What's your approach to dungeon-building?

[Thanee]

I would start with the idea why the dungeon is there and what it represents.
Also, why the party would come there.

Then based on this basic idea, either start with a map and then fill in encounters, or start with encounters to place somewhere in the dungeon and then draw up a suitable map.

That is obviously the hard part.

Bye
Thanee

 

[Hand of Evil]

There's my problem! I'm just randomly drawing a map on graph with no real rhyme or reason. I need encounters first to define the dungeon.

There is nothing wrong with that - you just have to concept them, naming them helps out.




oh some maps (if interested)

http://www.enworld.org/showthread.php?t=101452

 

[cmanos]

come up with a plot, a villian or a treasure, then build trhe setting around that. About 25% of my adventures consist of dungeon crawls.

 

[The_Gneech]

I start with what the story calls for, figure out the Main Event for the scenario (i.e., "rescue the princess," "encounter the beholder," or "the hallway of infinite gelatinous cubes") and put that in place, then flesh out the rest of the map with more or less random encounters that fit the overall theme. (Thus, if it's a sewer, I'll drop in rats, oozes, crocodiles, etc., or if its a trek up the side of a mountain I'll put in displacer beasts, wyverns, and owlbears as appropriate.)

Certain recurring motifs I use a lot are:

* Displacer beasts, hook horrors, and orcs are my favorite monsters, apparently, with minotaurs, bugbears, and beholders getting an honorable mention

* Arcane experiments generally go horribly wrong, leaving dungeons full of dangerous encounters and peculiar situations -- and dead wizards

* Often there's an encounter that's way beyond the abilities of the group, but which can be fairly easily avoided or escaped if the players keep their heads and don't go rushing in

* Usually, all or almost all of a single "dungeon" (even if it's multiple levels) will fit on a single sheet of 5x5 graph paper; any given scenario may have from one to four such "dungeons," depending on the complexity of the scenario.

* Traps are rare, but when they appear, they tend to be vicious

-The Gneech

 

[Mallus]

Just say no.

 

[Darkness]

Ideally, I'd do it about like this:



Start at the end - generally just the very end itself. Create the final scene - including opponents' tactics! - in loving detail. Oh, and the important parts of the treasure, I guess.

Make a list of what locations/events/whatever I absolutely need or really want to include.

Create the beginning - generally a small to medium area.

Figure out the basic theme of a few general routes the PCs can take to reach their destination.

Decide where my needed/wanted locations are relative to these routes.

Start to include ways the PCs can learn about this general layout if they try to ("West seems frozen, Southeast smells bad" or "troll caves to the East, dragon-worshipping kobold cult to the North"), so they have to make an actual decision, rather than randomly choosing one of a number of ways they know - and care - nothing about.

Complete the details of these routes and look what side-areas need (or would be cool) to be created.

Create those side-areas. (Actually, I will often create some of them while still detailing the areas they're near, depending on inspiration.)

Check over the entire thing - especially NPC stats - and, where necessary, modify things to make more sense to me and/or potentially be more interesting for the players. Further, add additional things that can make things more interesting for the players (e.g., side quests).

 

[jgsugden]

Step 1: Decide why it is a dungeon. What is the purpose of the location? Was it built? Found?

Step 2: Go back, in my mind, to when the location was created. If it is natural, how would it have developed? If it was built, what would the builders have wanted to put in place? Do rough diagrams of the place based upon these thoughts.

Step 3: Figure out who lives there currently. Don't drop anything into place yet, just figure out who the current residents are going to be.

Step 4: Bridge the gap between the creation fo the place and the present. Figure out how the place would have changed over the years. Were there different residents that modified it? Were there natural disasters? Modify the place to accomodate these changes.

Step 5: Drop in the current residents and figure out how they'd personalize the place.

Step 6: Send in some adventurers that miss out on 80% of the details you've added. ;-)

 

[Celebrim]

1) Decide what the mission is. Why are the PC's going somewhere?
2) Decide where the mission will take place. Where is it logical for this mission to occur? Is there anyway to make such a place more fantastic without blowing believability and versimulitude?
3) Decide how built the place where the mission is going to take place? How ancient is it? What is it's history?
4) Who lives in the dungeon now? Why are they there? How do they survive?
5) How has the dungeon aged over time.
6) What is an appropriate challenge for the PC's? How can I scale the challenge of the dungeon dwellers to fit the PC's capabilities. If I can't find a suitable answer, either shelve this plan for later and go back to #2, or else go back to #4 and think of similar but more appropriate challenge.
7) Map the dungeon, taking care to make each room interesting and functional. Generally this means small dungeons (15 rooms and under) are preferred to big ones.
8) Add stat blocks, treasure, and brief description. Put a couple of skill checks in each room.

 

[Wombat]

Keep them small

Keep them architecturally logical

Whenever possible, base them on real designs

Have them as a background TO the story, rather than BEING the story

Don't have 30 orcs camping out in a small room with no idea as to what is going on 30' away

 

[LordBOB]

location location location


Weither in the middle of a cliff, the underside of an old Abondaned house, or below a city

Make some dengeons sturdy and fairly new and others weak and about to cave in. Also the inside of a dungeon is important. What is looks like and smells are always fun to play with