The sandbox railroad part 2 (plus ULTRA-LINEAR encounters)

Another take on this post.

---

Important caveat before we begin 

"Railroad" has a negative connotation for some readers. If you dislike my use of this term, I offer an alternative by the end of this post.

I love random encounters and I think they are an important part of the game. No matter if I roll them at the table or half an hour before that, or if I already rolled 100 examples for each terrain. I've tried all of them and each has its pros and cons.

I don't think you can understand this post if you assume I'm criticizing the way you play, which is not my point. I'm just describing how random encounters might work in theory, not how you use them in practice. 

I'm not saying there is a problem, necessarily. If you seems random encounters rolls as simple suggestions, this is definitely not what I'm discussing here. I'm assuming you're using a table and sticking to the rules and the results you rolled].

---

I called "random wilderness encounters" a railroad because if the PCs are in the middle of a forest and the next encounter is an ogre, there is no choice but to see the ogre - no matter if they go North, South, etc.

[Assume the GM has already rolled the next encounter].

Usually, it doesn't even matter if they stop and rest, because then the next random encounter (ogre) will happen as they make camp.

Think of a dungeon where you have 4 doors and the module says "the ogre is behind whatever door the PCs choose first; if PCs decide to rest here, the ogre will enter the room they are in".

It would be obvious to everyone this is a railroad/"quantum ogre" situation.

Of course, while the encounter is presented in a linear fashion (you WILL see the ogre), they way you choose to deal with it is not linear.

You could even have the possibility of AVOIDING the encounter altogether.

But the same is true of the dungeon described above.

And, no matter what you do, you face the next encounter.

[Again, assume the GM has already rolled the next encounter: 2d6 goblins].

The ORDER of encounters remains linear - or even "ULTRA-LINEAR".

In a dungeon with 3 linear rooms (say, ogre-goblin-skeleton), you can avoid the next room by turning back or simply not opening the next door.

In the middle of the forest, turning back or stopping usually leads to next encounter!

This is not necessarily a PROBLEM; this is how random wilderness encounters work.

To add CHOICE to the next encounter, you'd need PCs to have some knowledge in advance.

For example, they'd have to be able to look for tracks or see foes at a distance.

This is not impossible to do, but requires you to MARK some hexes. 

For example, if they avoided the ogre, now they know that they are likely to meet him again if they enter that same hex (instead of just rolling the next encounter).

Filling all hexes is tiresome. 

In practice, you can use your memory; if the PCs avoided an encounter yesterday, going back might trigger the encounter.

If they go back there after a week the ogre is probably gone and forgotten.

This entire thread is descriptive and not prescriptive.

I'm not complaining or giving advice, other than, maybe: if you want to avoid railroads, give the players some options BEFORE the encounter begins.

But the PC's entered the forest and put themselves in this situation!

I agree. Although they might not have an option (if the starting point is surrounded by forests for example).

But I don't roll my random encounters in advance!

I don't think rolling in advance makes any difference here.

Because the roll is not affected by the PC's decisions.

If I rolled 39 before the game begins or if I roll 39 when the PCs say "we go North", the result is the same, not any more or less organic IMO.

I still don't see why this is railroading, or this is not what I call railroading, because the PCs can talk to the ogre etc.

I think the best way to understand my point is comparing the wilderness encounter to the dungeon with four doors, described above. 

Or think of it this way: "no matter what the PCs do or where they go, they'll find an ogre tomorrow because I rolled it".

This is not a problem, necessarily: most people would be fine if I said  "no matter what the PCs do or where they go, they see rain tomorrow because I rolled it", or "they'll see the ghost that is haunting them".

But if you don't want to call it "railroading" maybe "ultra-linear" would be a better description.

Additional reading:

https://methodsetmadness.blogspot.com/2022/05/railroading-in-space-and-time-and.html
https://thealexandrian.net/wordpress/36900/roleplaying-games/the-railroading-manifesto

https://thealexandrian.net/wordpress/36982/roleplaying-games/railroading-manifesto-addendum-random-railroads

NOTE: The Alexandrian has a good definition of railroading in the second post above. Random encounters do not seem to fit, at first. HOWEVER the last post above indicates that the CAN be railroading, in the exact same way I discussed today:

The core distinction here is whether or not the players are making a meaningful choice. In this hypothetical hexcrawl scenario, the choice of direction has been rendered meaningless (since you’ll have the same experience regardless of which direction you go). [...] This taught me a really important lesson as a GM: In order for an exploration scenario to work, there has to actually be something to explore. If all choices are equally likely to get you to your goal (because your discoveries are being randomly generated or because the GM has predetermined their sequence), then your choices become meaningless. And meaningless choices are boring and frustrating.

Yam-Shaped Campaigns

This is not my idea; the term may have been created by Sly Flourish but I couldn't find the exact source. In any case, I think it is a concept worth discussing and spreading.

Many people discuss "railroad" campaigns and "sandbox" campaigns as if they are opposite choices. I don't think this is entirely accurate. In any case, there is a third alternative that is probably my favorite: the "Yam-Shaped Campaign."

A "Yam-Shaped Campaign" is "narrow at the beginning and end but wide in the middle". In other words, it has a clear beginning (possibly with clear goals) and one (or preferably, a few) explicit endings. However,  HOW and IF you'll get there is up to the PCs.

In 5e D&D, Tomb of Annihilation (ToA) and Curse of Strahd (CoS) are good examples. In B/X, my favorite is probably B10 Night's Dark Terror.

Instead of adding a picture of a yam to this post, let's try this.

The PCs start the game at point A, in an exact place and time. Although this might seem to be the case for all campaigns, it is not. For example, if you're running a random wilderness, the PCs might be in a world of quantum randomness.

Points B and C are two possible "endings": let's call them the good ending and the bad ending.

In this kind of campaign, the PCs will usually have, at the very least, a vague idea of B or C. Conversely, they might have a vague idea (or a few options) on where to go NEXT and they eventually find out about B or C as the campaign progresses.

The black lines represent the multiple paths the PCs can take. The grey lines represent possible paths that lead the PCs away from the proposed ending.

Let's try some concrete examples.

In Curse of Strahd, the PCs begin near the village of Barovia. They know one possible "bad ending": they get stuck in there forever. This is unlikely to actually happen, because both the PCs and the villain are unlikely to sit and wait. So they have a clear goal instead: get out of there.

With that in mind, they can wander around looking for an exit, and they'll eventually realize that the only way to do that is to face Strahd. This is "point B". When they face Strahd, they can defeat him, join him, replace him, etc. 

The book even contains a page describing what may happen next, but this is beyond the scope of the campaign. Having this is very common and useful, because you might want to continue playing after Strahd has been "resolved".

