Hi, Jonathan here. I was asked to share my techniques for Caverns of Carthage dungeon creation and here they are. I don't claim to be an OSR expert or an expert on writing dungeons. I'm an old hand at RPGs, but relatively new to Old School. I've written 5 dungeons so far and players seem to be enjoying them (or, at least, finding them challenging). So here's the secret sauce for those of you that are curious or want some inspiration.
To create a new dungeon I follow these 8 steps:
I'll go over each point in more detail in the following sections. Note that I often need to iterate through the steps multiple times (especially steps 1 to 7), completing a bit more each time I go through them, before I commit to formalizing a final result in step 8. I'm not afraid to set aside or throw away ideas that aren't working.
I find that every great dungeon has some kind of central conceit, an interesting story that explains its existence, unique monsters, traps, etc. A completely random dungeon can still be fun to play in (e.g. see B/X Misadventures in Randomly Generated Dungeons). However, as Gus L suggests, if every room is unrelated to every other room then the dungeon might be bad. If the dungeon has some kind of unified theme and raison d'être then it both makes it easier for me to write the dungeon and more satisfying for my players to explore it.
Hence, the first thing I do is come up with some seed ideas for my dungeon. Here are some sample questions I consider:
I don't have to answer all the above questions all at once. But whatever ideas I come up with now as core concepts for the dungeon will help me make decisions later.
For example, for one of my dungeons I brainstormed that I wanted it to be a former museum that displayed magical items. The items would be clearly visible but protected by fiendish traps or other defenses. I wanted there to be some secret doors leading to some kind of central administrative room. I imagined a giant diorama filled with treasure, and some kind of golem that smashed pennies flat to imprint the museum's logo. For me, this was enough of a starting point to move on to the next step.
Dungeon Level: it's a staple of the genre that dungeons have a difficulty level and that the difficulty increases as you delve more deeply. Caverns of Carthage has a greatly squashed power curve (characters can rise to an equivalent of OD&D level 3 at their strongest), so my dungeons typically have a level from 1 to 3 as well. I find it can be useful to establish the dungeon level early on because it helps me make decisions later regarding how strong the monsters should be and how much treasure should be available. As a caveat, I see the dungeon level as a rule of thumb and not a straitjacket. While I'm not a believer in Anti-Dungeons, I do agree that dungeon dangers and rewards shouldn't be perfectly evenly distributed or this would be bland. I find it really fun (and challenging) to include some over-leveled critters!
This initial brainstorming step is potentially the hardest part of the whole process. I see it as what's going to make or break my dungeon, and distinguish a masterpiece from a humdrum dungeon the players will quickly forget. I keep brainstorming until I feel I've got some great ideas and I feel excited to start fleshing things out.
Ideally, I perform my brainstorming while I'm on vacation or when I'm otherwise in a carefree mood. Since carefree moods are a rare thing for me (due to work and parenting) I use additional techniques like drawing and meditation to open my mind and unleash my creativity. For example, if I'm stuck then this great Drawing Meditation video combines both drawing and meditation and really helps put me in the right creative mood. I call this mood an open mode, as opposed to the closed mode I usually find myself in.
When I'm in a closed mode, I'm ready to analyze and solve problems quickly and efficiently. However, my best creativity comes from the open mode, where I'm more whimsical, my attention wanders easily, and my brain makes more unusual connections. I think these are both natural modes for the mind and brain that everyone is capable of. The human mind intuitively adopts closed or open modes depending on what it needs to do at the moment. It's not hard to switch between these modes but in our crazy busy modern world people tend to get stuck in the closed mode and forget that the open mode even exists!
I watched a really great presentation by John Cleese on this topic of how to be creative. I can't share the presentation directly, but here's a synopsis: the major difference between creative and uncreative people (aside from things they can't do anything about like their upbringing and education) is their ability to reliably go from the closed mode to the open mode. Usually, creative people set up a process that allows them to do this. It typically means setting aside a regular space and time for creativity and using the right techniques to open their minds. Everyone can do this. The right techniques for you are whatever work for you. Experiment with different options until you find something that works.
