make "kills of opportunity" out of all the wildlife they see (just how many birds, squirrels, etc, can you see while walking through the woods for 8 hours? I'm thinking "a lot", which is a lot of CR <whatever DC 15 Will save translates to> challenges to attempt to overcome), maybe not, depending on the underlying physics engine (ie, does it disrupt their "traps" plan to do so?).

Eventually (probably really quickly), they make enough XP for enough levels and enough money for enough equipment that they a) can feel confident walking the really safe woods "alone", even if a stray cat should leap out and attack them; b) aren't actually alone, thanks to (Improved) Familiar and Leadership; c) can start facing off against the slow, mindless, uncontrolled zombie creatures that are animated for their training purposes (by someone who can do it for free / without material components). Even if they have to pay for casting services, a 20 HD zombie should run... 600 gp? It's a bit of a temporal setback to make that much money, but they're an elf, and this is just about the safest way to level against low-mid level foes I can think of. Or maybe their Mirror Mephit Improved Familiar could make a Simulacrum of a Marilith for free Animate Dead? That's about the least cheesy thing you can do with a Mirror Mephit, no?

Once they get Fabricate (and maybe a level of Arcane Archer, just to get their dad off their back), they can spend a decade Crafting (and using Fabricate for "Crafting") to make up for their rather low wealth relative to their level. And by "make up for", I mean likely put them well into Epic wealth, depending on just how clever they are with their crafting. Maybe they have Item Creation Feats to nearly double their effective wealth, maybe not. Shrug.

At this point, they've got to leave the Elven Woods if they want to find anything "worth XP" that isn't an ally. Hopefully, someone has some Knowledge: Local that they can show them good spots on a map... and hopefully, nothing's changed in the decades or centuries since those elves last visited those hunting grounds.

This sounds like a terrible plan from a Simulationist perspective. So, alternately, they could go hunting the Planes... or have creatures from other planes summoned for them to fight. I don't have any specific ideas yet, other than maybe go with a larger party, especially if hostile animals are likely to be a potential issue.

So, there's my idea for the safest way to level grind an Elven Wizard in this scenario. Doubtless it's highly suboptimal, and the build will be quite the head-scratcher, but that's what I've got so far.

At low levels, it should be extremely safe, repeatable, low on complications, able to generally handle the complications of "Random encounter: stray cat" with the high-level supervision, be in character (if dear old dad is disgusted with his squeamish child, and said child is trying to make dear old dad proud), possessing minimal cheese, able to work without a communal level grinding infrastructure (requiring only the Wizard's class-mandated teacher, and the biologically-mandated parental unit), and legal. At mid levels, is should still be extremely safe, repeatable, have virtually nonexistent complications ("oh no, a stray cat or friendly elven child random encounter while we're kiting this slow zombie!") all of which should be trivially handled, and still not require an existent level-grinding infrastructure. It's in character if the 2,000-year-old NI level Wizard who trained the Elf Wizard suggested this training regimen and the Elf accepted "if you animate and kill them, they cannot be animated again" as a valid counterpoint to SEP (Standard Elven Procedure) of "burn the witch... and everything and everyone else who is dead, so they don't come back as undead". Cheese factor is... questionable. But since I didn't want to go Tainted Sorcerer with this build, didn't want to count on PF content for Blood Money, and was too lazy to think too hard about other ways to get free undead, and wanted to add it into this build in order to circumvent the need for communal leveling infrastructure, I circled back to, "well, what would the 2,000-year-old, NI level Wizard who trained them suggest?". And I still feel this is one of the least cheesy uses for a Mirror Mephit Familiar. The legality of animating the undead within the Elven Empire might pose a problem; perhaps the character can get a license for their specific use case? It's something to consider.

Quote:

---

Originally Posted by Maat Mons

On the other hand, the neighbor kid with lycanthrope parents was born with 26 hit dice and the kind of crazy numbers that come with it. Kind of disheartening when you contemplate how much work it would take you to get to where he started at.

---

OK, first things first: what kind of Lycanthrope is looking at 26 HD?

That said, yes, what you're born as also matters a whole lot. Just look at the true Dragons. :smallwink:

As to how good the Lycanthrope is... on the one hand, even with cross-class skills, that's, what, 14, 15 ranks in arbitrary skills? That's pretty respectable, especially on a chassis that's that hard to kill. On the other hand, it'll level as a fairly high level epic character, so approximately never. This is one character who's pretty legitimately stuck as a 1st level Commoner (or whatever class they took that 1st level in).

As for this human emotional response? "Disheartening"? More like "heartening"! It's great that our community has such a skilled family doing X, Y, and/or Z, so I can [choose one: apprentice under them, concentrate on other letters of the alphabet].

