Why Build a Megadungeon in 2017?

It's 2017, and here I am playing with maps for doing another megadungeon. (Not the one in the picture, which is from the Blackmoor maps in The First Fantasy Campaign by Dave Arneson.) Which brings me to the fundamental question: why build a megadungeon in 2017?

When I got married, I was reading some books on wedding preparation. There were a lot of topics they talked about, but one was whether or not to rent a tuxedo. And it came down to the question: do you really want to get married wearing someone else's pants? It was a weird moment but it put the whole rental idea in an interesting light. I wound up wearing my own pants, from the suit I would still wear to a wedding or funeral today.

That colors my thinking about dungeons. Running someone else's megadungeon is a weird and slightly personal thing, and it just seems off to me. And I'm not the only person who thought this way; Gary Gygax and his cohort at TSR thought so as well. It took them four years to publish G1 Steading of the Hill Giant Chief, because they figured that referees wanted to design their own dungeons. TSR wouldn't publish a portion of a proper megadungeon until the Forgotten Realms boxed set Ruins of Undermountain in 1991, long after Gary had been pushed out. (A rather disappointing effort at Castle Greyhawk was published the year before.)

For me, the megadungeon is all about the design process. The original advice that Gary Gygax gave was to "construct at least three levels at once," and said that a good dungeon would always have "new levels under construction so that players will never grow tired of it." The megadungeon, then, is the original cure for the referee's interest flagging. Castle Greyhawk was known for the riotous diversity of its levels, and there is no reason you can't just grab whatever module or idea your favorite creators have come up with lately, steal its ideas, and put them in your megadungeon.

The process centers around vision and re-vision. You need three or four factions to start with; maybe the PCs ally with the gnomes and wipe the kobolds out. Then you invent (or steal) a race of fungoid men, and design a whole set of caves around them. After that you watch the new Alien movie and decide that you want to do something with that, so you have a buried spaceship with a homicidal xenomorph analog. Then you decide to move on to Vikings. And you can do it all within one megadungeon.

You could, of course, just do each of those as a separate adventure by putting them on a hex map and having NPCs come up to the PCs with juicy dungeon leads, or let them discover each new area as they do a larger hex crawl. But with the megadungeon they have a much better chance of organically coming upon the nice juicy bits. Let's say that the PCs discover the hidden wreck of the starship while they're trying to find a way around the fungoid cavern; they now have an important choice about which of the threats they want to face. Good megadungeon design always involves that choice between sub-optimal paths.

The "living dungeon" also means you can always fix your mistakes (well, except for TPKs). If you stick a bunch of undead in a sub-level and it turns into a bit of a grind, you can always have some carrion crawlers come through and eat them, and now have a new threat lurking the halls. The PCs might find that the goblins were a pushover, but with the goblins are gone the trolls are expanding. The megadungeon always has something fresh to throw at PCs.

Megadungeons of course also have that part that doesn't change. I particularly like this when you have things like the "goblin market" or the established neutral / allied factions that the PCs don't usually get violent with. This lets you take one of the advantages of a fantasy city and put it underground. All the neutral monsters aren't in OD&D by accident or to pad the page count; they're meant to be there as encounters that can go any which way.

Another reason to stick with the megadungeon is that the dungeon stocking rules in old D&D were really, really good. They put things at the optimal density for an exploration-focused game: there are enough empty rooms that most paths won't feel like a grind, but enough nasty stuff comes up to keep the players on their toes. Designing small dungeons using the rules in OD&D vol. 3 or Moldvay (the two best books for dungeon design) feels empty, and there is always the desire to put a "boss" or a "prize" at the end. There is some point in going down into the Upper Lowlands Dungeon of Death™, after all. Whereas level 5A of your megadungeon gets explored purely because it's level 5A.

(Incidentally, that's a mistake in Rappan Athuk: megadungeons don't have a boss monster.)

The other best reason for megadungeons, and this is something I decided after dropping in to Eric Hoffman's excellent B/X game for the second time in a long time yesterday, is that they are ideal for open table type games. This is hardly an accident, as the megadungeon grew up around this style of play, and it's mirrored in OD&D's suggestion that "four to fifty" players can be in a campaign. A tentpole megadungeon is a structure built for being able to throw out a notice, "I'm running D&D," and having a game to run in a jiffy. You show up and you go down to the deepest level anybody knows about, and see what they can find there.

Finally, if the players grow tired of the megadungeon per se, they can go somewhere else for a while. The outdoor random encounter tables are probably unsafe for low level PCs, but all of the classic megadungeon campaigns involved wandering about in the wilderness. Heck, you can even find other entrances to the megadungeon itself.

But no matter how many megadungeons there are in print, at the end of the day I think they're worth making on their own. Because, after all, you don't want to run D&D in someone else's pants, do you?