Streamlining My Megadungeon Design

I like to have things simple and precise. A megadungeon would seem to defy that, but I'm going to try anyway. Here are some things I'm thinking as I pound out a draft of the place.

I'm not going to describe each door. For most doors, I'm going to leave it to the GM to determine the nature of the door. I'm thinking that the beginning of the book could have 6 or 12 options for doors you could encounter. For example:

(1-2) Door is wide open. It looks like someone else secured the door with stakes to keep it from shutting; (3-5) The door is shut, but is otherwise unremarkable; (6-7) the door is shut and swollen, requiring a might check to open; (8-9) the door is shut and locked, requiring a might or thievery check to open; (10-11) the door is locked and trapped; (12) the door is magically sealed.

Any time you encounter a door, either roll or pick one. Or, use likelihoods starting from the beginning of the list and moving down. If it indicates that it is trapped, then roll for the type of trap present (and this could be a magical trap); if it is magically sealed, the GM can decide whether it needs a key, a password, or a dispel to open it. It could be a key that he's already given the party (they started with a magic key that opens several special doors) or it could be a whole side quest the GM adds. I want there to be flexibility for GMs to personalize the dungeon without asking them to do much prep work. It should be a zero-prep experience - you can improvise how and why it works using imagination, the rules for likelihoods, and your own personal goals for the dungeon experience.

This still leaves open the possibility of special doors throughout. I could decide that a door is a monster in disguise, works using a series of intricate mathematical locking mechanisms, or can only be opened when the statue on the other side of the room has been turned to look away from it. The presence of the random options doesn't prohibit other inventive doors, but it keeps me from having to write descriptive text for every relatively common type of door one might encounter. The same is true for traps, pools, statues... there can be statues that do common things, and special statues once in a while that do special things.

In terms of encounter areas, I like the number ten (although maybe twelve works better because of the game...). If each area has a fixed number of encounters, I leave room for open spaces and for the GM to be creative. In looking around, I found a thread with this advice by Gary Gygax from the Dungeon Geomorphs:

Roughly one third of the rooms should remain empty. One-third should contain monsters with or without treasure (possibly selected randomly using the Dungeons & Dragons Monster & Treasure Assortment), one-sixth traps and/or tricks, and the remaining one-sixth should be specially designed areas with monsters and treasures selected by the DM (rather than randomly  determined). Slides, teleport areas, and sloping passages should be added sparingly.

This seems to align pretty well with my basic idea, which is to create maps that have 12-15 or so areas, stock ten of them, and have two or three of those be 'special' in some way, providing unique challenges or unusual encounters. I will try to make it more than 'you open the door and find five goblins', but sometimes five goblins are in a room eating, sleeping, playing cards, or on guard duty. However, not every room can have five goblins in the middle of trying to solve a trap with moving floor panels as a statue overhead casts disintegrate on those who fail to move a floor tile appropriately. The mundane encounters are needed to help keep the special encounters special, and a few 'empty' rooms give places for the GM to add to the dungeon, or to give characters places to rest. 

A megadungeon has to be a little bit of a grind to feel like a megadungeon. If you get through a level in ten minutes, it's not a megadungeon. You should have to spend a good deal of time exploring, fighting, and thinking your way through it.

Module Dungeons and Gaming vs. Worldbuilding

I have waffled between two different fundamental approaches to designing the megadungeon in terms of its physical layout and mapping.

In option one, the entire thing is a series of carefully-connected maps. It is clear which pieces connect to which other pieces. It is likely that the various small maps connect together into one massive map - or even that the thing is just one massive map. I find this a bit intimidating, both as a GM and as a player. At the far extreme, I've considered designing the entire dungeon as geomorphs. I find this too random for a true megadungeon. The middle ground, where I tend to end up, is loosely-connected adventuring areas. This was how I built my only published 'megadungeon' (kind of), the Vault of the Goblin for Saga of the Splintered Realm.  

In that design, I kept the old-school idea of player agency in this way: when you go up the stairs or climb that dark shaft, you know you are going into an easier part of the dungeon - as you descend the stairs or rappel into the dank hole, you know things are likely to get harder.

This is linked to the idea of 'dungeon levels'. That the level of the dungeon is the level of the monsters seems like an intuitive way to go, but it's also quite limiting, and stinks of heavy-handed game design. This links to a larger philosophical concern - is this a dungeon complex in a rich fantasy world, or is it a scenario for a game? Yes and yes. However, 'realistic dungeon' does not necessarily equal 'fun game', and the opposite is true. How do you create a fun game dungeon that at least 'feels' like a real, authentic dungeon in a fully-realized fantasy world?

