Monday, March 11, 2019

Mastering the Megadungeon Take 2: Tips for 5e

My First Megadungeon

My first attempt at this received some flak for using some Old School D&D blogosphere jargon terms.  I'm going to try to do better about that this time and make it more friendly for people who haven't been reading about these things for a decade now.  What better way to do that than to talk about how to make them for the latest version of D&D?

I am not going to cover why you would ever make a megadungeon in this post, but may handle it later if there is interest.  Instead I'm going to assume you already see the fun in exploring a big, weird space and in making one to be explored.  Based on your hypothetical desire to run this style of game, I'm going to assume you would benefit from a houserule: half xp for monsters but you get 1xp per 1gp of treasure found.  This will, hopefully, encourage players to spend resources going deeper into the dungeon to find treasure instead of trying to play the world's weirdest version of Rainbow Six.

Another assumption I'm making is that you want to follow the traditional megadungeon model where the level of the dungeon is roughly equivalent to the level of the PCs exploring it.  Unfortunately, this is difficult in 5e given the nature of low levels.  A fix I recommend (and that I use in the 5e version of Nightwick Abbey) is to have the first level of the dungeon have challenges for characters levels 1-3, and then each level after that is equal to one character (so dungeon level 2 is for character level 4, 3 is for 5, etc.).  If you have c. 70 rooms on a level this should work fine with the other methods mentioned in this post.

Nightwick Abbey was inspired by films such as the Blind Dead.

As I said in the first post, if you want to run your megadungeon for a long time and not get too bored with it, you're gonna need to have a theme that resonates with you.  The old "built by a mad wizard" potentially allows for a wide variety of potential content but may not grab you enough to work on it in the long term. My two most developed megadungeons (Nightwick Abbey and the Uz undercity) were based on the midcentury horror cinema I liked. A Tolkien fan might want to do a riff on the ruins of the Pits of Utumno or Angband. A fan of classic sci fi may want to do some version of the huge Krell machine in Forbidden Planet. Someone who is more familiar with the works of the big three pulp authors might want to base their dungeons on Mt Voormithadreth or the city in "Red Nails." There are a lot of possibilities but the important thing is that they resonate with you and are broad enough in terms of genre that you can draw from a wide variety of sources.

Once you have determined you're overall theme, you'll need themes for the different levels of your dungeon.  Gygax suggested starting with 6 levels, but I think 3 is plenty and you can get by even with one with the judicious use of something like the Greyhawk Construction Company (in Nightwick I used purple mist). Each level should have a theme, like "the catacombs," "the sunless garden," "the orc spawning pits," "the house of portals," etc. This theme should fit within the broader theme of the dungeon (catacombs in a haunted abbey, for example) but provide a different flavor from the levels adjoining it.  Each sub area should also fit within this theme.

Your New Best Friends

But what the hell do I mean by sub area? Your dungeon is going to be made up of smaller complexes with themes connected together on a level that has a theme in a dungeon that has a theme. M A R Barker, the creator of Tekumel, called these "Saturday Night Specials" - the dungeons within a dungeon.  What I did for Nightwick Abbey, and what i recommend for your first time, is that you generate a level using a 4 x 3 set of dungeon geomorphs like those created by Dyson Logos. His geomorphs and those of his imitators are 10x10 sections of dungeon that all have exits at the same points so that they can easily be connected into larger complexes.  Using Dave's Mapper can help you make a level with very little effort, and you can replace geomorphs that don't fit the theme very easily.  However, you'll most likely need to add the connections to other levels to the map in the forms of staircases, ladders, pits, chutes, etc. I cheated with Nightwick Abbey and made the nature of the dungeon such that space is bent within it and thus the stairs didn't need to match up. You could also draw your own, which I have done with more recent iterations of Nightwick Abbey.

There are of course other ways to create these sub-complexes, but we're going to stick with geomorphs right now because tools like the ones above make them easy to work with.  If you follow my advice, you should have about 12 geomorphs (with the open edged option on Dave's Mapper) to come up with sub area ideas for.  Now some of your complexes might just be "storage" or some similarly vague thing like that that doesn't differentiate that geomorph much from the background themes of the level and the dungeon as a whole.  That's fine, but you'll definitely want at least a few that are very unique.  The more you have generally the better.

But What's Inside a Room?

To stock any dungeon, I use the following algorithm to quiet my body thetans: between 1/3 and 1/2 of the rooms will have monsters in them. This number is taken from old school D&D but I find it works just as well in 5e.  I use a set of monster types including Bosses, Sub Bosses, Grunts, and Mooks.  The CRs of these will be dependent on dungeon level. For every thirty rooms I have 1 boss encounter, 2 sub boss encounters, 3 grunt encounters, and 4 mook encounters.  I've found that 12 geomorphs usually works out to 75-90 rooms, so a given level is likely to have three sets totaling 3 boss encounters, 6 sub boss encounters, 9 grunt encounters, and 12 mook encounters.  I tend to determine these encounters based on the whole level, but one could hypothetically stock every geomorph with a smaller version of this setup; however, I find this produces too many boss encounters, so I use total dungeon rooms to determine how many monsters exist.

There will be a number of rooms with treasure in them equal to the number of rooms with monsters in them.  Half of these will be a room that already has a monster, and half will be rooms where the treasure is merely hidden, trapped, or unguarded.  To determine the total amount of treasure in a dungeon level, I generate treasure as though the level was the lair of each boss on that level.  Typically this means 2-3 bosses. Look at their CR, and then roll in the DMG or pick based on whatever method you prefer the amount of treasure they would get for a creature of that CR.  I recommend then using courtney's treasure document to divide treasure into parcels and make the stacks of money something more interesting.  I then group these parcels into however many hoards I need to sprinkle around the dungeon (again equal to the number of monster encounters) and sprinkle them based on what I think is appropriate.

The map to a metroidvania, Castlevania: Symphony of the Night

I hope to do an example of this kind of stocking soon, but this post is already one of my longest and I still have another topic to cover: Metroidvanias. This is a style of video game that has nonlinear paths but some are only accessible after certain progress is made in the game.  Usually this is in the form of new powers granted to the character that allow them to overcome previous obstacles; however, for D&D megadungeons I recommend just using interesting keys. Have areas on some levels that can't be accessed until a key is found or an object manipulated on a lower level.  This can make having to go through previously cleared areas to get to new dungeon levels more interesting because the players may realize "wait a minute, there's something still to be explored on this level and we just found that weird octagon stone like the hole in that wall."

I love that kind of stuff but it has a potential problem.  The first several times I tried to implement it, the PCs found the key before they found the lock and ended up selling it between delves. This is part of why I specify that the key should be on a lower level than what it unlocks.  It is still possible they'll sell or get rid of a key before using it, not realizing its true value, but sometimes it's more interesting if a door remains secret forever. If it wasn't, why have secrets in the first place?

1 comment:

  1. For the "We sold the key we need" problem, I've had the buyers of the key (or the person the fence themselves sold it to) contact the party with an offer to either sell it back at a higher profit or hire the party to open the hidden level to retrieve some secret treasure. This had two effects: 1.) The party could still explore the sublevel, 2.) they got much more careful about selling potential keys.

    ReplyDelete