The kind of Lycanthrope is Were Legendary Tiger. Or Legendary Were Tiger. Or whatever. It’s based on Legendary Tiger, which appears in both Monster Manual II and Epic Level Handbook, is what I’m trying to say.

I was figuring on 30 ranks in a Craft or Profession skill, as long as it’s a class skill for whatever the character’s one class level is. Most classes have one or the other, if not both. The exceptions I’m aware of are Aristocrat, Knight, and Warrior. That makes 3 gp per day feasible, which is enough money that most people would probably be content. Still takes most of a week to craft a masterwork suit of full plate though. If you’re human, you can take Able Learner to avoid needing to spend double skill ranks for your animal hit dice, which don’t have Craft or Profession as a class skill. It isn’t necessary though, if you don’t mind being unskilled in other areas.

Yeah, the Were Legendary Tiger will never get more than one class level. However, they’ll start with stats that most mundane characters will never achieve with a lifetime of training. I think some might consider it a good deal. Not if you’re aiming to be a caster, of course. But if they’re aiming to be a tradesperson or guardsman, slam dunk.

Yeah, being born awesome is the key to success in this game. Though if there’s a tribe of Lycanthropes who are willing to induct outsiders into their numbers, an average person still has a fast path to being a badass. Actually, I think in a realistic setting, some Lycanthrope tribe with expansionist tendencies will naturally wind up composing huge swathes of the population in humanoid lands. The major hurdle is getting all the new members trained in Control Shape ASAP. Don’t want to have any accidents giving the tribe bad PR. If skill retraining is anything to go by, it’ll take more than 8 months to learn to reliably control the transformation. Gonna need some adamantine manacles for those first few full moons.

Ray Of Stupidity, as a scroll, wand, eternal wand, or however you can get it, would be a valuable tool.

1d4+1 Intelligence damage makes an elephant (CR 7, Touch AC 8, Intelligence 2) comatose. Kill at your leisure.

Quote:

---

Originally Posted by Daisy

When discussing the world general populace and why everyone isn't a wizard or level 10+, what no-one here seems to have considered yet is motivation.

---

There are far simpler reasons for that. If the simulation actually runs by 3.5's population guidelines, the answer is that these kinds of worlds are not what they model and not what they spit out.

From a deeper first-principles viewpoint, different classes have different training and infrastructural needs. For example, a human wizard will gain their 1st level in 2d6 years. A human rogue, barbarian or sorcerer gains their first level in 1d4 years. This is after reaching young adulthood (~15 years).

What does this mean? Consider a hundred would-be wizards, a hundred would-be barbarians, a hundred would-be rogues and a hundred would-be sorcerers begin training at the same time. By the time the first three wizards finish training and achieve 1st level, they will have to compete with 25 barbarians, 25 rogues and 25 sorcerers of the 1st level, in addition to whoever's left of the previous years 25 barbarians, 25 rogues and 25 sorcerers who had an entire year to get stronger.

Additionally, wizards require spellbooks, which requires a society that's already invented writing and has requires industry to make them, since a 1st level wizard is not particularly good at making those. A barbarian or a sorcerer has no such need.

Why not just be a sorcerer instead of a wizard, then? Because sorcerers are supposed to gain their powers through inheritance. Being able to pick sorcerer at all at character creation is a gameplay convenience, in a simulation, the capacity to be a sorcerer could actually be modeled by an inherited factor or random chance.

This leads us to ability scores. The game suggests using arrays, but these arrays are a compromise so a dungeon master doesn't have to roll every character's abilities by hand. A computer simulation can ditch arrays in favor of using 3d6 and 4d6b3 functions to stat a population, meaning someone who has the required inheritance to play a sorcerer, does not necessarily have ability scores high enough to be a viable caster at all.

So, you get a lot of people who can't be wizards worth a damn even if they wanted to. Similarly, most starting characters won't exist in favorable starting conditions, such as an existing elven empire, nor have 2000 year old mentor. Those are less strategy and more "pray the simulation allows for this". If it's set up closer to how games of this genre usually go, you'll start in or near unpopulated wilderness, with only small low-level communities nearby. This impacts strategies such as the toad farm too: where are you going to find your first toads and can you ensure your character has the necessary Knowledge (Nature) and Survival skills to get it off the ground? How confident are you that this won't piss off the Toad God and cause a gigantic Boss Toad to murder you and your children in turn? Or piss off other predators that also eat those toads? Remember, PCs and NPCs in 3.5. are highly symmetric, so in a simulation, you being able to advance in level can mean other beings are able to advance in level. If you can power level by killing toads, so can other things, which means going to a toad pond to get your farm started can mean going face-to-face with a whole host of large and huge snakes.

Quote:

---

Originally Posted by Vahnavoi

Similarly, most starting characters won't exist in favorable starting conditions, such as an existing elven empire, nor have 2000 year old mentor. Those are less strategy and more "pray the simulation allows for this".