In the diagram above, I used "time" as an axis. This can be some type of "STRICT TIME RECORD" or something more abstract. Strahd has no clear timeline, but the confrontation gets closer and closer. Tomb of Annihilation has a strict time limit: if enough time is passed, the PCs simply fail and everyone is doomed (IIRC).

Both approaches are possible, but I think SOME time pressure is needed in this type of campaign or the PCs might wander around aimlessly.

In any case, there is at least two possible endings; probably more. I think of point B as some kind of "final showdown" that can result in victory or tragedy (C).

Even if the "endings" are clear, the GM might have to choose what to do if they are skipped. What happens if there is a TPK, for example (all PCs die?). Do they start again with new PCs? Turn back time? Find a new campaign? Or maybe advance the time line a few years, and create PCs that must serve/oppose the (now victorious) antagonist?

What if the PCs leave the continent or ignore the villain? Same thing. Most decent campaigns will at least give you an idea on what to do next, but you're mostly on your own.

When I wrote my own adventure, I didn't know this "Yam-Shaped Campaign" terminology. But I did include an "aftermath" section that I'll add here as an example, as it includes:

- Two possible "boss" endings.

- What happens if the adventurers just fail or leave.

- Ideas for more adventurers if they succeed.

These three possibilities are the bare minimum, I think, to include in a yam-shaped campaign. Here is how it looks:

Aftermath

Here is what happens after the end (or in the middle, in some cases) of the adventure.

If Malavor is slain, the bee-people will immediately destroy the remaining demons. The Queen (now free from mental slavery) will telepathically ask the adventures for death, but she will resurrect from its own carcass (as an ordinary bee) in less than one minute and fly away. The fortress will collapse within 1d6+6 days. The demons in the underground will be buried alive. Some might survive.

If the Queen dies, bee-people will disperse immediately. It will take Malavor 3d6 days to summon another avatar or come up with a new use for his fortress. His success is not guaranteed. He might try a different plan. In any case, the hive will still be a menace as long as Malavor lives.

If both the Queen and Malavor are still alive, the hive expands. In 2d4 weeks, the number of demons and bee-soldiers is doubled, and the hive’s defenses are reinforced. In another 1d6 weeks, Malavor manages to mutate himself into a bee-demon, half-insane, but with full control of the bee-people. The bloated and sick avatar dies after a while, but this no longer affects the bee-soldiers, that can now be cloned in the underground. Three months after the characters left, Malavor unleashes his army against the nearest village.

The underground is a different matter. The underground is currently running its own schemes. It supports Malavor but only because he lets them do their own stuff. If Malavor leaves the hive, they will quickly take it. They will seal all doors to the outside, leaving a couple of secret passages. They might demolish the towers to avoid getting attention from the outside, and cover the whole fortress in dirt to transform it into a mound. They are digging their own underground tunnels, leading to somewhere miles away… or miles below. There are more demons in the Abyss that spawned Malavor and the biomancers.

In addition, unless Zothaq and more than half the biomancers working underground are killed, the underground keeps expanding until it becomes some kind of megadungeon, full of demons and hybrid life forms. The forest around the ruins of the hive becomes progressively weirder, with mutant beasts prowling around. Fortunately, they have no plans of conquering neighboring towns right away (the idea here is that the characters find this out and come back after a while, hopefully when they are a bit stronger).

In conclusion, I have run multiple campaigns. Some could be described as "Railroads," others as "sandboxes," but lately, I have realized that Yam-Shaped Campaigns are my favorites. 

In my sandbox campaigns, I have noticed that the PCs (Player Characters) often get lost easily and have no clear direction. As a DM, I then have to decide abruptly where the game should end, or I let it fizzle out. Conversely, railroad campaigns feel terrible: it is like the PCs have no choice and I have the sole responsibility of taking the game to the end.

Yam-Shaped Campaigns give me the best of both worlds: the PCs are free to wander around and surprise me, but I always have an idea about what is going to happen next and we can even have an epic  - or tragic! - ending before moving on.

Contains affiliate links. By purchasing stuff through affiliate links you're helping to support this blog.

Unused parts - Reduce, Reuse, Recycle?

Lately, I've been reconsidering the amount of work I put in my campaigns.

As an example, let's talk dungeons. No matter if you create your own, use a published module or random tables, I expect you to have a minimum knowledge of the thing you're running.

The problem is an interesting dungeon has lots of content - and, while the players will probably never see half of it, the GM has to at least give it some consideration.

It is not just the ignored rooms and secret doors that were never found - it is also the NPC with relevant information that the PCs immediately attack, the writings in languages no one speaks*, random encounters/events that do not happen, maybe even the villain monologue that gets interrupted, etc. 

[*For example, I ran Tamoachan the other day and it has SEVERAL writings and dialogues in "Olman" or whatever. Nowhere in the adventure summary I found any warning for the GM about that. I might have missed it. I used both the 1e and the 5e version].

This is what I call this "unused parts".

It seems that the DM will always have a lot more work than the players - the fun part, for me, is that during the actual game (my favorite part) I can sit back and see what the PCs do, as a neutral referee, without having to author anything.

While I love the sandbox format, I think it encourages "unused parts". A sandbox needs huge unexplored parts, and the freedom to allow PCs to go where they want. I takes the GM a lot of work to create or get acquainted with an entire setting.

Here is an example from X I've seen today (from @ericbabe3):

The PCs decide to follow the main corridor, found the treasure behind a secret door, and left. All the work the GM had choosing this module and reading it beforehand was wasted. 

I definitely LIKE the fact that the PCs can do that.

But how to deal with all those unused parts?

Let's see.

- Force them on the PCs (quantum ogre). This is just putting each encounter/trap/room you want to have in your game in front of the PCs, no matter which door they choose. I dislike this - it makes PC's choices meaningless and ruins any sense of surprise for me, the GM, as I know exactly what will happen.

- Randomize everything (random ogre). Roll 1d6 to put a random trap in front of the PCs, no matter which door they choose. Like the example above, it makes PC's choices mostly meaningless, and you still have a few traps/encounters that will simply not happen.

- Talk to the players beforehand. In my sandbox, the PCs can go anywhere, but I need at least a vague idea on where they are going. I've scattered several dungeons around the map, but I can't read - let alone remember - each one. So I just ask before the next session about where the PCs intend to go.

This is harder to do with dungeons. Depending on the size of your dungeon and your sessions, maybe you can read only a few levels beforehand.

- Minimum prep. Just do the bare minimum:  give a quick glance to the module and read the entirety of the room IF the PCs get there. This is what I've been doing, but it has led to some confusion on occasion, when the module is not clear enough and I'd need more time to understand. For example, I was once running a module that gave the PCs an "Amulet of the Phoenix" that was simply not described anywhere. I had to improvise in a case of life or death (that deserves a post in the future). Later I found out there was a "Talisman of the Phoenix" or something in the magic item list.

