Problems that trouble me underground

1. Detailed dungeon prep (e.g. room-by-room topological map with full brief keying) is impractical for Immergleich. There are thirty-odd districts and each has at least one conventional dungeon. Many houses are basically small dungeons. Meanwhile my “maybe do” list is ninety items, and most of those aren’t dungeon mapping.
2. When I do some dungeon prep, I often feel that I’m getting very low ROI, especially given the dungeon may never be entered. I’m not confident that I’m using my time well — if I have thirty minutes, I want to spend it in a high-yield way, not pacing out three miserable rooms¹.
3. I am mighty in improvising. I see places in my head and you can follow my voice into them and be metatasized by a slime. But when I do this, I worry that:
   • I’m using only a narrow subset of the possibilities of dungeons.
   • I’m arbitrarily deciding how large the dungeon is, where important things (e.g. quest objects) are. It is harder to see where player skill fits into this, at least player skill of the dungeoneering kind valorised by the OSR.
   • I am also haunted by the Quantum Ogre. I am not sure I care about him, but i see him crouching over there by the old fireplace.²
4. Because players often enter dungeons in response to jobs from the Immergleich job list, I often need to seed job goals into dungeon spaces.

Goals that I have formed in the dark, like mushrooms or a weird egg

1. Substantially reduce the prep time for dungeons and the like (especially to support new jobs) …
2. … while maintaining the variety, balance, structure etc that careful planning and pre-planning can provide.
3. Provide a “fractal structure” that guides me to use my prep time well. One that can do that this given profound uncertainty about what the PCs will do — I want explicit processes that guide me to spread my time well over all the places they are likely to go.
   • Specifically, I want to be able to “Level 1 prep” a dungeon in (say) three minutes, then come back and expand to “Level 2 prep” in fifteen, without having wasted much of the three minutes. And so on, until at a hypothetical level 7 I am drawing individual bricks without having to waste any of the eighteen hours I spent terrain-typing 5′ squares.
4. Give me a principled approach to what I am doing. Because I am all about the principles. I want to have an explicit practice that I can declare, e.g. if players ask how I roll.

Resources that I have found beneath a filth-besmirched altar

There is some material on the Internet concerning how to generate dungeons. Much of it seems indifferent in value, either generally or for me, now. It is not clear what it gives me over just pouring out my mind onto the page.

There are some that stand out, both in that they do have clear USPs and they are relevant to my particular desires:

• The “dungeon as a monster” model from Dark Heart of the Dreamer (free download), or the similar but more developed one from The Perilous Wilds (this part is free)
• The Angry DM’s “Zone Exploration” model
• “Matthias” on enworld — Representing a dungeon as an abstract track

I’m keen to here about your experiences with these, or with any further examples.

<sup>1. I like my rooms dungeon rooms miserable. Pleasant, airy rooms are ill-fitting to my goals.
2. I fear I need to read and think more about the ogre. A lot more. In particular, I need to re-read and analyse the article series I link to above. I imagine the ogre will wait.</sup>