This post is part of my series on developing the sandbox. You can read other posts in the series here. The series will, when necessary, go into detail on the development of my sandbox for Dungeon Module B1: In Search of the Unknown, but will also discuss sandbox development more generally. You can read play reports from my B1 campaign here.
I'll be honest - these days, I don't actually much enjoy making dungeons. You have to figure out what the dungeon used to be, who built it, what happened to it, who's been using it since, who's using it now, how they get along with their neighbors, where they live, (sometimes) what they eat and drink, what they're doing there, what stuff they have, what stuff is in the dungeon that they don't have, why they haven't gotten it yet, why are there traps, what the traps do and how they work, what a "special" room is, why whoever built the place would put it there, what the layout is, what the rooms used to be and what mundane stuff would be in there to play with, what are the doors like, which ones are secret, which ones are locked, how do the inhabitants get around, how many levels are there and how many rooms are there per level, where are the stairs, what's the percentage chance of encountering such-and-such while wandering around the place - the list goes on and on.
To compound all of this, there is so much advice to be found on the subject, whether it's in rulebooks throughout the many editions of D&D, other TTRPG systems, or online on blogs. It is really overwhelming to take it all in, decide what you like and what you don't, synthesize it all into a coherent methodology, and apply it when designing dungeons for your home game.
It's a lot of level design. There was a time when I really enjoyed that part of the game a lot, but lately, making dungeons has made me very frustrated more often than not. And then you have to write that all up in a key that's comprehensive while also being simple enough to run at the table.
You have to put in similar work when designing a a stronghold or monster lair, but there's plenty of information about real world strongholds to draw upon, and monster lairs are often smaller, simpler, and easier to imagine. There are dungeons in literature, but few that actually emulate the D&D dungeon-ass dungeon. You're better off just reading old TSR modules to figure out what a dungeon should more or less be like.
I'm more of a big picture guy. I like making big maps with varied terrain, then finding out what all is there, whether it be cities, strongholds, dungeons, or monster lairs. Who lives there? Who rules it? What kind of monsters plague this or that region? How do they all relate? For me, that's where the juice is.
That's why I've basically just started stealing my dungeons. I originally decided to create my B1 sandbox because (1) I thought the dungeon-focused adventure was crying out to be placed within a larger context and (2) I really liked the idea of one of the more tedious parts of the sandbox development process (designing the dungeon) being done for me.
B1, however, does not feature a finished dungeon. The DM is instructed to roll to determine if each of Quasqueton's rooms contains monsters, treasure, or neither. This is relatively easy, however, since the module provides lists of monsters and treasure which the DM can place in each room when a positive result is determined. All that's left is to use the module's thorough room descriptions to figure out where the treasure is stashed, to determine how the monsters relate to one another, and create new wandering monster tables based on the monsters found in my version of Quasqueton.
I used some loosely interpreted opposed reaction rolls to determine monster relationships, eyeballing the map to find out who lives near who (combining groups into factions when appropriate) and who controls what territory.
What I ended up with was some goblins who control the dungeon's entrance, some orcs deeper within who threaten the goblins and have lured away some of their numbers, and a few smaller groups of intelligent creatures and roaming non-intelligent monsters which are indifferent to the conflict. This has worked out quite well so far, with the players deciding to try to stay on good terms with the gatekeeping goblins, the goblins asking the characters for exchanges of favors, and the orcs launching attacks against the goblins when the party is away from the dungeon.
The DM is also instructed to roll to determine when doors are stuck, which requires more eyeballing to determine who can get into what areas, whose territory they'll need to pass through, and how that affects the dungeon's politics and who is aware of whom. For example, if the bandits must pass through goblin territory to get to the area of the dungeon in which they've been stocked, then the goblins must know about the bandits and have some opinion about them, and there must be some reason why they haven't killed each other.
For the wandering monster tables, I add up all of the monsters in the dungeon to get a total number of monsters. Then, for each monster type, I divide that monster type's total by the larger total to get a percentage. That is the percentage chance of encountering that type of monster wandering the dungeon.
Let's say I have a 20 room dungeon, and there are monster in five rooms: 11 brigands, 6 hobgoblins, 4 gnolls, 2 rot grubs, and 3 troglodytes. I don't think of the rot grubs as "wandering", so excluding those, I have 24 total monsters - 25% hobgoblins, 46% brigands, 17% gnolls, and 12% troglodytes. My wandering monster table would look like this:
01-46 1d6 Brigands (from Area 8, looking to threaten, plunder, and rob)
47-71 1d3 Hobgoblins (from Area 17, killing any they outmatch)
72-88 1d3 Gnolls (from Area 11, willing to trade information for treasure)
89-100 1d2 Troglodytes (from Area 20, looking for sacrifices for the God of Stink)
(I'm assuming this is a 1st level dungeon, and the number appearing is based on the assumption that there will be 4 to 6 characters of that level, using my cataloguing of AD&D 2e monsters by level to determine roughly how many of each monster would be appropriate. You'll want to make it work for your own assumptions about party size and whatever system you're using.)