Another example: There are huge unmapped areas in my sandbox that are simply called "giant territory", "goblin territory", and "undead territory". No dungeons in most of these areas - I'll detail them when the PCs get closer, but they at least have a vague idea on what to expect and can make meaningful choices.

- Zooming in an out. Another way to think of this is "zooming in and out". Ideally, modules should be prepared in such a manner. You should be able to get the clear picture from a sentence or two, preferentially in bold, and the have other paragraphs describing what may happen if the players do X or Y.

Any important secrets or details must also be emphasized/ marked to make sure the GM will not miss them.

Likewise, for a overland map, with hexes or otherwise, you could have brief descriptions of each are and then a more detailed description - with a corresponding map - later on.

I know there are several attempts to do that in OSR games. I never read a module that really stands out in this manner, but I'm always searching! Have you? 

- Reduce, Reuse, Recycle. This is especially important if you're creating your own stuff.

Reduce: make it terse, keep prep to a minimum. Don't fill all the details, just create whatever you want to inspire you. "This room has two sleeping ogres, Torg and Gurps, not aggressive but starting to get hungry".  If the PCs kill them while they sleep, you haven't wasted much details.

Reuse: have recurring NPCs and factions, and let they repeat their tactics/traps/behavior sometimes. Let the PCs go back and forth in your wilderness maps. PCs usually need a home base that will be more detailed than other cities. 

Some times, the PCs will enter the same dungeon multiple times (although it is useful to consider how the dungeon changes as they come and go). In my latest "haunted castle" adventure, the undead simply rose again every night.

Recycle. Traps/rooms/encounters that are cool and unused can be kept for later. This is not a "quantum ogre" thing. The PCs must have the option of ignoring it again. You have to deliberately change where the thing is placed, and you have to do it carefully -  the new place must make a last some sense.

Let me know in the comments if you have any other methods to deal with unused parts!

30-mile hexes

I've considered several hex sizes before. Today I will think about 30-mile hexes.

Look at this map (source):

The hexes are about 31-miles each (31 miles across, with sides about 17.32 miles).

Good enough so they can fit all major cities and the shape of Italy is still recognizable.

You'd need to add several random villages and things (ruins, caves, castles, lairs, etc.) to each "empty" hex.

The whole country could be described in 150 hexes or so - sounds reasonable to run an entire campaign., maybe giving at least one page to each hex. And EVERY hex could have one or two interesting features.

At this size, I think you could just give the map to your players. If your setting is anything like medieval Italy, it wouldn't be hard to know the general direction of Verona or Genova. 

Rome fits in an hex, but does not FILL it.

You lose some interesting detail; the map doesn't show you Venezia (Venice) as an island, nor the Lago di Garda (the largest lake in Italy - with an area of 143 square miles, it would fill one or two 6-mile hexes).

According to google maps, you can cross the entire country (from Bova Marina to to Milano, for example) on foot in 302 hours (1310 km, 814 miles).

Apparently, "Google Maps uses average walking speed of 5km/h to calculate walking times" - which is 3.1 MPH.

This is NOT a straight line, and while there are many more roads in modern days, ancient Rome had several, as you can see in this interesting site:

https://orbis.stanford.edu/

OSE says "the number of miles a character can travel in a day is determined by dividing their base movement rate by five. For example, a character whose base movement rate is 120’ could travel up to 24 miles in a day."

Of course, if you travel 24 miles a day, you might as well use 24-mile hexes (although maintained roads will increase travel speed by 50%).

Anyway, that is 34 days to cross Italy on foot.

This will probably result in around a dozen encounters - most of them avoidable or uneventful.

Most expeditions will NOT require you to cross the entire country, of course.

What I'm thinking is that hex travel requires a huge map to be meaningful.

And MAYBE it requires RE-USING hexes. Getting lost, searching several hexes for a landmark, etc.

In my current campaign, the hex map only made sense when the PCs went exploring beyond the limits of the know regions, where there are no more cities or roads.

Herein lies an important distinction others have mentioned: do you want a map for EXPLORATION or TRAVEL? 

Maybe these should be distinct procedures.

Another thing to consider is "Medieval Europe x Old West USA", and which influenced D&D hexcrawling the most. 

I reckon Europe is too dense for most D&D settings. Mapping the USA would require more than 120,000 6-mile hexes, or around 5,000 30-mile hexes (Europe is of a similar size, but in the 1400s-1500s the population of Europe was more than 10-20 times bigger).

Maybe you should consider this BEFORE choosing hex size...

For now, I'll say that big hexes look better for travelling, and you can always sub-divide them as needed when PCs go exploring (or settling in) a particular area.

Additional reading:
https://methodsetmadness.blogspot.com/2024/04/how-big-is-6-mile-hex.html
https://coinsandscrolls.blogspot.com/2019/06/osr-sienas-6-mile-hex.html

The sandbox railroad

Can a sandbox be a railroad?

Despite numerous definitions, sometimes I think of sandbox as "you can go anywhere in here" and railroads as "you can ONLY go this way".

But last week the PCs were exploring an unknown area of the hex map, looking for a ruined castle of legend. 

They were surrounded by forests, and had no exact clues about the location of the castle (only that it is "near the river" and a couple of clues they can find in the way), so they just wandered around a bit, looking for more information, "opening" new hexes.

(BTW, Hextml has this cool "fog of war" effect that can help you manage that).

They still haven't found the castle; I decided that it is big enough that they will see it if they cross the right hex.

So, you could say they are in a sandbox... they can go anywhere, but... until they find the castle, there is no difference. There is just more forest and rolling for random encounters.

This is not exactly a railroad, but the result is the same no matter the direction they go (again, until they find the castle).

Let me illustrate this:

Lets say the green area is a forest, brown is mountain, blue is sea.

In an hex map, if you're going from "B" to "A", you can try different ways - assuming you cannot simply draw a straight line because you don't know the EXACT location of B.

But it doesn't make a difference. It is all forest. Same speed, same chances of getting lost, same random encounters.

In fact, even if the PCs suddenly decide to go THE OPPOSITE DIRECTION, to C, or if they get lost and wander into the mountains of the North, or travel South until they see the ocean... you STILL know exactly what is going to happen in the next week or so (if that's the distance from A to B).

Of course, this does not feel like a railroad if you have interesting, preexisting, features in EACH hex (which I don't).

Random features and random encounters, however, do not change the railroad "format" of this experience. If you roll to find out there is a cottage or an ogre in the new hex, it still doesn't matter if the PCs went North or South: the result would be the same.

