I was reading the Alexandrian post about the 5th Edition D&D announcement and was struck by a couple of commenters saying they wanted to see a guide to dungeon creation; that no such thing yet exists. This is a real shame, since one of the primary tasks of a DM is creating environments for the PCs to play in, and making a fun and useful one is hardly an intuitive process. Well, allow me to attempt to fill the need to some tiny extent.
Keep in mind that in this context, “dungeon” doesn’t just refer to an underground fortified construction; it’s the term for any set-piece area that the players explore on a local scale, often with a relatively high density of challenges for them to overcome and rewards for them to collect. “Dungeons” are usually clearly separated from the outside world in some way and subdivided into units – rooms, halls, open spaces, filled spaces. That said, dungeons can have a variety of purposes, and these should be the primary consideration when you design each one.
What kind of play is the party interested in? Exploration? Tactical combat? Puzzle-solving? Do they like short, fierce encounters with downtime in between, long running battles, or carefully-planned surgical strikes? Do they want to sneak past enemies or negotiate with them instead of fighting at all?
And nearly as importantly, what is the function of the dungeon in the campaign world? A temple or other religious location would be relatively accessible for its worshipers, even if well-hidden, and most will be open, with the construction focusing on central holy spots and relatively large public areas. A military installation, on the other hand, would be better defended, not only by its inhabitants but also as much as possible by its construction and location within the landscape. A animal’s home will be simple, with greater numbers of exits if it supports a large colony or if its inhabitants have natural predators. And so on.
Suddenly, a Dungeon!
Sometimes, a random encounter table produces something that would logically have a home base, and you need to create one on the fly. Sometimes, the party wants to do something that you’re not ready for yet and you need a quick distraction that you can pass of as a random encounter. Either way, there may be times when you want to design a dungeon in short order. I suspect that your best friend here is randomness. Need a floor plan or map? Do an image search. Find an online dungeon generator. Zak S. has some excellent advice on whipping up a dungeon in short order, or just making a really creative one in your spare time.
Let’s say you’re not online, though, and you need a dungeon right now. It’s time to break out the dice and let randomness guide you. I’m going to include an appendix at the end of this with an example of using dice rolls to create a dungeon on the fly, but for now think about some aspects of a dungeon that could be whipped up that way. In the meantime, there are some elements of design that should inform any dungeon, whether you’re making it up as you go or crafting it lovingly before you even have a campaign to use it in.
General Guidelines
1. Know the dungeon. Is this a natural cave? An animal’s den? If it’s an artificial structure, what is the culture of the people who made it… and of anyone that’s lived there since? This isn’t something you need to tell your players directly, but knowing a dungeon’s motif makes it easier to add detail on the fly, and can give the whole thing verisimilitude. Little details like the tools being made of bronze instead of iron, or the walls being decorated with a particular art style, can go a long way toward full player immersion in the world.
There’s more to this than just coherent flavor text, though: knowing a dungeon helps you draw the floor plan. If it’s a kind of building or locale that exists in real life, look at those for inspiration; otherwise think about what design choices the builders would have made. A place made with defense in mind will have narrower halls and more chokepoints and death-trap rooms (think of the inside of a gatehouse) than a place made mostly for living or practicing a trade in. Poor builders will leave more plain construction and smaller spaces than rich ones.
Knowing a dungeon also tells you what kind of challenges the players are likely to face in it. The tombs of wealthy kings will have lots of durable traps; well-known ones may have rival treasure-hunters or signs of previous incursions. Outlaw hideouts, on the other hand, are more likely to have lookouts, alarms, and hostile inhabitants.
2. Make each one interesting. Each dungeon, whether it’s a one-room cave with a bear inside or a sprawling high-tech compound full of guards and environmental defenses, should be distinct from the others the PCs – and the players – have encountered. In part this should follow from rule 1: if each dungeon has its own unique character guiding its setup in your mind, then the cumulative effect of all the little details, from mere wall decoration to the forms the challenges and rewards take, can add up to something memorable.
Beyond this, of course, it can be nice to go out of your way to give each dungeon a special feature. It’s divided into halves by a narrow passage or precarious walkway, allowing the inhabitants to set up a defensive chokepoint. It seals itself shut at intervals, limiting both intrusion and exit. Its only defense is one massive, elaborate trap. Or perhaps it’s full of tiny, harmless, infuriating booby-traps. It’s built around a central pit, pillar or other structure that allows for non-standard ways of traveling through it. It’s home to a foe or hazard that could wipe out the entire party if they don’t plan properly – and signs that that danger lies in wait.
That said, don’t worry if the “memorable thing” you planned isn’t what the party remembers. Maybe what stays with them longest is the random-encounter NPC they accosted to ask for directions, or the loot they came away with. That’s okay. But make sure that there’s something you can point to that distinguishes each of your set pieces from the others.
3. Put in useless things. In the service of both world verisimilitude in general and rule 2 in particular, you should be able to tell the players things about the dungeon that don’t give them XP, don’t have any obvious use in combat, don’t give bonuses to stats, can’t be easily sold. Put in things that are just there to be interesting, or mysterious, or realistic. This doesn’t mean that you need to calculate the population of each dungeon and make sure that there’s an appropriate number of chamber-pots. It does mean that if the players find a chamber-pot, they think of it as a piece of local color rather than vital key to some part of the plot to be investigated and taken.
I include here something that could also have been in rule 2: one of the tools that makes a dungeon unique and memorable is item or treasure that the players can find there, and nowhere else. This could be a plot-vital resource or McGuffin, or just a unique bit of flavor text: the treasure stash contains rare red pearls, for example.
4. Put in important things. Part of the appeal of RPGs is the feeling of progress and gain. If all they do is go into a place, kill some enemies, and then go to another place, no amount of colorful description on your part can keep the game fresh for very long. Every location should have something in it that makes the PCs glad they went, or interested in going back: a mystery, a puzzle, a tool, a toy, treasure, allies, information. (By “toy” I mean anything intriguing that the party can have fun fiddling with even if it isn’t a mystery or puzzle, such as a stage, a party hall, or just a nice fishing spot.)
5. It shouldn’t be too big. It shouldn’t be too small either, of course. But it’s better to leave the players wanting more than to make gameplay feel like drudge work. And it’s both easier and less wasteful to add new content – Oh look, a secret door that you missed earlier!; Oh look, a sinkhole opened up in the ground! – than to try to prune off areas that the PCs haven’t explored or noticed yet. When in doubt, leave things a little small; the worst that can happen is you’ll finish the dungeon earlier than expected.
This has gone a little longer than I had anticipated, so look for a part two in a few days with a process (hurrah more ordered lists!) and some examples.