---

Sure, this isn't the character's strategy for leveling, any more than it was the character's strategy to be born a Grey Elf, or to be born with "Wizard Stats", it's the Player's. Just like it could be Society's strategy to have set up toad farms, or equip newborn babies with Amulets of Intellect +6, to maximize their skillpoint development.

IIRC, the OP specified, "how would you...", so I was giving a player's strategy, along side "in character", and thus including the character's strategy in parallel.

Thanks for giving me the opportunity to explicitly tease those apart. Later, I may address "but what if the player gets no hand in strategy?" of a 3d6 random lost soul.

@Quertus: the OP also specified it's a Dwarf Fortress-style procedurally generated world. This means what's in it depends on what an algorithm spits out based on a random seed. You, as a player, are not guaranteed access to elven empires or 2,000 year-old-mentors unless the generation algorithm specifically includes a part saying "and this is where the elven empire, with 2,000 old wizards, goes".

We're both talking of player strategy; I'm simply talking from the perspective of a player dealing with largely unknown generation algorithm, which means a distinction can drawn between "this is a choice I can reasonably expect given basic character generation rules and hence always available for strategy" and "this is a favorable initial condition I have no control over and should not expect to be available for strategy before confirming it exists in the world".

Quote:

---

Originally Posted by Vahnavoi

@Quertus: the OP also specified it's a Dwarf Fortress-style procedurally generated world. This means what's in it depends on what an algorithm spits out based on a random seed. You, as a player, are not guaranteed access to elven empires or 2,000 year-old-mentors unless the generation algorithm specifically includes a part saying "and this is where the elven empire, with 2,000 old wizards, goes".

We're both talking of player strategy; I'm simply talking from the perspective of a player dealing with largely unknown generation algorithm, which means a distinction can drawn between "this is a choice I can reasonably expect given basic character generation rules and hence always available for strategy" and "this is a favorable initial condition I have no control over and should not expect to be available for strategy before confirming it exists in the world".

---

I’m not familiar with Dwarf Fortress; I am marginally familiar with D&D worlds, and very familiar with basic… “this is dumb” error checking.

D&D has concepts of towns, cities, and class and level population distribution within such. This is the type of thing I was expecting to be generated by the AI. If the AI generates, say, 20 nations, but puts their towns and cities randomly around the map, rather than, you know, generally grouped by nation? Then this fails my “this is dumb” check.

So, I can’t speak for how idiots might choose to build such an AI, but were it written by me? The arboreal elven empire my playthrough required was on the low end of Elven safe zones it could produce.

Well, no. If I wrote it, it’d be needlessly complex, and “elves as refugees”, “space elves only” and “no elves” would actually be possibilities (as would “what’s a forest?”, per Athas). But the zone I described was the minimum level of safety one could find in a “standard” world with elves in it - I didn’t have my sample characters benefit from Mythal level magic, symbionts, “Mother hen” deities, or any of the homebrew elves have gotten on various worlds.

As for the Wiz2k, it’s a reference to a running gag of mine on my opinions on limited information. Don’t get me wrong, I played older editions of D&D before Internet forum optimization, I know just how suboptimal *real* builds are. Heck, one of my catchphrases is “people are idiots”. Still, it’s not the case that *everyone* in the world is a complete ignoramus; Wiz2k is my way of saying, “push mandated idiocy too far in the name of ‘realism’, and I’ll combine Simulationist logic and background optimization to say, ‘sure, but I’ll run the guy who wasn’t raised by wolves and was trained by someone who wasn’t a complete moron, thanks’.”. Wiz2k spends those years of training wisely, bothering to mention or do things that are actually useful to their apprentice.

Oh, and it’s a reference to the idea that Wizards are actually *trained* - rather extensively, as your own calls to “starting age” demonstrate. Unlike plebeian Sorcerers, who legitimately should be ignoramus…es?

And we’re talking Elves - even without age magic, millennia is kinda normal.

As far as “what can we reasonably expect?”? I sadly lost a post on that earlier; short answer is, if I wrote the AI, and you chose “completely random settings” for the world and “no user input” for your stating character? You couldn’t even guarantee elves or humans existed in the world, or that the character wasn’t born mute or armless, so I suspect most strategies would be able to fail.

Lastly, I’m not interested in a “largely unknown generation algorithm”; thus, I’m trying to nail down (or simply arbitrarily choosing) some basic known facts, like “normal Elven empire generation”, “can everyone level through murder? (and is that known)”, or “Simulationist vs Minecraft logic for monster spawns?”.

As I said in the lost post, if you choose different assumptions, you get different results.