Anyway, I don't think this is an actual railroad, because, while the PATH seems identical, the destination is different. The PCs haven't had their choices negated. Their choices are irrelevant for this week (they literally have a random encounter with an ogre whichever direction they go), but ultimately they'll arrive in a different destination.

But this is an interesting aspect to consider. There are SEVERAL dungeons scattered across my sandbox, and I'm always a bit worried that the PCs choose to go to a dungeon I'm not entirely familiar with (I cannot memorize all this stuff, TBH). 

I don't want to "force" them into a dungeon , but I'd really like to spend at least half an hour reading it before they enter.

The fact that I know exactly what happens on the way to the dungeon - NO MATTER WHICH WAY - might be an interesting tool to give me time to prepare. 

Maybe I could even roll/prepare half a dozen encounters beforehand (I'm currently using this) and know exactly what is coming in the next session - regardless of the direction the PCs take.

Corrupt + Evil = good? (factions)

A quick thought about factions.

In my current campaign, I have two "empires" fighting for a land that is divided into several unaligned clans. The PCs are within the eastern empire and assume these are the "good guys", despite my (too subtle) hints that they can be brutal invaders.

They heard tales about the western empire and it looks undeniably evil. But I didn't want to turn this into a "good versus evil" game. What am I missing?

One thing might be the absence of selfish, corrupt traitors in the OTHER side.

The PCs met several bad people on THEIR side (although they haven't always realized that). But the opposition seem too orderly; everyone is a zealot for their evil cause (well, most of them are undead...).

Against such a determined enemy, there is not much to do - except fighting.

Now, if we add some opportunists to the opposition... the PCs have someone to deal with. Evil, despicable maybe, but someone who can be reasoned with.

Not simple turncoats that "become" good like Han Solo or even Lando, but irredeemable foes that can still be USEFUL.

Come to think of it, adding some corruption and backstabbing to the "evil" side might be an important aspect of dark fantasy. In "epic" fantasy, it seems you only leave the "dark" side by joining the good guys.

Of course, making deals with mercenaries, opportunists and sadists has a price - and they are likely to turn on you later on.

In having a hard time coming up with examples other than Elric, who uses the help of the Lords of Chaos until they finally turn on him (or vice-versa).

In addition, dark fantasy may include people that ARE good fighting for the dark side BECAUSE they have an oath etc. to fulfill. But these are people the PCs are not likely to reason with, no matter how noble they are.

Amarod (I) - Creating a sandbox

I started my current sandbox campaign in late 2022. 

I'm using Dark Fantasy Basic with a few updates.

We play a bit more than a couple of hours every couple of weeks, so it went slowly... but the system is simple and I hand-wave some stuff, so things hopefully are not too slow.

As it approaches a (possible?) ending, I thought I'd make a brief recap - but this post also has some advice on how to build your own sandbox campaign, including sharing some of my mistakes.

The first thing you have to do is to draw a map. This is really easy - I do NOT recommend random generators here. 

I used (and recommend) Hextml. It allows you to expand the map as you go. The PC map currently looks like this:

MY map in Hextml is a bit bigger, but the PCs haven't explored much of it. Their map was even smaller when they first arrived in Belarte (the capitol) by boat, from a distant land that is NOT part of the campaign.

My advice here is start small and add more stuff as the PCs explore.

I wanted to try an hexcrawl, but I'm thinking maybe a pointcrawl would be better, with an old-looking map. 

In any case, as you can see I eventually wrote down the distances in the map (6d = 6 days, and so on) to make things easier. There are few roads beyond Aplos, so they are currently back to counting hexes.

I wrote a brief story of the region (Amarod), which is very simple: an ancient (western) empire fell, now the land is full of warring clans, but the (eastern) new empire is invading these lands. The PCs are mercenaries trying to make a buck, owning no allegiance to either "side", but taking advantage of the eastern invaders relative safety.

These are some important aspects of most D&D settings: ancient empires, ruins/dungeons, a social order that is between stable and flexible, factions and shades of gray.

The second step is scattering a few dungeons, caves, towers and villages around. In my case, I used various BFRPG modules plus Doom of the Savage Kings, The God That Crawls, etc. Mektlan is Tamoachan - the PC's latest exploit.

I did not have all these modules memorized, of course, which caused some issues. Read this post - it contains some important sandbox experience IMO.

Using mostly 1st-level modules was sub-optimal; when they got to level 3 or so, things quickly became too easy. In a future campaign I might start them at level 2 or 3 so the curve is less steep.

I assigned the area around Belarte as "safer", meaning encounters are less likely.

Now, I could say that PCs were free to roam around... but they'd be lost. So I offered them a few rumors and quests about nearby places, and several NPC expeditions they could join. 

It kinda worked. By level 3 or 4, they were planning their own expeditions and hiring their own guides without me having to offer anything.

I recommend starting with three suggestions per session. Some of them should be time-sensitive, but I haven't really implemented a system for that. This deserves further reflection, as the PCs have been treating time as an endless resource ("let's rest for three days to recover", "we will come for the goblins some other day", etc.).

NPC guides and hirelings are very useful, but RUNNING them was a hassle. Now I leave this mostly to the players.

So far, we have the PCs basically going around exploring rumors and places. As they went, I wrote down questions about the consequences of their actions:

1) What will happen in the village of Marval after the priest’s death?  A: A new priest (Father Ostid) has been appointed, you don’t know him well yet.

2) Did Polgrim and  Wolson - the NPCs who went with you to the church in Marval - talk to anyone about what happened? Remember they left when you planned to attack the evil priest.

3) Who sent a letter to the priest saying “the Lord suspects nothing”, signed only with the letter S?

4) Are the kobold ruins empty now the PCs have cleared most of them?

5) What will happen to the Jarl of Savakir (and his seer) now that the beast is dead?

Also some notes, such as "The PCs became heroes in Peranegra after defeating the local elves", etc.

One mistake is that I should have written MORE notes. Maybe every session. It would help me to get a clear picture of the events potentially unfolding.

I've been answering this questions as the adventure advances. Again, the question of TIME deserves reflection. The answer may be different depending of how long it takes for the PCs to go back to any location.

I organize ALL of this in discord channels. The players have access to maps, NPCs, questions, etc. One channel is exclusively for momentary concerns (monsters, dungeons, etc.), other channel is for dice rolls. The list of text channels goes more or less like this:

# general

# rules

# setting maps

# local maps

# adventures (momentary concerns)

# PCs and NPCs

# rolls

# questions

The map channel contains a summary of known cities and events. Each entry after the first was added as they visited new places.

Belarte: the largest port, where you arrived. Lord Belarte. 

Peranegra: famous for its pigs. You expelled the elf king who kidnapped children, and you are well-liked there. Lord Figworth (count). 

Marval: a small town with a church that has a giant slime underneath. You threw an evil priest down there. 