However, adapting B1 for my campaign wasn't as simple as stocking the dungeon, making a wandering monster table, and then using the module as written. Like, this isn't really usable:
There's an art to turning this into something that functions at the table. You can't read it as is. You could maybe skim it and put it into your own words on the fly, but you're wasting everyone's time. You could memorize it somehow, but good luck with that. My process has been to rewrite it in my own style, in a way where the interactable elements and the hierarchy of information is plain to see. In my style, this room looks like this:
Descriptive, but not verbose. Concise, but comprehensive. Important stuff is bolded. Bolded items are described in further detail. Additional information is available upon further inspection. I can eyeball the initial description and rephrase it in my own words, then go into further detail when the players spend time examining any one thing. This works for me. Yours might look different.
This is all very specific to my B1 sandbox, or any sandbox in which your dungeons are repurposed from existing published material. That isn't always the case. Though I don't love doing it these days, when I choose to do so (and I'm not wholly opposed to it - my sandboxes for both B2 and B3 include dungeons beyond those detailed in the published modules, after all), how do I design my own dungeons from scratch?
As with other location types, I'll start with a minimalist sketch. In the case of dungeons, this includes the type of structure it once was, who built it, how it fell into ruin, and (broadly) who or what dwells there now.
In the AD&D 1e DMG, the ruin types are limited to deserted strongholds (with "mazelike dungeons thereunder" and the occasional monster using the surface ruins as a lair), villages (here I'd include the other classes of small settlement like hamlets and thorps as well), cities (or towns), tombs, and shrines. Tombs and shrines are classic dungeon fare. Deserted strongholds are good fodder as well, especially when considering B1 and how it depicts the maddening, hostile subterranean environments which bored, retired adventurers are likely to carve out beneath their seats of power.
Villages and cities are a bit trickier. They're not quite "dungeons" in the traditional sense. I tend to build ruined cities and towns as I would normally build a city or town, then wreck them. The characters can still crawl through the settlement's ruins district by district just as they would crawl through such a place were it not ruined. Perhaps it is now overrun by the doppelgängers, fiends, lycanthropes, and undead which have always prowled its streets. Navigating the settlement itself might be more akin to a point crawl. It's the locations which were once important to the settlement that are the proper dungeons.
Using my B1 sandbox as an example, if Timbershore were to fall into ruin in the far away future, perhaps the remains of the Evil wizard Lambrecht's castle is one such dungeon, and the ruins of the Evil cleric Blanchefleur's keep another. Add in some crumbling manors or a merchant consortium's sealed treasure vault for variety. A ruined village, hamlet, or thorp could be much the same, albeit smaller and easier to get around. Instead of ruined castles and manors, a dungeon there might be an old mill or granary - a diversionary location with just a handful of rooms.
As for who built the structure originally, it depends. Strongholds are easy for me, since AD&D is quite explicit as to who can build strongholds and when, and the DMG has nifty (if not, at times, wholly consistent with its own rules) tables for determining who a stronghold's ruler is (or in this case, was).
Shrines are built by clerics, so it's easy to determine who the cleric in question was - just determine their race and alignment and either pick some appropriate classic D&D god and genericize it for your setting or choose some god unique to your setting which you wish to highlight. If there were to be a ruined shrine in my B1 sandbox, for example, I would make it a shrine to the Horned Mother (Blanchefleur's deity), or I would come up with some other deity in opposition to her, its worshipers long since eradicated. I tend to favor either Evil shrines or Good shrines which have been desecrated by Evil powers. Remember that Evil undead are empowered in such places!
The ruins of cities/towns, tombs, and villages/hamlets/thorps are a bit trickier. I quite like the 2014 D&D 5e DMG's tables, which you can roll on to determine who built the dungeon, what type of dungeon it is, and why it was abandoned - so much so that this framework strongly informed my own ideas about creating the initial sketch of such a location. Still, who and what the tables include can be a bit of headscratcher. A dungeon might have been built by dwarves and elves, but not by gnomes or halflings. It might have been built by hobgoblins, but not by orcs. Beholders, giants, mind flayers, and yuan-ti, but not any other monsters.