@Quertus: not knowing how Dwarf Fortress or other extensive procedural generation simulations go excuses you somewhat. You may find the following elaboration interesting:

A procedural generation algorithm starts with a random or pseudorandom seed - you can imagine this as a long string of die rolls used to check various world generation charts, as that is how you'd do it by hand. The Dungeon Master's Guide has its own tables for population demographics and settlement creation, but I'll leave those aside.

Instead, imagine the initial position created by the random seed as the beginning of a grand strategy game: a tribe of elves is placed is here, a tribe of orcs is placed there, so on and so forth. They all start with equal numbers, with no ready-made infrastructure present.

Here, like in many grand strategy games, a phenomenom arises: some races have great long-term potential, but no particularly great starting advantages. Other have great short-term advantages but level out in the long term. As noted by the argument about class starting ages, many powerful classes are in the former category. The same applies to longer-lived races, as they tend to take longer to achieve 1st level in any class. This creates an unstable equilibrium, where small changes in initial conditions can change who ends up on top.

For example, if a tribe of elves gets unlucky and is placed next to a tribe of orcs, you may never get large elven empires with 2,000 year old wizards. In competition for resources and living space, the orcs are incentivized to commit to an early rush strategy, such as, say, choosing barbarian class over others - this allows them to wipe out or enslave the elves before elves get off the ground at all. The downside of this is that choosing the early rush strategy and eradicating the elves may condemn the resulting orc society to illiteracy and low magic for extended period. So when the simulation reaches "modern day" when you, the player, step into the created world, you may find yourself having no elves, no wizards and really no large civilizations at all.

The same algorithm may be able to spit out hundreds of thousands of variations. Sometimes, the elves are placed far enough from quicker-maturing species to get off the ground, but majority of living space has already been occupied by other races that have had years to secure their foothold - the elves remain an isolated minority that never live up to full potential in either class advancement or age. Some times, the elves can only survive if they adopt a variant rush strategy, meaning that instead of very old magical elves, you end up with lot of young elf barbarians who never live very long because they keep dying in wars against other races. So on and so forth.

So, "everyone can level through murder" and "normal elven empire generation" are not in the same boat at all when in it comes to a simulation such as this. The former statement can be answered true or false just depending on whether basic rules of the game are applied universally. The latter depends on uncertainty about the initial conditions of the simulation. Again: you are not guaranteed an elven empire with 2,000 year old wizards, unless the algorithm specifically includes placement rule for an elven empire with 2,000 year old wizards. You cannot assume elven empires to be normal without knowing that.

It isn't simply about lack of user input. A generation algorithm can have all kinds of toggles from map size to what kind of terrain is favored, and still be unpredictable in its output. There a huge difference between "when I enter the world, I can pick my race, class etc. according to normal rules" and "when I press 'generate new world' button, I can be confident it will have this specific thing in it".

@Vahnavoi: ah, interesting. I was, indeed, thinking in terms of "starting at the moment generation is done", not "let the simulation run for an extended period after it is generated before entering". My playthrough was based on it being generated "at the end", and I was going to comment on one generated "at the beginning". But the "the simulation runs for a while before you enter" mechanic throws quite a wrench into things.

As you said, there's no guarantee Elves aren't extinct. The same is true of the OP's dogs, or even the question of whether wolves have been domesticated into dogs at all.

Under this scenario, I'll... echo myself from the last (and first!) time I heard of the idea of starting a simulation at 0: my money's on the were-rats. When you combine their high DR and low ECL with an ecological niche with little real competition, they're one of the ideal species to successfully develop real power. To which I'll add, Deities are another decent choice as far as "likely to still exist"; the problem is, with their 20 Outsider HD and Epic ECL, they should have a doozy of a time leveling.

So, in a Dwarf Fortress setting, I'd probably aim for Wererat or Deity, since they're the most likely to still be around.

A Level 1 Wererat Wizard is ECL 5. If I cheezed (heh!) that down from... uh... "natural" to "infected", they're ECL 4. Or if I went full cheese of Quasilycanthrope, they're ECL 1. I'd say that, after Wererats' all but inevitable initial success, the ensuing Quasilycanthrope communities would be very well positioned. If, somehow, 3e's "free literacy" didn't happen, or the Wererats were otherwise unable to develop the industry necessary for Wizardry, it's easy enough to go Cleric instead - which I expect would be a rather popular choice, given the diseases many traditional Wererat haunts are rife with.

The Wererats cannot go the same route as my Elf Wizard playthrough, because even with Improved Familiar, they cannot get a Mirror Mephit before they level out of earning XP from trapped animals. And, IIRC, Lycanthropes don't get the Grey Elf bonus to Intelligence, but do get a bonus to Wisdom. All in all, it sounds like Cleric is the way to go here. (We'll just ignore the Quasilycanthrope, and assume they manage to go a Wizard route not entirely unlike the Elf).