Savakir: a town surrounded by palisades. You killed the local monster but fled because the Jarl was chasing you. Bernier (the Jarl) and Sylle Ru (seer). 

Ilmare: a swampy city with rumors of monsters and bandits roaming the area, as well as abandoned mines. Lord Montegel. 

Mektlan: a cursed place with ruins of an ancient buried pyramid. 

Glani: a prosperous city in the region, but occasionally attacked by native clans. Duke Darvon. 

Aplos: the largest city in the region, famous for its large cathedral, with somewhat labyrinthine streets. Cardinal Godebert lifted Pete’s curse and spoke about the Heretic. You rescued the Cardinal during the siege of Glani.

To sum it up, if you want more advice: TAKE NOTES and KEEP THINGS ORGANIZED.

Anyway, that is what I've got for now. Let me know if you have any questions so I can expand this further in an eventual part II!

UPDATE: I just realized I was missing one important channel: #calendar. KEEPING TIME is extremely important. See here:

https://methodsetmadness.blogspot.com/2024/09/random-ticking-clocks-simplified.html

Random Wilderness is too random

Last post, I briefly mentioned some reasons why I dislike randomly generated dungeons. I think they can be fun, just not my favorite.

I feel random wilderness is a lot worse.

Theoretically, you could run a "no prep"* hexcrawl, deciding randomly upon entering an hex if you're in a forest or desert, and if you see ruins or nothing.

BTW, "no prep" is the idea that you can run adventures with no preparation - by using random tables, improvising on the spot, using other procedures to generate adventures/situations, etc. This issue deserves a post of its own in the future, but I think this "random wilderness" idea illustrates why I think "no prep" is a bad idea if it requires random terrains.

One problem with random terrain is that even the best tables I've seen (the ones who default to a "next hex is similar to this one" rule) cannot create a simple, coherent map like this:

Now, think of how many random hexes you need to run a campaign. Sure, you could set the entire campaign in a single hex, but Outdoor Survival - the original hex map - uses more than a thousand and is representative of an area much smaller than the US.

If you draw a big mountain range, and maybe add another small mountain range, and decide for yourself which way is the ocean... you have created a map that looks more believable and saved yourself more than a thousand rolls.

There are other reasons to set the mountains and the ocean before the PCs start travelling:

- Mountains can usually be seem from several hexes away. It would be absurd to walk from a plain to a "sudden" mountain in a clear day.

- Most people in history have at least a vague idea to which way is the sea.

- Mountains and seas function as natural barriers to your sandbox - crossing them requires more preparation than walking over plains.

- Once you have mountains and seas, rivers are very easy to figure out. Draw rivers like trees, with a trunk that ends in the ocean and several branches (tributaries) towards mountains.

The red lines represent the tallest mountains: rivers do not cross them.

Notice that the presence of a second, significantly smaller mountain range to the east makes the map more interesting.

You might say that US topography is too simple, but it is like that over most of the world. 

South America is similar. Russia has basically sea to the north, mountains to the south and east. Asia has many mountains but the Himalayas can be used as the primary delimitation between various regions (and the source of many rivers). 

Etc.

I'm not saying you need to establish every hill and every trade route. Maybe you can even generate vegetation randomly as you go (although forests and deserts are related to rivers and mountains). I'm just saying having a good outline is incredibly useful - and easy.

Letting PCs wander around with no preexisting terrain has other problems. For example, why there are no map in this land - and what happens when the PCs find a map in random treasure? How can there be significant rumors of goblins "in the North" if the DM doesn't know what lies in the north?

But anyway, this is about drawing maps - and I reckon the easiest way to do that is simply:

- Separate land from sea.

- Put a big range of tall mountains to a random direction.

- Add a smaller mountain range, with hills etc.

- Rivers are easy to draw once you have that.

- Most cities are near the water (rivers or seas).

Additional reading:

https://methodsetmadness.blogspot.com/2024/08/the-mythic-underworld-controversy-and.html

https://methodsetmadness.blogspot.com/2021/08/easy-worldbuilding-with-pareto-price.html

Fallout: Carcosa

Nine years have passed since I first started thinking of creating a "Dark Sun-ish" setting.

(Nine years. Whoa. That is quite a lot. And it was fast. I have to think about that. Anyway.).

At that time, I mentioned the Fallout series as one inspiration - besides Barsoom, Carcosa* and Dune.

A few things have changed since then - maybe one or two authors mentioned in that post turned out to be contemptible at best, so I'll avoid them - but, once again, I've been thinking of Fallout and Barsoom, and wishing to run a campaign in such setting.

It was probably the TV series that made me want to play Fallout again. I had only played Fallout 3, and now I'm playing New Vegas. 

It really makes me want to create my own post-apoc setting - "with blackjack and hookers", as the saying goes, but New Vegas already has some of those...

Carcosa* is one of these games that I always wanted to run but never quite got around to it. It is full of awesome ideas, but I think it stops just a bit short of being an amazing setting like Dark Sun. I hear that the author has other modules in the same setting, but these are not in DTRPG.

Here are a few elements from New Vegas I'd like to add to Carcosa in order to make the setting more interesting.

Vaults

Fallout vaults are a source of endless possibility and fun. 

First, each one is a dungeon to explore - but a dungeon with a real purpose and history. 

Second, vaults are created as different experiments, so each vault can be unique and even contain different creatures (including mutants, robots, etc.). Some experiments are social, so different forms of government are to be expected - you can check this post or Dark Fantasy Places for ideas.

Factions

New Vegas has several interesting factions. Some evil, some neutral, some trying to do 
good, with many shades of gray,

"Everyone wants to save the world, they just disagree on how" as the show mentions.

There are several factions that think that things will be a lot nicer when THEY are in power, even if they need to brutalize some folks to get there.

Carcosa doesn't have much of that - all groups are similar, and governed by a single individual. Add a few details and things start to get interesting.

In addition, Carcosa has "men" int he encounter table, but doesn't give you anything else - are these merchants? Explorers? Adventurers? We should probably add some ideas here.

Lost technology

This is another fun part of both Fallout and Carcosa - projectile weapons, power cells, power armor, robots and cyborgs are in both settings.

Maybe the system deserves some kind of "repair/tech" skill, allowing some PCs to try to access technology they cannot fully understand.

Mutated monsters

In Carcosa, "The spawn of Shub-Niggurath are the innumerable and typically unique monsters [...] These [...] are the most common type of monster on the planet of Carcosa". There are also numerous oozes, jellies, worms and dinosaurs.

Which is fun. But I think mutated/giant insects (or other beasts) might be an interesting addition to the setting. 

Unique monsters are cool, but not every monster needs to be unique - sometimes, fighting a bunch of giant cockroaches or giant scorpions can be fun.

Sandbox & Hexcrawl

I have been critical of hex-crawls for a while now. 

It doesn't make sense that an adult wouldn't know at least vague directions to nearby lands. The idea that you enter a new hex and suddenly see a new city (or the sea!) sounds absurd. There are exceptions - maybe when discovering a new continent, etc. Still, if there are intelligent people around, it should be easy to ask for directions.

Unless...

What if the PCs are vault dwellers? When they come out to the world, they have no idea where they are! They still can see a mountain at a distance, but not much more than that. It would be a nearly ideal setting for a hex-crawl.

Can we use Carcosa hexes?

Carcosa hexes are often interesting but sometimes very repetitive and terse ("Castle of 29 Black Men led by a neutral 5th-level Fighter"). 

Human groups are very small, somewhat similarly to Fallout.

Maybe we can combine existing hexes with some random tables to add detail.

Do we still need magic?

Fallout doesn't have magic spells. Carcosa has rituals, but they are very unique - unlike anything I've seen in other RPGs. Dark Sun has proper spells. But do we need them?

Maybe psionics would work better here. And, of course, technology that is "indistinguishable from magic".

But the rituals in Carcosa are so interesting and unique that I'm doubtful about getting rid of them.

---

Well, that is it. I hope I can actually do something with this before nine more years have passed!

* Affiliate links.

How big is an (6-mile) hex?

I've been thinking of hexes in abstract terms, but I think something more complete would be useful.

I default to the proverbial "6-mile hex" describe here.

Apparently, the area is about 31 square miles (correct me if I'm wrong).

This is bigger than Manhattan (23 square miles).

The Isle of Wight would cover about 2 hexes.

Siena - plus about 20 towns and monasteries - would cover one hex.

The City of London, within the walls, is HALF a square mile - so it covers about 1/60 of the hex.

Notice how thin the Thames would look in a six-mile hex map! It is about 0,16 miles wide near London.

Rivers such as the Danube, Rhine, and Mississippi have an average width of less than one mile.

In short, this means that crossing an hex will not necessarily allow you to see every relevant site. In the plains you are likely to see the entire hex in a clear day, so a village is not difficult to find - especially because there are roads etc. But the village is not hard to AVOID, either.

Anyway, there is no way you can picture every village in a six-mile hex (which I used to do - I marked the hex with a small house).

At this level of detail, it would be better to use 3-mile hexes, which are four times smaller than the original hex - and have their own advantages. 

A 3-mile hex is enough for a village with countryside, or maybe a very large city surrounded by 4-6 hexes of countryside and villages.

In a 1-mile hex, river thickness may start to vary in the map (and maybe even change from season to season). 

Traveling becomes less abstract - you do not cross "the mountains", but choose specific paths.

This is probably too much detail for me. Too much choice with little consequence. I'll stick to 6-mile hexes for now.

---

Additional reading:

http://steamtunnel.blogspot.com/2009/12/in-praise-of-6-mile-hex.html

http://steamtunnel.blogspot.com/2018/09/the-ergonomic-3-mile-hex.html

https://coinsandscrolls.blogspot.com/2019/06/osr-sienas-6-mile-hex.html?m=1

https://coinsandscrolls.blogspot.com/2017/07/osr-fast-mapping-part-3-barony-terrain.html

https://silverarmpress.com/down-with-the-6-mile-hex-a-modest-proposal/

https://ruprechtsrpg.blogspot.com/2020/01/6-mile-hexes.html

https://the-robgoblin.blogspot.com/2024/04/it-takes-village-to-stock-hex.html

Update on the wilderness encounter book

I spent most of last weekend working on my wilderness encounter book for B/X.

The first part is a simplification of the random encounter procedure. I like how that turned out; I'll publish most of it in this post blog soon.

The third part describes 366 days in the wilderness. Basically, weather, encounters and a few details for variety.

These parts are mostly finished by now.

The second part is the hardest. It describes, very succinctly, ONE THOUSAND random encounters, with most necessary rolls (number appearing, surprise, reaction). This is how it looks right now:

This was a bit harder than I thought. Rolling dice thousands of times was easy with the help of apps and AI. Same for adding names to NPCs. 

The hardest part right now is "confused wolves". 

I had a hard time thinking of hundreds of variations for what's the monsters are doing. 

I could certainly take some inspiration from this amazing blog d4 caltrops, but I still feel it will get somewhat repetitive anyway. Most aggressive animals and beasts are just hungry or territorial - I can think of a few alternatives, but not hundreds of them.

At least intelligent NPCs, dragons, etc. are looking pretty good IMO.

B/X descriptions are very terse, so I have to add a few details from other sources.

Anyway, it's lots of work, but maybe it is worth it if other people find the book useful. 

Well, at least I've been using it in my own sandbox campaign.

And I've been learning a lot about B/X monsters and encounter tables (mountains are VERY dangerous, and troglodytes are SCARY - 5d8 appearing, 2HD, camouflaged, attack everyone).

So for now what I ask you to do, is let me know if that looks interesting/useful, or if you'd change anything.

If this looks interesting and you and want to provide even more feedback, consider joining my discord channel to discuss it.

The book will take a couple of moths or more to complete, but at this point I am pretty confident that it WILL be completed.

BTW: let me know if you have any reservations about the use of apps and AI for dice rolls, random names, etc.

Wilderness level tables

Just a quick update to the last post.

Another method of achieving a similar result, if you prefer using a table to adding dice, etc.

Roll 1d20, add the number of hexes away from civilization (maximum 10), and check the table below to see how many monsters appear. 

E.g., if you roll a total of 12 and the "number appearing" is usually 1d6, then 3 monsters appear.

Special cases:

Rolling 25 or more: There is a chance that you've found a monster lair (25% or whatever the GM deems appropriate).

Lairs: if you find a lair, roll number appearing as usual (for example, a lair of orcs containing 5d6x10 orcs - just roll 5d6x50 regardless of the initial result, or keep the initial result as a minimum - see the examples below).

0 monsters appearing: either the encounter doesn't happen, or the monster is of a smaller/weaker variant  (whatever the GM deems appropriate).

Natural 20: if you roll a natural 20, roll again and add 10.

Natural 1: roll again and subtract 10.

Examples:

- You find some ogres (number appearing: 2d6), 8 hexes away from civilization. You roll 1d20+8, and get a 21. This means 5 ogres were found for each d6, for a total of 10.

- You find some orcs (NA: 1d6x10), 5 hexes away from civilization. You roll 1d20+5, but you get a natural 20. Then you roll again with a +10 bonus, getting a total of 31. This means you found at least 70 orcs, but there is a chance you have found their lair. The DM decides this is a viable place for an orc fortress and roll 5d6x10, getting 150 orcs.