I am not sure, but I'm going to try. 

My current thought is to try to thread the needle, and an image of Hogwarts helped me - when you go up those stairs in Hogwarts, you cannot be sure where they are going, because some of the stairs are always moving. It could be that there is a pattern, or it could be that the spider is always creating new threads between different parts of the web. 

To try to find the middle ground of all of these various competing ideas, I'm going to go with a modular design (at least for now). Each module of the dungeon is designed as a 'stand alone' encounter area; its ecosystem, forces, and overall vibe. There might be a frozen encounter area (I'm sure there will be). This is an icy hall filled with icy stuff. It doesn't necessarily draw ice from elsewhere, and it may not have an impact on another level that is trangentially connected.

Each encounter area would have a handful of ways 'up', and a handful of ways 'down'. There would be no built-in lateral moves between sections: Beyond a locked door requiring a magical key might be a stairway, but it's not just going to be a hall to another area on the same level. Stairs level a level will be marked as stairs; whether they go up or down is entirely at the discretion of the GM. 

The GM then decides, either in play or before the campaign starts - how and where these connections take place. This might be how the dungeon works in your campaign world, or it could be how Moridis has decided to structure this particular version of the dungeon for this particular group of characters. I would presume that once you have moved between sections (you descended from the lonely watchtower into the imp's abode), you would always go between those two areas with that set of stairs. Once the GM makes a decision on how two areas fit together, that's how they fit together. 

The GM has more flexibility then. If the players really want to go deeper, but the GM knows that they are going to get destroyed, 'down' takes them into another comparable level, but still closer to more difficult challenges. 

The GM also decides where the McGuffins go. For example, I have an idea of a pair of chests that acts as a magical teleportation system. A group could have a first mission to be to recover the two chests. They know that the first is in the hands of a group of kobolds (who don't know what it is), but the second was lost underground. They could recover the first, take it back to town, use it to 'enter' a deeper part of the dungeon (one of the delvings), and then either try to carry it out (continuing to look for passages 'up' to eventually get out), or just start exploring where they are, not sure where they are or how it connects, but knowing that they can head back to town at any time by jumping in the chest and emerging in the village. This is an instant campaign starter, and allows the GM to keep a lot of mystery about how this whole thing fits together.

That idea transfers to everything: you need a magic key, but the GM gets to decide where it is; you want to recover a holy sword, and the GM decides where it is currently locked up. Important items would be detailed and have a bit of lore, but where they ended up in the dungeon is up to the GM to place as desired. It wouldn't be out of character to Moridis to grab a staff of the archmage and move it to another location where those heroes might find it a little easier. She is curious to see what they do if they get it...

Here's my current thinking on this level of organization:

Tier 1: The Surface - These are designed to be entry points for level 1 groups or even solo players. These areas touch the surface (or are largely on the surface). These are filled with menials and some level 1 foes, with an occasional level 2 foe sprinkled in. They are generally easy to get into, easy to get out of, and contain at least one entrance into a lower tier. This could include a cave complex inhabited by goblins with a pool that leads into a deeper water, a ruined gatehouse with a floor grate that opens to reveal a way to descend, or a watchtower with a rusted porcullis in its small cellars, its stairs descending into darkness. These would probably follow five-room dungeon design philosophy, giving small, focused adventure possibilities, a way to take 'control' of an entry point, and some early adventuring success. 

Tier 2: The Delvings. These are the levels within maybe 200' of the surface. These are filled with foes of menials to level 3 or 4. They contain connection points to other delvings, to deeper tiers, or to the surface. Each area is themed in some way, with 10-20 encounter areas. The GM could then connect these as needed; if the first delving area is 100' deep, the PCs could ascend to another delving or to the surface, or they could descend to another delving, the catacombs, or even the pits. The GM has latitude to determine how prepared the PCs are, and 'scale' encounter areas somewhat appropriately. 

Tier 3: The Catacombs. These are levels as deep as 500' below the surface, filled with foes of levels 2-5. A party of really, really capable and well equipped level 2 characters could find some success here, but it is likely you will want to be level 3 or 4 to be regularly adventuring here.

Tier 4: The Pits. These are the deepest levels, hundreds if not thousands of feet below the surface. These are filled with foes of levels 4+, and contain portals into pocket realms of shadow and fiends. Ancient evils linger in the pits. A party below level 4 should have no chance of surviving this deep, and even a powerful solo character has little chance.