I listened to a live talk by David Usher, where he basically completely agreed with John Cleese and gave very similar advice (with the additional advice that playing an instrument takes a lot of practice and that he didn't need to practice anymore—LOL).
In my own life, I set aside every day from 11 PM to midnight for creativity. I'm tired after a long day and this helps to push my mind out of the closed mode I needed all day long to solve the endless series of problems that fill my days. If my mind is still not transitioning then I do a bit of drawing about the topic I'm interested in today or watch the Drawing Meditation video. Within 15 minutes I'm ready to go!
This above technique works great for me but it might not work for you. Maybe you're more creative in the morning? Maybe exercise or yoga work better to open your mind? Do whatever works for you.
For Caverns of Carthage, the touch points I want to see in dungeons are always (1) mythology, (2) a cool moment in history, and (3) something quirky. I aim to include at least one of each of these things in every dungeon. I call this process mismatching because I brainstorm at least one element in each category and then throw them together to see how they combine with the ideas I've already come up with. I find that the more ideas I come up with, the more interesting the connections can be.
For example, for one of my dungeons I brainstormed: (1) a Fountain of Youth, (2) a 17th century Protestant revolt, and (3) a civilized dragon and an undead plague. I threw all these concepts together to come up with an origin story for the dungeon, which then informed all my other choices. In real life history, the mercenaries from the Protestant revolt were trying to sap the walls of the city they had under siege. Well, what if they were using gunpowder to dig through the dungeon, thinking they were going under a city wall? Those explosions were making a lot of noise! What if this was making the dragon angry and she wanted it to stop? She'd offer adventurers some of her treasure if they'd make the noise stop so she could get some sleep. What if the mercenaries discovered the fountain? They'd probably want to guard it and keep it for themselves. Or maybe the dragon was guarding the fountain and using it to heal herself, making herself invincible? But there was a secret door into her lair even she didn't know about, and it would be possible to roll in some gunpowder barrels and blow up the fountain (and dragon)? These could all be past stories that explain the current state of the dungeon, or they could be potential interactions for the players to discover.
If you're trying to adapt my technique then you'll need to figure out what themes or touch points are important to you and make sure they're in your dungeon. Maybe you like body horror, and there should always be mutants? Maybe you want a dragon and a wizard in every dungeon? If you're not sure, go back to the previous step and start brainstorming your thematic touch points.
I find that drawing greatly improves the quality of my ideas. I'm not particularly good at it, but it doesn't matter. It helps me develop nascent ideas and come up with new ones. I'll draw out whatever cool ideas I have so far, and maybe sketch out a part of the map. I use a pencil and either blank paper or graph paper. Sometimes, I use lined paper. I usually regret it.
For this step I recommend using Donjon (actually, it's
the only tool of its kind that I'm aware of). I play with the
settings until it's producing a dungeon I find interesting for what I'm
aiming for. A good recommendation
from Nick LS Whelan is to use option Map Style of
GraphPaper. This uses less printer ink and leaves you more
blank space to write notes.
I use the Donjon map as a starting point. I like it better than starting from a blank page, which is hard to spring from. This is where I apply my previous brainstorming and drawings. They serve as great inspiration and provide me with direction and focus. I might end up using only sections of the Donjon map. I might generate another Donjon map and frankenstein them together. I impose my vision on the map, modifying as I see fit, adding new rooms, removing rooms that don't work, adding secret doors, adding connections, and so on.
I use the pre-generated doors, traps, and encounters as inspiration. Sometimes I keep them mostly as-is, sometimes I change the types of monsters or traps, sometimes I do something completely different.
I think through how PCs might traverse through the map and consider that some ways might be easier than others. Probably (but not always!) the secret or clever path might be easier or more rewarding for the PCs to traverse.