- You encounter dragons  (NA: 1d4), 2 hexes away from civilization, but you roll low and the table indicates 0 dragons. The GM decides wether this encounters doesn't happen (e.g., you see a dragon flying far away in the distance, you find dragon tracks, etc.) or if you find a smaller dragon, drake, etc.

Variations:

You could use 2d10 to make the outcomes more predictable.

You can use the table below if you want fractions; this is useful in multiple situations (e.g., when you find 2d6 ogres or 1d6x10 orcs, using fractions gives you more nuance).

Yes, my version of excel uses commas instead of decimal points.

Round to the nearest integer.

The formula:

The formula I used is: D*R/25.

D = Dice size (4 for a d4, 6 for a d6, etc.).

R = Result of d20+hexes.

Wilderness levels

This is an idea I have mentioned before. I've been thinking about it for a while.

Despite my dislike of "encounter balance" as it is commonly understood (i.e., in relation to the PCs), I appreciate the need of letting the PCs know - and, ultimately, choose - what kind of encounters they are willing to face (of course, their place we will often go awry).

Traditionally, "encounter balance" in old school D&D is not calculated in relation to the PCs, but in relation to dungeon levels. So, the third level in a dungeon has harder encounters than the second, and so on.

I have several issues with this (e.g., each dungeon should have its own tables, how do dragons live so deep if there are no apparent big entrances), but that is another subject. 

My main concern right now is encounter balance in the wilderness, something I also already analyzed here.

Basically, B/X has TWO "encounter balance" tools for the wilderness. First, do not go unless you're level 4. Second, some terrains are more dangerous than others (although there is an good chance of finding 1d4 dragons even in settled lands...).

This is not satisfactory because it lacks the nuance of "dungeon levels" - and also because it stops levels 1-3 PCs from traveling around to get a "taste" of what to come.

The obvious solutions is analogous to dungeon levels - the farther you are from civilization, the greater the wilderness level (you are "deep" in the forest, etc.).

To avoid creating entirely new tables (something that 5e almost manages to pull off with XGtE and ToA, to some extent), we could adapt exiting tables for these idea.

For example: roll 1d20 for each hex to the nearest the nearest civilization/ safe outpost (maximum 10). 

This is a percentage of "number appearing". 100% means the maximum number appears, but that will be rare. When you get "1d4 dragons" it is more likely a single dragon unless you're a few days from civilization. Smaller numbers (e.g., 10%) usually indicate a single monster (but see "what if number appearing is one", below, for alternatives).

To add some spice, 20s explode; roll again and add 10, "ad infinitum". If the FIRST dice rolled is a natural 1-2, the rule is simply ignored - roll 1d4 to find how many dragons. 

Even in the safest areas there is a chance of fighting a group of powerful monsters, but it is small (also, they might be there for a reason...).

[Alternatively, if you don't want to throw that many dice, just count some of the d20s as 10 without rolling. For example, for six hexes roll 3d20+30].

What if the result is over 100%? It is up to the GM. Rolling 130% might indicate a 30% of having found an actual lair (with up to five times more monsters), or just finding five dragons instead of 1d4.

What if "number appearing" is "one" (e.g., hydra)? In this case, the percentage indicates a chance of the encounter happening at all. An hydra next to a city? Unlikely to happen. Other alternatives would include changing the number of heads or even HD for some creatures (a dragon found right next to town was undetected for being too small, ).

What about humans? Humans and maybe some humanoids, of course, are exempt from this system; they are likely to be encountered near civilization. Elves, etc., will be more or less common according to setting.

[EDIT: as suggested in the comments, maybe changing die size could serve a similar function. So, 1d4 dragons become 1d2 near civilization, 1d3 a bit further away, then 1d4 and even 1d6. Come to think of it, this looks like a more elegant solution. Maybe results that are greater than the usual maximum  - e.g., rolling 6 for dragons - indicate a chance of lair].

Building a castle or keep is automatically useful, as it makes travelling less dangerous. Roads could provide a similar result (especially if patrolled), although traveling faster is useful enough.

Add modifiers to taste: greater chance of encounter near the darkest dungeons/castles, smaller chance if travelling light with a small group, more giants in the giant territory, etc.

A more nuanced version would be creating "danger zones", each adjusting number appearing. Ideally, each area would have its own danger "level" and encounters, so that we are not "balancing encounters" but "balancing encounter tables".

Not all encounters in the same table are equal, of course. Instead, as suggested in AD&D 2e, some creatures are RARER than others regardless of HD. 

It is about a fictional world - not about a game of HDs.

Of course, if you're using the tables in Xanathar's (they are decent), this is even easier: maybe 1d8+hexes, with 1-4 meaning tier 1, 5-10 tier 2 and so on.

[Notice that in 5e, similarly to B/X, Pcs of level 1 and 2 might have a hard time anyway, although the difference is smaller].

Another thing to consider, maybe in future post: cities and outpost, despite being static, might have occasional random encounters of their own (or should this be part of a "disaster check" table?). that small village encountered 4 dragons when the PCs were elsewhere? It might be toast by now...

Additional reading:

https://methodsetmadness.blogspot.com/2023/12/encounter-balance-why-i-avoid-using-it.html

https://methodsetmadness.blogspot.com/2023/11/how-d-5e-xgte-encounters-succeed-and.html

https://methodsetmadness.blogspot.com/2023/11/good-generic-wilderness-encounter.html

https://methodsetmadness.blogspot.com/2023/11/bx-wilderness-encounters-are-bonkers.html

Clans of Amarod

These area few clans for my current hexcrawl. If my players read this: there are dozens of clans, you might encounter these ones or not. 

Otherwise, feel free to use them in your own settings! Some entries might be specific to my setting, adapt as needed.

One thing that occurs to me is that I should have decided this beforehand - see this. I think I'm doing this for every setting from now own, as I cannot stand orcs anymore (I unfortunately already introduced dwarves and kobolds from published modules - dwarves are useful, as goblins IMO, but kobolds scream "we are in D&Dland" for me).

I think you should try it too - even if you like orcs, it is nice to have distinct clans, some more aggressive than others. And having FEWER creatures gives a setting a stronger theme and feel.

The last part is from Dark Fantasy Places (which you can get for free/PWYW), and not particular to this setting. I pasted here for convenience.

Uzumaki (goblin cannibals)
One of the most hated groups. Easily recognizable for their sharp teeth and spiral-shaped ritual scars. Hide in the woods, attack on sight, and laugh as they fight to death. Probably insane. When fighting other goblins, they prefer capturing to destroying, and will buy/ "rescue" goblin slaves if they are available.

Moonchildren (tabaxi mystics)

Curious students of the occult. Wear silver amulets, usually moon and starts. Like spellcasters, enjoy wine and singing. Collect books and memorize songs, exchange information. Usually nocturnal.

Ironfolk (silent dwarves)

They do not speak much Common, but do not look for trouble either. Found in mines and forests. Use body paint, usually white, with red and black details. Not much armor (bare chested), but carry heavy iron weapons. Their caves are decorated with stone statues - rectangular cuboid carved to resemble dwarves.

Skullfolk (human skeletons)

These thin folks use white and black paint to resemble skeletons when they are in battle (or raiding, invading, stealing, etc.). They are pale and have no body hair at all. A few of them cut their own noses. Their weapons are poisoned and they are extremely hostile when in paint (50% chance).

Dragon claw (wild elves)

These tattooed, scantly clad elves have a deep respect for dragons, although they recognize not all dragons are good. They are distrustful of anyone wearing heavy armor, as these are more common in the invaders. Discreet body paint, long braided hair, curved blades. Nomadic, claim no lands as their own.

Glowtoads (shy frogfolk)

They live in the marshes and avoid getting close to civilization. Sometimes they use special paint (usually curves and dots) that make them glow at night. Will use as a diversion tactic when needed. Otherwise, they are usually friendly to humans but hate mantisfolk.

Warp folk (human mutants)

These misshapen humanoids worship a green rock which has a strange glow. A few generations of natural selection made them a bit more resilient to negative mutations, but they still look ugly (scaly skin, asymmetrical features). Feared and despised by most, but not actually as aggressive as other clans.

Sky terrors (cupendiepe hunters)

Nocturnal, blood-drinking, winged humanoids that fight with moon-shaped axes. Very aggressive. Live in caverns near mountains tops, steal treasure for no apparent reason.

Firestarters (human/goblin pyromaniacs)

The worshippers of the Lord of Fire cover their bodies in ashes and burn their enemies to commemorate victories. Fight with oil flasks and fire traps. Occasionally worship at the crater of an extinct (?) volcano, but the location is secret to everyone except their shamans, who have been "touched by fire" (burn scars).

Greenscales (powerful lizardfolk)

The greenscales are the most powerful lizardfolk clan in the land. They take a nuanced view of the invaders; they know they are feared, but can make powerful allies. They might engaged peacefully with smaller groups. They were enslaved under some of the ruined empires and they hate magic and the cult of Apopep with its corrupted sorcerers.

Random clans

Type (1d20)

1. Tabaxi

2. Lizadfolk

3. Goblin

4. Cupendiepe

5. Dwarves

6. Elves

7. Mantisfolk

8. Frogfolk

9. Kobold

10+. Humans

Looks (1d20)

1. Naked

2. Tattoos

3. Body paint

4. Skulls

5. Sharp teeth (in mouth or necklace)

6. Demon mask

7. Colorful silk

8. Spikes

9. Ritual scars

10. Reptile hide 

11. Camouflage

12. Feathers

13. Wood armor

14. Jewelry

15. Snake motifs

16. Jaguar skin/motifs

17. Shell necklace

18. Light robes

19. Colorful hair/ beads

20. Braids

Like/dislike (1d20) - [or: respect, fear, worship, etc.]

1. Heavy armor

2. Strangers

3. Cannibalism

4. Spellcasters

5. Nearby tribe

6. Dragons

7. Castles

8. Undead

9. The Water Lord (main Lawful deity)

10. Alcohol

11. Commerce

12. The Ruined Empire (Apiaka, etc.)

13. Lord Belarte (or other nearby leader)

14. Monsters (like: 50% chance of pets/mounts)

15. Hallucinogenic plants/fungi

16. Other species (like: 30% chance of mixed clan)

17. Mercenaries / dungeon delvers (like: can be hired)

18. Agriculture

19. "The Natural Order"

20. Slavery/slavers

Some additions and variations

Tabaxi. Cat, cougar, red, blueish. A rare lionfolk might be found in the North. Many tell legends of ruined Muru, the circular city that stands in ruins - but ascended, "in spirit", to higher planes.

Lizadfolk. Red, mustard, blue, dwarf, horned, rarely white, black, chameleon. Most distrust magic, but a few are loyal to Apopep and the serpent-people that enslaved them in the past.

Goblin. Alchemists, tree-jumpers, burrowers (gray and earthy).

Dwarves. Albino, explorer, slaver, merchant.

Elves. Types vary, but mostly the typical "wood elves", closer to nature, or "underground elves" that enjoy tricks and illusions. Both are familiar with fey, dryads, etc.

Mantisfolk. Dark, mustard, red, green. A few clans are psionically held to a "hive mother" like bees and incapable of independent thought while near. Some gain individuality to perform outside jobs, and some gain it permanently due to genetics, accidents or death of a hive mother.

Morlocks.

Other clans

Orcs - None. Replace by lizardfolk.

Hobgoblins - None. Replace by morlocks, pictured above. As hobgoblins, "dwell underground, but commonly seek prey above ground."

Aaracroka - None. Replace by cupendiepe, a more aggressive species of bat-people.

Etc.

Villages & Cities

From Dark Fantasy Places. Unlikely to be used until the PCs decide to interact with one of the clans.

d20 Appearance

1. Intertwined with trees

2. Buried

3. Lifted from the ground

4. Made of moving tents

5. Camouflaged

6. Suspended over water

7. Built amidst ancient ruins

8. Huge and mostly empty

9. Dug up in rocks

10. A single building

11. On a mountain top

12. Brightful colors

13. Tall buildings

14. Misty

15. Narrow streets

16. Inside a crater

17. Constantly flooded

18. Poor and ruined

19. Beautiful and frail

20. Heavily fortified

d20 Society

1. Property is communal

2. No concept of privacy

3. Identity is defined by masks

4. Rigid protocol for every conversation

5. Universal vow of silence

6. Adults are cast out

7. Appropriate clothing is mandatory

8. Visitors have no rights

9. All burials must be in pairs

10. Children are raised by all

11. A proper season for each action

12. Travelers welcome for a single day

13. Weapons are forbidden

14. “Couples” are always three

15. Those who can’t fight must serve

16. No activities during daytime

17. Violence is never the answer

18. Everything is permitted

19. Might makes right

20. Casual cannibalism

d20 Problems

1. Besieged by monsters

2. Ruled by tyrants

3. Plagued by poverty

4. Infected by disease

5. Fiercely territorial

6. Infiltrated by demons

7. Filled with criminals

8. Resentful of outsiders

9. Ongoing power struggle

10. Cursed with madness

11. Violently expansionist

12. Falsely utopian

13. Mutated by Corruption

14. Malignant religion

15. Screwed morals

16. Widespread panic

17. Hopelessly defeated

18. Hiding dark secrets

19. Inhabitants cannot leave

20. Periodic human sacrifice