I don't think it would be wise (or particularly 'realistic') to connect a surface level directly into the pits; I would think at the very least you'd have to move through one encounter area at each tier to get to the bottom. A high-level group entering this for the first time on a mission to slay a shadow dragon dwelling in the pits might have a map with the fastest route there - but this still takes them from the Wailing Cave (surface) through the Forge of Godon (delving) into the Carotic Maze (Catacombs) and finally into the Pendular Pit. 

By the way, each of those has to be a real encounter area. Each of those names already gave me some ideas! Last thought: I am going to go through the many maps I have drawn over the years (okay, decades) and use the best ones as inspiration. This is my 'one dungeon to rule them all', so I plan to take the best ideas of my last four decades of gaming and put them all in here. It is likely that some of the levels from the Vault of the Goblin will end up re-packaged and dropped into this dungeon. The map at the top of this post is a re-draw of one of my favorite maps, and one that appeared in the Vault of the Goblin. This will be one of the delvings.

More Thoughts on my Megadungeon - Philosophy Time!

The metaphor of Moridis being a spider and the dungeon being her web is sticking with me, and it's helping me to provide the place with an overriding philosophy.

I think the tendency with a megadungeon (or my tendency at least) is to see it as 1) a stronghold where a powerful creature is holed up (hence it would be designed by nature to be defensive and resist intrusion); 2) a prison where something is trapped (and therefore is actively trying to get out of); 3) a warehouse where something has put its cool stuff.

These all provide context for the dungeon, but also some inherent problems with why things are the way they are. None of these provide a good reason for questions like why puzzles? Why tricks? Why strange statues and magical pools? What purpose do these have in a prison? Some might be defensive, but others seem horribly inefficient as a means of defense. If this is a storage facility, then building a maze in the middle seems kind of dumb.

Insanity seems to be the default answer. Oh, the gnome inventor went insane, so now the dungeon is filled with his strange creations. Okay, that kind of works. But, it doesn't give him a purpose. It's still random. Insanity is (sorry) a rather lazy way to explain randomness. 'It's random because of crazy people.' There. Now I have an excuse to include lots of random stuff. 

But for Moridis, the purpose is to play with her food before eating it. She doesn't want to get out, because she built this for the purpose of luring stuff in and bringing it to her. She wants adventurers to come exploring. In short, she's a game master.

And, like a game master, she needs to fill it with treasure so that people keep showing up. "I have a deadly pit filled with things that can kill you" is not very attractive, but add some gold and magic to the mix, and suddenly you'e got idiots lining up to get in. She's got lots of cool stuff. Furthermore, she gets excited when a powerful hero brings a new item in that she hasn't seen before - she's going to get to eat that paladin and keep his holy sword! Or, she can turn him into an undead death lord, and force him to spend eternity in torment at his failure. Fun. Then, she can use the sword (and its legend) to draw in even more people. 

And, like the casino, she needs people to win sometimes. If nobody leaves with money, then people figure things out quick. The casino only stays in business if people think they have a chance to win big. The bugs stop flying by if that light isn't shining. So, she allows some adventurers to successfully delve her depths, grabbing a few things and collecting some coins, so they can go out and show off to the rest of the world, and more people will pick up a ten foot pole and walk through the front door.

It's also possible that, in some cases, fate (or other gods) have told her to keep her hands off certain visitors. She can't openly defy the fates, so if a mortal has some larger purpose and just happens to swing by her dungeon, well, she can afford to lose some gold and those monsters can always be replaced. The sooner that mortal who has been chosen by fate moves on, the better. 

Furthermore, I think that she's really, really persuasive. I like the idea that she is a shadow that whispers. For one of the first enounter areas, I think that there's a goblin leader who has gone over the deep end with his devotion to her; she keeps whispering in his ear, and he keeps doing everything he can to please her. He's got a small garrison of goblins, has captured a tree growing through the stone that has a dryad inside, and he keeps drinking the magical sap and it's doing all sorts of nonsense to him. She just keeps whispering and whispering and whispering.

One other real gaming benefit her in terms of world building. Moridis doesn't have some larger plans. She's not seeking world domination. She's not pulling the strings in far away places. The High Marshal is building an army to conquer large chunks of land. The last thing he wants to do is waste valuable resources with a massive military campaign on her vaults, when it could easily cost far more than he gains. She poses no threat to his aims. She poses no threat to anyone, really, so nobody has a reason to mount a significant campaign against her personally. However, the High Marshal could hear rumor of an item that could increase his power or give him particular influence, and hiring a group of mercenaries to scur her dungeons for that item is not beyond the pale. There are lots of different reasons why lots of different people could be here, and there are lots of strange ways to get in and out.