With Rat Empathy, it's really easy for a Wererat community to scout and keep appraised of nearby threats, and put them in the mind to think in terms of animal friends. Domesticating wolves into dogs seems unlikely (unless they stole the end product from other races), and cats seem right out, but Giant Rats (obviously), plus Elephants and Rhinos (large, distracting, trampling creatures) seem the kinds of things Wererats would want on their side. Other good choices include Giant Spiders and Dragons. Maybe Hydras.

Anyway, just from killing stone-age Goblins, or whatever "not-threat" we've allowed to live nearby to let our young cut their teeth on (toad ponds, whatever), we'd hit Wererat Cleric 5 (if breed) or Cleric 6 (if infected). That's high enough level to cast Animate Dead. And Cure Disease. Between Casting Services for Animate Dead for our Community's defenses, or Cure Disease for the nearby settlements for diseases we totally didn't cause, the Wererat Cleric 5 could in theory earn money quite quickly; in practice, we're looking at mediocre Crafting checks, or the ridiculously OP "Craft Magic Items", 1k gp per day solution to financial problems. And it only costs us 500 gp, and 200 gp worth of XP per day! :smallsigh:

In other words, while the Infected Wererat could craft items for 500 gold profit a day while taking out his frustrations on squishy stress balls in toad form, the *real* Wererat has to actually kill "real" threats in order to keep replenishing their XP from crafting magic items for sale. Thanks to their DR, even things like Brown Bears can only hurt them on a crit, so it's not too hard to replenish those XP (and Diplomacy is a Class Skill for Clerics, so it's fairly easy to rely on others / utilize the rat scout network to find suitable prey).

The Wererat isn't as nigh-immortal as the Elf, but even after a season worth of crafting (possibly spread out over a year, if they're also gaining XP and setting up plots ingratiating themselves into various communities), that's more than enough $$$ to give ourselves level-appropriate items. Maybe even including a War Rhino mount, or a Dragon Egg to hatch and train.

As suboptimal as it likely is, we're probably looking at Animal and Trickery for our Spheres (if those are even actual spheres in 3e), although Luck and Disease sound like valid options for a hypothetical "god of wererats" to have, as well. Or we could have rolled up a really weird world, where the only Dragons left are the ones in captivity, that the egg-stealing Wererats train, giving their deity the Dragon sphere. Shrug.

Taking a Wererat Cleric 5 out into the world, though? I'm not liking those odds near as well as the Elven Wizard. We'd definitely want some speed boosters, and a Potion of Invisibility or two (Trickery probably gives us that spell, doesn't it?) to escape from threats before even considering heading out on our own.

Or... with the "Dwarf Fortress" premise... would anyone really do that? Or would the Community just some day up and ravage an area? I'm not quite clear on the "world-level roleplaying" that the AI has forced upon the world.

Level grinding a Deity... that's really hard. I'll have to give it some thought.

I'll also think on whether there's any other races I'd believe were highly likely to still exist after a realistic "Dwarf Fortress" run from Re:Zero.

Quote:

---

Originally Posted by Quertus

Under this scenario, I'll... echo myself from the last (and first!) time I heard of the idea of starting a simulation at 0: my money's on the were-rats. When you combine their high DR and low ECL with an ecological niche with little real competition, they're one of the ideal species to successfully develop real power.

---

Lycanthropolis is inevitable!

Quote:

---

Originally Posted by Quertus

Deities are another decent choice as far as "likely to still exist"; the problem is, with their 20 Outsider HD and Epic ECL, they should have a doozy of a time leveling.

---

Every deity can cast all spells for domains they offer as spell-like abilities at-will. Make sure you offer domains that include Wish, Miracle, and Shapechange. Having those three SLA's at-will makes being stuck at low levels much less onerous.

It doesn't make sense to talk about deities in context of player strategy - whether they exist or not, they are not typically player characters so not a valid pick when play starts.

As for infectious templates, Dwarf Fortress proper has vampirism. Vampires can infiltrate colonies and are often much sturdier and more powerful than normal. The problem there is that vampires require mortal blood - not everyone can be turned into a vampire or a community collapses. So a player has incentive to either restrict spread of the template or come up with an elaborate trap to kill the vampires without losing too many resources.

In D&D, not all were-templates have such obvious restriction, and if they happen in a primitive community, nothing much can be done to limit spread. Some kind of prey-predator-relationship can still limit their numbers. Damage reduction is a significant boon against most low-level threats, but if I recall correctly, does not apply against other werecreatures.

When thinking of infectious templates, one has to consider level draining undead. In Dwarf Fortress, a localized zombie apocalypse (due to necromantic mist of whatever it is) is a real threat. Similarly, a procedurally generated D&D landscape may end up with one or more no-go zones in the form of undead wastelands.

So, at least the way I would write it, there won’t be any Vampires until you get a Cleric capable of casting Create Greater Undead, or a Deity who earns a domain granting them that spell (directly, or indirectly through Miracle) who would also actually create a Vampire.