Instead, it's probably best to look at what's already been established about your setting, and pick or randomly determine some appropriate creator based on that. My B1 sandbox is located in a heavily forested region, so perhaps a civilization of elves once lived there in the ancient past. Rogahn and Zelligar used orc slaves to build Quasqueton, and went to war with barbarians in the region - perhaps the orcs or barbarians had a civilization of their own and built structures which fell into ruin as the two powerful adventurers asserted their dominion over the place. There is an ogre lair in the area - perhaps their ancestors built things as well. Perhaps Timbershore once had outlying villages which were reclaimed by the forest, or perhaps it was once subservient to a larger city which met a similar fate.
The same approach works for dungeon history. Barbarians, ogres, and orcs may have brought ruin to human or elven civilizations and structures, and may have likewise been ruined by the campaigns of Rogahn and Zelligar. Perhaps there was a conflict between worshipers of the Horned Mother and whatever deity's worshipers had a foothold in the region before them. In addition to conflict from war and conquest, you could work in plagues, curses, and magical catastrophes.
All of the preceding information - the type of ruin, who built it, and what happened to it - can inform what lives in the ruin in the present day. A fighter's stronghold which was ruined when a subdued dragon rebelled might still be home to that dragon. A wizard's stronghold might house summoned guardians or magically-altered experiment creatures. Evil and desecrated shrines and tombs will likely have undead or fiends. An enclave of barbarians, elves, ogres, or orcs might still call the ruined city home. A village abandoned due to plague will be home to disease carriers. A curse upon a city may have turned its aristocracy into wererats or vampires.
I try to pick a few different monsters types - intelligent, organized monster groups with lairs and factions inside the dungeon, unintelligent scavengers and predators which roam its rooms and halls, and singular, powerful monsters with or without minions. In AD&D terms, if I'm making a 1st level dungeon, I'll want it to have a mix of 1st, 2nd, and 3rd level monsters, according to the DUNGEON RANDOM MONSTER LEVEL DETERMINATION MATRIX:
I might go so far as to use reaction rolls at this stage to determine who is friendly, hostile, or neutral towards the party, and which monsters are allied or at odds with one another. The Dungeon Checklist is a great resource - make sure there is fodder for the characters to kill, thinking creatures to parley with, and dangerous creatures to be avoided or carefully neutralized.
This all will give you a pretty good sketch, but dungeons aren't the kind of environment you can easily riff on using minimal information. You need to know specifics. How well-lit is it? How high are the ceilings? How many rooms are there? How many levels? How are the rooms laid out? Are they connected by passages or doors? Are the doors locked or stuck? Are they wood, stone, or metal? What's in each room? How many monsters are there? Where are the tricks and traps? How do they work? What damage do they do? What saving throw is needed? Where is the treasure hidden? How much of it is there?
As I've said, there are so many ways to determine these things that it's nearly impossible to suggest best practices. It will differ for everyone based on their taste, system of choice, and campaign setting.
One of the best tangible tips I can offer is to design dungeons in chunks. This somewhat solves the question of how many rooms and how much of each thing (empty rooms, monsters, tricks/traps, and treasure) to include. I use 20-room chunks along with the AD&D 1e dungeon stocking table:
Thus, if I have a 20-room dungeon, there are 12 empty rooms, two with monsters only, three with monsters with treasure (these are my monster "lairs" in the dungeon), one "special" room, one trick or trap room, and one treasure room (like a vault behind an especially secure or secret door). This to me is a small dungeon (boo, hiss, five-room dungeons stink). If I want a bigger one, I add 20 more rooms. If I need more, I add another 20, and so on.
Since each 20-room chunk contains a "special" room which might also contain a stairway, I tend to include one stairway per two 20-room chunks. A level can have multiple stairways up or down, of course, so this isn't necessarily a limitation upon how large a single level can be.
If I want to create a big 100-room dungeon, I know I need 60 empty rooms, 10 rooms with just monsters, 15 monsters with lairs and treasure, 5 trapped rooms, 5 treasure rooms, and so on.
You might use a different dungeon stocking table to inform the size of your dungeon "chunks". For example, from OSE:
Empty rooms are not truly empty, but contain environmental storytelling, hints at what awaits in other rooms, time-wasting set dressing, or interactive elements which might come into play in the future, such as if a wandering monster shows up there, or if the room is restocked with a monster on a future delve.
I like using Courtney Campbell's resource on empty rooms to determine what a room's original purpose was, and from there determine what sort of set dressing to include. I do this with rooms that are otherwise occupied as well. The 1d6+6 orcs are not just standing around in some barren chamber, but are instead in an old kitchen scavenging for food, in the wizard's bedchamber scraping the gold leaf from his headboard, or in the lounge gawking at the marble statue of a nude woman.
The specific monsters I use could be determined by rolling on AD&D's dungeon encounter tables, or I might pick a few based on the dungeon's history. Tricks, traps, and treasure would be similarly informed. An elven ruin might have more magical traps, while a dwarven one might have more traps involving shifting masonry. Gnomes use clockwork devices, kobolds use poison and pots of stinging insects, and defiled shrines are warded by curses and negative energy. Elves treasure ornately carved wooden objects, spellbooks, moonstones, and magical bows and swords, while dwarves treasure stone carvings, gemstones (but not pearls, because dwarves HATE pearls), and magical axes and hammers.
The number of monsters appearing, gameplay statistics of traps, and value of treasures will all be informed by your system of choice, of course.
As for layout, you can create your own dungeon and still steal a little bit. You could take an existing map from one module or another and simply restock it to your liking. This is useful if, for example, you know you want to run the hideout for a cult of Elemental Evil, a fire giant stronghold, or a Tomb of Horrors-type thing, and know where to find one. I ran Tomb of Annihilation's Fane of the Night Serpent, Princes of the Apocalypse's Sacred Stone Monastery, and Storm King Thunder's Forge of the Fire Giants this way, using the maps and room descriptions but changing the monsters, treasure, and mechanics of traps and the like to suit my needs, and it worked quite well.
Alternatively, you can steal a more generic map from somewhere online. I've used a lot of Dyson Logos maps in my games. I often find this unsatisfying though, as the types of dungeons I can make are limited by the maps that are available. I have to either compromise my vision to suit a map that wasn't created with that vision in mind, or find a map I like first, then come up with the vision after. Neither is ideal.
So, I often create something from scratch. I don't usually bother with detailed maps these days. I've tried using AD&D's Appendix A to randomly generate dungeons. It's really confusing and it creates a map that kind of stinks. I feel the same way about random dungeon generators found online. I've basically given up on it.
So, if a relativistic flowchart-stye map is good enough for the players, it's good enough for me! I start with a bubble in the center of the page (with stairs leading up to the surface), then roll 1d4 to determine the number of exits. For each room branching off of that, I roll d4 again to determine the total number of exits (that is, if I roll a 1, there is only one way into the room, and no other ways out). I do this until I've exhausted my 20-room chunk or end up with only dead ends (to which I might add secret doors to additional rooms to fill out any remaining rooms from my 20-room chunk). I can add additional notes as needed, like "This room is really big and takes two turns to map," or "This hallway is really long and requires a wandering monster check when passing through it."
I tend to put important rooms like monster lairs (those with treasure), "special" rooms, and treasure rooms in those areas farthest from the entrance, so that the characters will have to endure the longest route through the dungeon to reach those rooms. Rooms with only monsters or only traps will be those at important intersections or choke points, ensuring that the characters either go through them or find a long way around them.
After that, it's a matter of writing the key, which I've demonstrated in my rewriting of the room from B1, above. I like to find a middle ground between writing something descriptive but also usable at the table. Highlight interactable elements. Try to anticipate what the players will want to investigate and provide additional detail as needed. Try to think of all the information you might need. Think of what questions the players might ask, and come up with answers. Include game mechanics whenever possible to avoid looking stuff up during the session. Write it and reread it and refine it.
As you run the dungeon, it will change. Make note of which monsters have been killed, which have had changes in attitude towards the PCs, which traps have been disarmed, which treasures found, which doors unlocked, and the like. Once the players make contact with the monsters, feel free to move them around. They should be doing something, not just hanging out in one room all the time.
When the characters leave the dungeon, I like to change things. Monsters which don't like each other may come into conflict, killing or capturing one another. Those that do like each other may form alliances. Their territory may expand as the characters open doors, disarm traps, and kill other monsters. They may recover treasure themselves, find secret doors of which they were previously unaware, and update their own dungeon maps with scouting missions.
This is also when I restock the dungeon. I go through each room the characters have already explored, checking again for monsters, traps, and treasure. Perhaps a new monster has wandered into the dungeon, or wandered up or down from some other level. Perhaps a trap has been reset, or a new trap placed where there was none before. Perhaps the monsters have moved their treasure to a previously unoccupied room for safe keeping.
This can be a boon or a bane to the players. Perhaps the friendly goblins have thwarted a counterattack from the orcs, and now the orcs are lesser in number. Perhaps the orcs triumphed and carried off the goblins, and now the characters have lost a valuable ally. The rival adventurers may have perished in their own quest, or maybe they succeeded and carried off some valuable treasure, or simply made progress and disarmed some trap. Discard any results that don't make sense, or stretch your imagination until they do. No two delves should be exactly alike.
As I've explained, this is much too large a topic to possible cover in its entirety in a single post. Hopefully, I've provided a somewhat comprehensive explanation of my approach to getting started and to doing the long, somewhat tedious work of preparing a location like this for gameplay. Everyone's process will be different. This is a only a peek into mine.
No comments:
Post a Comment