If the spider web is far enough out of my way, I tend to just let it be. Why knock down a spider web when it's not making any problems for me?

When a bug is stuck in a web, the spider doesn't have to rush over and poison it. The spider can sit there and watch the bug exhaust itself trying to break free from the web, which is a fruitless endeavor, and then casually saunter over and deliver the fatal bite when that spider is good and ready. 

Moridis has lots of time. And she likes watching the bugs struggle.

What Is With Me and Megadungeons?

I have mentioned at least twenty times how I want to build a megadungeon for my game world. In reading about published megadungeons, I often find similar problems. They are either repetitive and boring (how many bugbears are we going to kill here?) or they are discordant (oh look, a cool magical trap that has no reason to be here, but at least it's interesting). The real megadungeons of suburban myth are largely that: the Ruins of Castle Greyhawk were never a fully-realized set of detailed maps: they were notebooks of random scrawlings and poorly-conceived maps that were held together with duct tape and chicken wire. I read a little bit about Undermountain, and found analyses that others had done showing how the maps Ed Greenwood originally crafted were almost unusable by others, and how the maps that were released were actually cut and pasted from other maps in other TSR products.

The reviews I've read of some of the most iconic current megadungeons point out how these tend to be large maps with lots of similar, redundant encounters. Open door, kill monsters, repeat. 

I keep coming back to some text I wrote a while back but never did anything with. This is my hook for the megadungeon project 7.324 (or whatever):

***

“Moridis? Oh yes, dear. She’s an ancient evil. Quite terrible. More tea?” Mrs. Kunnelbreck didn’t wait for an answer, pouring boiling water into your cup for the third time.

Mr. Landiford wasn’t much different, “You might need this.” He offered forth a 10’ pole, “for trap findin’. I hear tell Moridis loves trappin’ all y’all.”

“Moridis is a spider, and her labyrinth is the web,” warned the Lady Eldritch. That wasn’t her real name. Rumor was that she was a Huffenfeffer, but they had disowned her and seemed quite relieved that she had taken on a stage name. 

For his part, Yrzek the High Priest stared into the decanter, “Ah. Moridis is pleased. She awaits your visitation.”

***

Moridis is an ancient evil of the first world. She was consigned to the depths, trapped and hedged in. She has become one with the stone itself, creating a vast labyrinth that she has filled with treasures and torments beyond measure. Her form is a mortal woman, a dragon, or a voice in shadows. She can be defeated, but will re-emerge within 1d12 days, assuming physical form again. The village ‘worships’ her, offering regular sacrifice. It submits an animal sacrifice once a week (typically a cow) but a human sacrifice once a year (it’s usually some bandit they catch). Other than the occasional human sacrifice and worship of a foul entity of shadow, they are a decent enough folk.

***

The locals refer to Daggerford as quaint.

Objectively, it is a village that is almost a small town that squats where the midlands break up at the base of the Palisades. Here, the potatoes grow large, the corn grows tall, and people know never to look East. 

Ever. 

The far borders of their village touch the edge of her lands. They have a symbiotic relationship; the town folk get along quite well, and Moridis generally lets them be. Since the primary entrance to her halls is 8 leagues north of the Dwarven fortress of Stonehold, the dwarves have been known to send raiding parties to keep her at bay and recover what wealth they can.   

And, when a new, fresh-faced group of adventurers sets out from the Blind Basilisk along the eastern road, townsfolk meet and cheer, and often exchange bets, as these fools attempt to delve the lair of the mighty Moridis.

Most are never seen again.

 

Maybe Why I Love Hack'D & Slash'D

Full disclosure: I struggle to create content for my games after I craft them. I love the IDEA of making lots of superheroes for my supers game, and scores of monsters for the fantasy game, and detailed deck plans of a variety of starships for my sci-fi game. But, I typically run out of gas partway into expansion projects. And it's the same reason every time - I get bogged down in crunching numbers. I have to sit with the book and flip back and forth and check things... wait, how many points was he able to allocate to strength at level 4? And how many bonus hit points does he get for this ability? And wasn't this a bonus to AC? And, oh shoot, his attack modifier is wrong because I forgot (again) to factor in the trick shot bonus he gets. And then I get worried that I'm releasing stat blocks that have errors (because they sometimes do), and the whole process stops being fun and starts to become accounting. I don't mind that process when I'm making the game as much, because then I'm testing the mechanics I'm developing, and seeing how they run in play. Once the game is finished, I'm just doing accounting. My mom was an accountant. It looked pretty boring. 