Further, I wasn’t actually going the “infectious” route for the lycanthropes - I was only assuming natural, “bred” wererats. They do get even more powerful (and easier to level) when you let them multiply via infection.

Yes, the wererats would die like rats to most other lycanthropes, who can indeed ignore their DR. For them, it’s a question of getting the levels - or, better yet, peaceful coexistence, if the AI allows such things.

There’s also the question of what’s available at Time 0. That is, I can see implementing this “by raw”, where everyone can level as everything, craft everything, etc, but that feels to violate the “in character” principle of the OP. Alternately, everyone can appear naked, and have to go through a Civilization-style tech tree to build the tools to unlock the tools to unlock the classes. In that scenario (which was what was described at the other time I heard of “start at 0”), wererats have those huge defensive advantages while still being fairly easy to level.

In addition to deities, I think Sprites and Beholders might be worth looking at. I’ll ponder their grinding.

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.

The idea is to use tactics that a character living in that world (a character, not a player) would be able to conceptualize and enact.

For example, a character could plausibly use the Craft or Profession skill for many months, or even some years, to earn money and equip himself better than a standard level 1 character. He could think about raising and training war dogs (which are fairly strong at low levels) to help him in hunting bandits, goblins, or wild boars.

---

Quote:

---

Originally Posted by Vahnavoi

A procedural generation algorithm starts with a random or pseudorandom seed - you can imagine this as a long string of die rolls used to check various world generation charts, as that is how you'd do it by hand. The Dungeon Master's Guide has its own tables for population demographics and settlement creation, but I'll leave those aside.

Instead, imagine the initial position created by the random seed as the beginning of a grand strategy game: a tribe of elves is placed is here, a tribe of orcs is placed there, so on and so forth. They all start with equal numbers, with no ready-made infrastructure present.

---

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”.

Also… I personally prefer the starting population to be based on CR (proportional to the square of CR, per Deity of Creation logic) rather than equal. It may seem to shoot my wererats in the paw, but I think an equal size “tribe of Dragons” (or “tribe of deities”! :smalleek:) might be a little too OP.

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.

---

In terms of the final printed timeline? Yes. In terms of game turns processed? Not really, probably not. A comparison can be made to Civilization 6: the nominal Ancient era starting point is 4000 BC or something, with every technology having to be researched from scratch. This time is not evenly divided among the standard 500 turns: the game will typically reached middle ages (500 AD to 300 AD) by turn 120.

Similarly, a world generation algorithm can cheat a lot to speed past ancient history. I don't remember how Dwarf Fortress in particular does it, other than it can print out highly intricate accounts of old civilizations, but it probably takes for granted presence of several basic technologies. Nonetheless, if you actually want a generation algorithm to spit out thousands of years of history, it will take a long time - in real time. This another obstacle for expectations of supremely ancient things: there obviously won't be any 2,000 year old elves if world generation was stopped after 200.

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.

So, it sounds like the simulation will be running for a long time, that the game will start a very long time after world generation. Meaning that even if Adam and Eve's contemporaries were all level 0, modern day individuals won't have that problem.

Thus, IMO the best answer to how to level-grind in a Dwarf Fortress style scenario based off 3e rules is to have several high-level individuals chaperone one or more noobs through "fairly safe" territory (ie, places known for those past few thousand years to produce "survivable" random encounters), letting them level by earning "their share of the XP" from any kills. And doing so in zones that are "adjacent" to "safe zones", just in case of trouble.

Note that relying on higher level characters instead of trained dogs doesn't require the society to have domesticated wolves into dogs, or even to have encountered wolves. "The math just works", regardless of any unknowns in world generation. :smallwink:

Further, IMO, the society that will prosper most in this scenario is the society that not only has such "murder mentorship methods", but also equips their citizens in defiance of WBL. Whether that's war horses, Amulets of Intelligence, Boots of Run Away Real Good, Mythals that extend into the "danger"/XP zones, or what have you, these advantages will hugely impact the ability to safely level. Call it the equivalent of a "college loan". Only, even a group of 10 leveling off random encounters of CR 1 animals won't take 4 years to graduate to level 9. Probably more like 4 months.

Also, the society that "terraforms", that a) literally changes the landscape to give themselves advantages; b) exterminates (and stocks) wildlife / threats in order to maximize advancement and minimize risk will have the advantage over societies that are attempting to grind while "leaving the dice where they lay"; ie, who just take what is given without being smart about it. Which, I admit, requires a little more advanced stuff than I'd expect most implementations of this as a program to include (but, honestly, would anyone write the code such that the random encounters in town are the same as out in the wilderness? I'd hope not. So "changing the random encounter table" should happen in any good implementation anyway.).