Jaquaysing: an important goal for me is to have sufficient jaquaysing. I will ruthlessly add extra entrances and connections until the dungeon seems sufficiently non-linear. Having at least two obvious entrances and an extra secret entrance is a good rule of thumb, and there need to be lots of different ways of navigating through the dungeon itself. I find this promotes player agency and it also makes gameplay more interesting for me as a Referee.
Resting Place: I try to add at least one quiet spot in the dungeon, which might be a good place to rest. In easier dungeons this spot might be easier to find, whereas it's probably a secret in harder dungeons. When I played Pool of Radiance as a kid, I remember always being so happy when I could find that resting spot in each dungeon. It was an achievement and also a milestone because it would enable more readily exploring the dungeon without always needing to retreat all the way back to town (and trying to survive all the random encounters on the way back).
By this stage, my brainstorming has usually filled out only some of the dungeon and left large parts of it blank. I have a formal dungeon restocking procedure I use for after PCs have depleted a dungeon, to introduce new stuff and maintain the dungeon as a living and dangerous space. I can also use this stocking procedure now to fill in anything that's missing.
Random Encounter Table: I can fill in the random encounter table based on monsters I've pre-populated in the dungeon rooms. I usually go for a simple table with 6 options and using a 1d6 roll. Early in the design process, there are probably only a few entries in the table. One of the nice things about Caverns of Carthage is that I don't have to care if everything in the table actually exists in the map. Chaos can serve up new monsters on a whim, so I can have some fun adding creatures to the table that don't have a lair on the map. At this stage, I don't need to include monster stats, I just have to name them and have an idea of what they're about.
Alternate Method for Encounter Table: for random encounters, many people might recommend rolling multiple dice to get a bell-curve—but I feel averse to this. I kind of want each potential encounter to have some reasonable chance of occurring, otherwise why bother? If you disagree with me, Nick LS Whelan has a radically different and interesting way of Structuring Encounter Tables that you might like better. In particular, I like his idea for adding recurring characters to the encounter list. Personally, as the PCs encounter creatures they might re-encounter later, I just roll a bigger die and if I roll above the highest listed entry then it's an encounter with a recurring character.
Chaos Events: Caverns of Carthage also has the concept of Chaos Events, permanent or semi-permanent changes that can happen to the dungeon over time based on the whims of Chaos. At this stage I'll spend some time brainstorming ideas that fit with the theme of the dungeon.
For traps, I use Gary Gygax's World Builder (p.113). It's got a big list of traps I can roll on randomly for inspiration. It's the most useful thing in the entire book, and it's not even in the table of contents!
By this stage I probably have at least two different kinds of intelligent creatures in my dungeon. If I don't, now's a good time to add some more! I think about how these factions might interact with one another. Dungeons are usually fairly closed environments where there aren't unlimited resources. If the polities are competing for basic resources like food then this could be an obvious source of conflict. Alternatively, perhaps the main resources are plentiful, or each faction is roughly equal in strength, and there is peace among the factions. What would it take to break the equilibrium?
Here are some interesting questions to consider for each faction:
I don't think it's essential to answer all of these questions. I answer whatever I can and move on.
After the dungeon is looking complete, in terms of overall content, I'll start inking in the details. It's important to me to do this last because I find that once I start doing this it transfers my mind to a more closed way of thinking and I don't come up with creative ideas as easily anymore.
I'll write down formal stats for the various monsters, traps, treasures, etc. in the dungeon. I'll do research like reading books and Wikipedia to get various details of mythology or history right. I'll check to make sure the amount of treasure in the dungeon fits its difficulty level (that it's in the right ballpark, anyway).
Personally, I type everything up and make a manuscript. This takes a lot of time, but the cleanliness makes it easier for me to find what I'm looking for during the game session. It would also be OK to just re-draw the map and nicely write out a note for each room on that map, plus add some monster stats and encounter tables. I've definitely had people recommend the simpler method to me.
I think I still have a lot to learn about dungeon creation.
Here are some sources of information that inspired my learning journey and that I highly recommend: