D&D 5E On Dungeon Design

[Lanefan]

One thing that often gets overlooked by dungeon designers is that the choice-making should start before the party even gets inside: multiple entry points.

Even a small keep can have a main door/portcullis, a postern door, a trap door in the roof*, and a high window or two that can be climbed to.

* - which of course the party won't see unless they think to climb up there or fly over.

Typical big underground dungeons never seem to have enough entrances/exits to the outdoors, never enough connections between internal levels, and almost unheard of are connections between levels that skip past other levels.

The party should always map. If they don't, it's on them when they get lost.

Lan-"and a with-it DM (unlike me) always knows who is carrying the map for when the fireball hits"-efan

 

[Gilladian]

Also, beyond the good advice you've gotten here, I recommend researching the "five room dungeon" concept. Makes for short, sweet, to the point scenarios of many kinds.

 

[Ratskinner]

While you've got some good advice on this thread, have you considered just skipping dungeons? I've run plenty of fine campaigns with nary a large underground complex to be found. I find most of the "dungeons" feel rather contrived to me, so I tend to not focus on that type of design.

When I do include dungeons (or any terrain-intensive scenes) I like to cut out geomorphs or even specific tiles from card stock. Its cheap and you can use freeware like Inkscape to customize them.

 

[Authweight]

I absolutely know what you're talking about. Here's what I recommend for designing dungeons (not a universal "this is how you do it," but rather a style that works well for me):

1) Instead of thinking of dungeons as connected rooms, think of them as connected sections. Each section should have a number of rooms that go together conceptually and spatially. Instead of building encounters in rooms, build encounters that span whole sections of the dungeon, and have the enemies in each section react somewhat intelligently to the intruders. This means you can build relatively mundane areas of castle/cave/whatever that are still interesting and relevant as potential combat areas.

2) Make each section interesting. If it's a fight, have something that can be used in a fight there. If you are just exploring, give them a reason to look around there beyond "well, maybe there's something here." If nothing else, give it an interesting description that helps convey tone, atmosphere, story details, or interesting lore. There should be no "pointless" areas of the dungeon. Think of each section as being a scene in a movie. Give the players a prompt for each scene - either something that happens to begin the action, or a detail that is immediately compelling for them to explore. Try not to just describe a bunch of stuff and then sit there and ask, "so what do you do now?" without being confident that your players will have an immediate response (unless your goal is to present a difficult situation and make them consider how they want to tackle it - in this case there should be a clear objective instead of a clear course of action. Really good dungeon design emphasizes these situations, but honestly you don't need to push these all the time).

3) Don't bother with having the party map. Assume that the party has a good enough sense of direction to get back where it came from. They are professionals at this, after all. If the dungeon is truly vast, handwave travel and/or have them make checks to not get lost.

4) At all times, have in mind what the point of each area is. A point doesn't have to be big - it can be as simple as "establish a small detail to make the environment feel more real" - but keep this point in mind, because it should drive the way you run the scene at the table.

 

[SilentBoba]

I just started playing Dragon Age: Inquisition. At one point I was exploring a few rooms of a castle that looked reasonably realistic as a design. Aside from one big knot of defenders and a few outlying ranged attackers, there was nothing of interest in the castle. Or perhaps I should say the castle interiors were lovingly rendered by the game designers, but aside from pinging my search sonar, I took very little time to note any particular marks of distinction in the place other than a quiet nod to the overall trappings and verisimilitude. The battle itself was fun, and raged through several of the adjacent rooms in a continuous fight, but once the fireworks stopped, they were just rooms that I traversed at a jog.

Let your PC's do the same.

If they insist on details, give them some, but point out that time is marching on as they painstakingly search. Or, if there is something to find, either use passive searching if it can be entirely overlooked, or telegraph the "secret" door with some particularly detailed description. Let the optional secret doors be found when the PC's get flanked by the second group of bugbears coming out of the one behind them.

Oh, and try to give the PC's a reason to not only be there, but to get THROUGH the dungeon in a reasonable timeframe. That will hopefully minimize the expectation for excruciating furniture and inventory cataloging.

 

[Heavy3p0]

I have the same issue where i don't have the time to create grand adventures and sprawling dungeon maps anymore. recently I've developed the philosophy of making my dungeons little more than flow charts and never very large. i will connect a half dozen or so 'flow chart rooms', each typically containing an encounter, trap, or requiring a skill check (with the occasional room requiring nothing) and connecting to one or more 'rooms' in the flow chart. i just ad-lib my way through the descriptions as the players wander through the flow chart. essentially anything that would hinder the parties progress through the dungeon (like a locked door) becomes a 'room' on the chart that the players will have to deal with when they arrive at that portion of the dungeon chart. If they players request to see a portion of the dungeon in more detail i just make up a sketch of the item in question.

 

[dd.stevenson]

I suck at designing dungeons.

Player-drawn mapping is a fairly specific play style that only really sings when combined with a certain dungeon design philosophy and specific D&D rulesets. Just like XP for gold, it ain't for everybody.

Anyway, it sounds like the weak link is how you run dungeons, if I'm to judge by your post, since you seem to hit the same problems whether it's something you wrote or not. I've recommended a DMing exercise in the past, and gotten good results from it. Maybe it'll work for you.

First, sit down with a few rooms of of differing types in a well-respected published adventure.

1) Describe the rooms as if you were spoon-feeding the room to players in a way to maximize tension. Out loud. To yourself. Imagine the player's responses, and continue spoon feeding as they interact with the environment. Note every decision point, and be sure to spoon feed them the info they need to make a decision.
2) Describe the rooms as if the PCs were fleeing pell-mell through the rooms, and only have time for a brief glance at each place. Again, note the decision points and be sure give them the info you want them to have in the most frantic way you can.
3) Describe the rooms as if the players were showing signs of boredom, and you really just need to get through it and onto something that interests them more. Again, note the decision points and give them the info they need in the most incisive, down to business way you can.

There are other techniques to try as well, but IME these are the "big three" dungeon DMing skills. Now, for added insight, you can try to find someone pod casting or you tubing the same published content and see what they do that's different than you.