-----

Also, by RAW, am I remembering correctly that "naked Adam and Eve" need to make Survival DC 15 to get food? Which, if there's 20 people, means they get food for... 15&16: 1, 17&18:2, 19&20:3, or 12 out of 20 people total. If all of them spend 8 hours foraging. So, am I missing anything, or do most 0-level "naked Adam & Eve" races just die off in the first couple days? :smallconfused:

Quote:

---

Originally Posted by Vahnavoi

In terms of the final printed timeline? Yes. In terms of game turns processed? Not really, probably not. A comparison can be made to Civilization 6: the nominal Ancient era starting point is 4000 BC or something, with every technology having to be researched from scratch. This time is not evenly divided among the standard 500 turns: the game will typically reached middle ages (500 AD to 300 AD) by turn 120.

Similarly, a world generation algorithm can cheat a lot to speed past ancient history. I don't remember how Dwarf Fortress in particular does it, other than it can print out highly intricate accounts of old civilizations, but it probably takes for granted presence of several basic technologies. Nonetheless, if you actually want a generation algorithm to spit out thousands of years of history, it will take a long time - in real time. This another obstacle for expectations of supremely ancient things: there obviously won't be any 2,000 year old elves if world generation was stopped after 200.

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.

---

I feel like, if a race is suited to a "Zergling Rush", they might well live highly nomadic lifestyles, meaning that they could know and encounter large portions of the map quickly. Such a lifestyle would, however, result in them having minimal infrastructure, minimal tech tree growth.

Point being, with such large turn sizes, in the first turn, they may well have encountered (and possibly exterminated) several races.

Quote:

---

Originally Posted by Samael Morgenst

Suppose you had to make your character grow in experience, levels and - possibly - wealth and power, while minimizing risks as much as possible.

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.

For example, a character could plausibly use the Craft or Profession skill for many months, or even some years, to earn money and equip himself better than a standard level 1 character. He could think about raising and training war dogs (which are fairly strong at low levels) to help him in hunting bandits, goblins, or wild boars.

---

Quote:

---

Originally Posted by Vahnavoi

You are all too concerned with whether a character knows specific game details and not nearly concerned enough about all the things that are genuinely unknown even to you, such as enemy placement and terrain.

---

Even if "starting from zero", if "society" has been around for the thousands of years necessary to turn wolves into dogs and get a D&D economy from "naked Adam and Eve", I feel it'd be horrifically unrealistic for the civilization not to know the area, and its denizens. So it's not "genuine unknowns", it's... "unknown knowns", I suppose. :smallamused:

And, again, that's for naive morons who aren't "terraforming" the area to suit their needs. So, with a smart race that has killed off all local predators they don't want around, it's less "unknown knowns" and "predicted knowns". How good that prediction is depends on how different the programmer made each individual forest, mountain, and plains's starting random encounter tables.

Quote:

---

Originally Posted by Quertus

Also, by RAW, am I remembering correctly that "naked Adam and Eve" need to make Survival DC 15 to get food? Which, if there's 20 people, means they get food for... 15&16: 1, 17&18:2, 19&20:3, or 12 out of 20 people total. If all of them spend 8 hours foraging. So, am I missing anything, or do most 0-level "naked Adam & Eve" races just die off in the first couple days? :smallconfused:

---

I just checked, it's only DC 10. So if you put in 4 ranks, take 10 on the check, and don't have a Wisdom penalty you can provide for yourself and two other people.

Quote:

---

Originally Posted by Maat Mons

I just checked, it's only DC 10. So if you put in 4 ranks, take 10 on the check, and don't have a Wisdom penalty you can provide for yourself and two other people.

---

DC 10? Oh, that's much easier. If Classes are a thing at T=0, only 1/3 of your population has to take classes with Survival as a Class Skill and spend their days foraging. OTOH, if classes are unlocked via some strange civilization skill and tech advancement tree, then they've just got their Humanoid HD... is Survival a Class Skill for a Humanoid HD? If not, then they can take at most 2.0 ranks, and 1/2 your population needs to be dedicated to foraging initially.

For your man-sized humanoids. Other creatures will have different results.

-----

So, were I really simulating a 3e world from T=0, IMO all of the Humanoid races "play the same", more or less. The races that are particularly interesting / have different tempos include