But for Hack'D & Slash'D, the process of making monsters and characters and spells and items is still fun, even after the game is published. And, I think in large part (okay, pretty much entirely), it's because it's so easy.  I remember when Hack'D & Slash'D was 'born'. It was probably twenty years ago now, and I was at a friend's house, and I was talking about Ogres. I think third edition D+D had just come out. I was arguing that the perfect game, to me, would be able to represent an ogre in a handful of statistics. You should be able to get an ogre down to four or five numbers, and run the game from there. Yes, dragons need more, but an ogre? Why do I need more than 5 numbers for that? This struck me, and still strikes me, as excessive:

That became the various iterations of Resolute. I was always after that 5-stat ogre. If I can get him down to 5 stats, and have a playable, robust game system around him, that would be the perfect game to me. I want to be able to do everything in can do in B/X, but I want an ogre to be 5 stats in a single line of text.   

Here's the entire stat block for an ogre in Hack'D & Slash'D:

In retrospect, the word 'intimidate' is unnecessary; 'bully' already suggests that. I could have this down to two lines and not lose anything. I saw this stat block in my head twenty years ago; it just took me a while to build the game around it.

The Marshalled Lands Need a Marshal

In thinking about the kind of guy who would call himself the High Marshal and would name a place and pay for a map and do all of this, I was reminded of the drama that played out in Russia in the last week, and thought about the kind of guy who'd be 'in charge' here. He is Abran, the High Marshal. He doesn't like the word 'tyrant'. It's such an ugly word. He's simply 'called'. Who else bears the sword Justicar? Who else has held counsel with the Iron Dragon? Who else has spilled blood on the edge of the Shadowed Vale and lived to tell of it?

Nobody. That's who.

And that is why he is called to Marshal the Broken Vale. 

And Marshal it he will.

Note: I've included the final wash drawing and the line drawing. Also, fun fact: if you look up 'lawful netural' in the dictionary, it's just a picture of this guy. 

Of Marshalled Lands and Messy Maps

In looking at my campaign map, which is the 'official' map I'm working with for Hack'D & Slash'D (which I had hitherto referred to as 'the Broken Vale'), I realized that the map showing the Desert of Desparing Souls was not compatible with the map I've been using for the Desert of Despairing Souls of my Arath solo game. And my first thought was, 'well that's a problem'. And then my second thought was (putting on my best Bob Ross), 'well, what a happy accident'. I made the maps a year apart for different purposes, so they have some shared concepts, but different locations.

Because let's talk about medievel maps. Until the late 1300s, they were messy and inaccurate and contradictory (and sometimes beyond 1400 too). They relied on a variety of accounts to generate their estimated locations, and then took a guess when they weren't sure. A desert, in particular, would be especially hard to map. What points of reference are you using? You were feverish and dehydrated and going through a sandstorm for three days, but SURE, we trust you that it is exactly twenty-two leagues between those points. No doubt.

Now, in a fantasy world such as this, magic would provide significant help - until it didn't. Yes, an Augury (or Dark Argury) spell might give you answers, but these answers might be cryptic, coded, or incomplete - so you make assumptions that may or not be true. If you ask Ubek where the Ruins of Vas Anok are, he might tell you to walk to that place, because that is where he intends to meet those on the road to Vas Anok - that is where you will 'find' it (from his point of view), even though it is not where it is located. You can only find it by getting lost first. 

Some cultures might have symbolic rather than literal maps. Why yes, the dwarf kingdom IS ten times larger than anything around it. It sure feels that way to us. The other lands don't really matter, so we made them really tiny on our maps. Your point is?

And then I thought that maps are stories (probably why I like them so much). They are stories about us. So, if a human is going to commission a map (because that is something humans do), he (this is totally a guy thing to do) would also want to be able to name it. And he would not name it 'broken', because that implies a lack of control. So, even though EVERYONE calls it the broken vale (because this place is pretty broken), he would decide that these are called the Marshalled Lands, because by mapping them and naming places, he has brought them under some form of control. They are under a banner of sorts, because they have been mapped and there is a written law that puts all things so mapped under the High Marshal's Dominion. 

Oh, well that settles things for sure. No doubt the elves are going to completely bow down and let you garrison troops. They didn't know that you'd put it on PAPER and everything. That's, like, really official. So, this map is very official. Until it isn't.