• Wererats - Good defense vs many things, great at T=0 (especially if nobody gets classes until they are unlocked) while still being able to earn XP. But they level with more difficulty, and most hit a much lower level cap than Humanoids.
• Most Lycanthropes - great early game; can't level easily or at all. Best strategy is to partner with other race(s), have them become Lycanthropes when they hit the end of their leveling curve.
• Dragons - Powerful, flying, ranged, magical, automatically improves with age. About the only way to level is to hunt really high CR targets (while at the right age). Best (or funniest) target/class combination? Hunting Deities, to level in Ur-Priest. Otherwise, Dragons work best when they don't get into fights, as those offer them nothing.
• Beholders - Really powerful from the start; about the only class worth taking is Beholder Mage, but XP is hard, and their power arguably drops with that 1st level and losing their Antimagic.
• Sprite.Pixie - Flight and Invisibility mean they can kite (or avoid losing scenarios) like nobody's business; small size means they need little to eat (Survival gives additional "man-sized portions" with no reference to the size of the being making the roll), and great stats for Rogues and (arcane) Casters. Unlike Wererats, these seem able to reach higher level IME, even if they die a little more often.
• Petal - small, easily fed fliers with even better stats than Pixies, with less LA and therefore easier to level to boot! Humanoids can't fit in their fortresses, and they lose most of their mobility advantages in theirs. And having almost no carrying capacity definitely impacts their options.
• Maugg (sp?) - reproduction based on getting enough money to Craft more of themselves. And, if in a run where Classes need to be unlocked, they start with Sorcerer. EDIT: Also, they don't eat, so no need to have anybody foraging. They also don't sleep, so (for example) their whole society could escape an Undead Hoard.
• Anthropomorphic Bat - Small size & Wisdom Bonus means minimal foraging, so most can be doing other things; flight for defensive evasion or kiting; easy to level. About the only real downside is not great stats otherwise, especially at the start.
• Deities - If on the world, they're just dead, so that's not interesting. If "projecting", they care about number of followers (as that determines how long they're locked out from "projecting" if they "die"). If (say) there's a Deity of each race, they care about their race's strategy, as it affects what Spheres they get access to. Completely different priorities and play than any other race.

So, were I to write something to try to simulate a new, artificial 3e world, I'd include at most 2 creatures of each of these playstyles (ie, 1-2 powerful Lycanthropes, 1-2 Dragon races), plus a bunch of Humaniods and other "playable" races (Treant, Troll, Ogre, whatever), each with their own Racial Deity who starts with no Spheres. The choices the races make determine the Deity's spheres (so the Dino-riding Elven Barbarians might find that their Deity gets Strength and Lizard as their spheres or something).

"Unplayable" races (those that exist) would exist as "random encounters" for terrain. If enough are encountered and killed in a given amount of time (varies by creature type), that element is removed from the random encounter table; over time, nearby populations (ie, random encounters from adjacent areas) may replace the "no encounter" entry that was left behind by their local extinction, favoring favorable terrain and a few other factors. This would allow races to custom tailor their environment. Note that "us" becomes a Random Encounter; ie, in a Human City, potentially all of the Random Encounters are "Human(s)". Animals would definitely be part of the Random Encounters. Anything that is a Random Encounter can also be explicitly "Hunted".

In a slightly more sophisticated version, "random encounters" might hunt one another, and, when a population in an area grows too great (creatures with no predators - Deer in an area with no Wolves, for example), their Random Encounter Rate increases, and eventually spills over into neighboring areas (even if their Random Encounters haven't been wiped out yet). Meaning that patrolling a wider area is important for keeping Undesirables out (the Elven Rangers with Chosen Enemy: Cats know what I'm talking about).

Some races (like any half-, including half-elves and half-ogres and half-dragon everything, but also the aforementioned Vampires, infected Lycanthropes, and Quasilycanthropes) are emergent - they don't exist at T=0, but can be created through play.

Anyway, that's how I might build such a thing, which impacts how I think about the question of "how would one (plan to) safely level-grind a character in a world with random programmatic AI advancement of a race, under 3e rules". And I doubt there's many better answers than, "high-level friends guide one or more low-level noobs through minimal threats in a carefully-crafted environment" "in a society that Empowers its Youth with Items / Mythals / etc".

Now, how many societies can get to the point where they can afford to send their children to college? That can depend on a lot of variables, but I think those that do, win.

even if it were dc 15
a dedicated hunter even with the standard array
human first feat skill focus survival
second feat self sufficient
4 ranks survival
wis 12

take 10 to get 20 every time and the base dc assumes you are traveling it would not be unreasonable for someone focused entirely on hunting to either get a bonus or be allowed multiple checks.

Quote:

---

Originally Posted by Quertus

OTOH, if classes are unlocked via some strange civilization skill and tech advancement tree, then they've just got their Humanoid HD... is Survival a Class Skill for a Humanoid HD? If not, then they can take at most 2.0 ranks, and 1/2 your population needs to be dedicated to foraging initially.

---

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.

Quote:

---

Originally Posted by Quertus

I personally prefer the starting population to be based on CR (proportional to the square of CR, per Deity of Creation logic) rather than equal. It may seem to shoot my wererats in the paw, but I think an equal size “tribe of Dragons” (or “tribe of deities”! :smalleek:) might be a little too OP.

---

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.

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

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.

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.

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.

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).