GreatWyrmGold's Megadungeon Development Thread

[GreatWyrmGold]

This first post is an index for a project I'm posting as much to motivate me to actually keep working on as anything, so that I can give the first post a unique post title without changing the thread title.

Part 1: Introduction and Idea Barf (http://www.giantitp.com/forums/showsinglepost.php?p=24068415&postcount=2) (okay, maybe that unique post title wasn't worth a double-post).
Part 2: Spreadsheets and Basic Structure (http://www.giantitp.com/forums/showsinglepost.php?p=24076488&postcount=3)
Part 3: Key Items and Themes (http://www.giantitp.com/forums/showsinglepost.php?p=24079594&postcount=4)
Part 4: Lore and Proto-Maps (http://www.giantitp.com/forums/showsinglepost.php?p=24094693&postcount=6)

[GreatWyrmGold]

I’ve been reading theangrygm’s Megadungeon Monday series (https://theangrygm.com/series/megadungeon-monday/), and it made me want to do something similar. That’s what this is—GWG’s MDM. Except that my storytelling interests and what I want to get out of a TRPG are different, so the goals and resulting project will be different, just built with similar methodology. And a worse schedule.

I should probably explain those goals, which I’ll share in a thought-dump of assorted seed-ideas, ordered roughly from most general to most specific:
A more active-participation story. Angry’s overarching story structure of “a tragedy happened here; now deduct facts, beat up things, and fix stuff until the problem goes away” is perfectly fine, but it doesn’t fit my writing or gaming tastes. I want to create a story where the PCs are in the middle of it, active agents in preventing a tragedy.
A focus on survival and macro-scale resource management. These two overlap heavily (for instance, managing your supplies of food and water), so I’m lumping them together.
A more exotic environment, giving me more freedom to create wild locales and convenient gating mechanics (though one specific enough to provide creative constraints).
I have this image of how the first few levels will play out. The players go into what looks like a normal dungeon where they learn some ritual is going on. The dungeon takes a couple days to properly penetrate, and at some convenient point of no return, the dungeon lifts into the sky as the ritual begins. The PCs fight their way to the top o the dungeon, and confront the ritual guy. Partway through the boss fight, he snarls something about the ritual being complete; all that’s left is the sacrifice.

With these seeds in place, we can sketch out what comes next.
There’s this monster called an astral dreadnought which is a cool idea (a big fantasy-space monster which eats things and shunts them into a pocket dimension), executed in the most boring manner possible (the pocket dimension is empty and you can escape it with a good Will save). So let’s make up a monster which is that cool idea done better and use it in this adventure.
The roof is ripped off the dungeon thanks to some magic thingamabob (Note: The boss fight at the end of the tutorial needs to include some kind of magic thingamabob that can rip off rooves at a dramatically-approptiate time), and the party sees a distant silhouette in the sky. After defeating the boss, they are left to stare at the approaching monster thing and scour their character sheets for anything their probably-around-level-3 butts can do to drive off a huge flying monster or escape a mile-high flying dungeon. After they’ve run out of ideas, the monster calmly devours the dungeon, with the PCs still inside.
The fortress crashes down in some pocket dimension, full of dungeon bits, scattered around the half-digested remains of the monster’s previous meals. But the digestion is a metaphysical process, not a literal digestion, which gives us a justification for all sorts of magical hazards and weirdly degraded landscapes.
Also, the monster’s gullet is a series of pocket dimensions; right now, they’re in the monster’s gizzard, marinading in its eldritch saliva energies and being smashed into other bits of landscape by whatever the space monster equivalent of gastrolites are. There’s some magical way to go from dimension to dimension unharmed, which the party finds along with the basic tools for survival and enough information to give them hope and a goal. Somehow, some of the dungeon’s residents slipped past the early stages of digestion (probably leaving behind hints or artifacts which let the PCs continue), and got into the dimensions roughly corresponding to its stomach, intestines, etc. They’re trying to escape, because the ritual didn’t end up quite right (the PCs killed the intended sacrifice in the boss fight) and they need to get out to do whatever they intended to do. Meanwhile, the PCs need to get out to stop them and also not get digested.
So the adventure continues, the PCs going through the megadungeon that is the giant monster’s gullet. They find magic items or learn spells that let them traverse strange hazards, learn more about what this monster is and what the original bad guy’s plans for it were, and then escape to send the beast back where it came from. Because by the way, it’s still flying about, munching on the occasional hill or town and sending debris into the gullet to be digested.

Alright. Now that we’ve figured that out, let’s throw together a rough draft of our progression table (which we’ll be modifying and filling out further as we go along).
(The image host I'm using doesn't work well with forum image stuff, which I only just realized, so have a link to an image of a spreadsheet (https://prnt.sc/onysfr).)
Quick note: I’m not using “tier” in the technical sense of a group of five-ish levels. I’m using it to mark out tiers of obstacle difficulty. The end of Angry’s first Megadungeon article (https://theangrygm.com/megadungeon-intro/) talks about a ratcheting difficulty curve; the players enter a new area, find themselves challenged by the area’s challenges, gain one of the big Power Levels, and start being able to handle the local challenges with ease. The tiers are where I’m incrementing the ratchet. There’ll still be some difficulty shifts within the tiers, but nowhere near as significant as between them. It makes sense to line these up with transitions between gullet-dimensions.
You might wonder why I’m writing the chart up to level 20 when there’s a big premise-destroying spell listed clear as day at level 13. The reason is simple: I dunno how long I want this dungeon/campaign to be. Of course, getting plane shift is a pretty serious milestone in this adventure. It lets adventurers leave the monster, ignoring all the hazards imposed within until they’re fully recovered. On the other hand, it also lets them pop into the main world to see what’s going on and get pointers from magical experts and whatnot, letting them fill in the blanks and figure everything out that they hadn’t before. This also lets us reward their diligent research, if any, by letting them convert their discoveries into powerful magic gizmos or whatever. So let’s include a chance out of the monster some time around that point. But its placement is inconvenient; either we need to divide the monster into two big segments and a vestigial bit where they learn how to escape, or the third part is going to get the shaft without some narrative reason to stay.

Let’s look at each possibility and see where each can go.
The monster’s gullet could be divided into stomach, intestine, and colon. The stomach is a storm of floating islands, smashed into each other and pounded with debris whenever the monster devours another chunk of land. Why does it do this? Maybe the monster is trying to kill everything it eats before it’s digested, or at least all the weak stuff. In the first case, maybe it’s trying to isolate the souls so it can turn them into something else that it mixes with the physical matter mush in its intestine. In the second, maybe the intestine is for “refining” the strongest creatures into something that’s important for the colon. Either way, it could be turning these creatures into something it needs to heal or grow, akin to its cells or organs or whatever; the colon is just where anything not turned into that thing is dumped, along with any creatures that avoided the transformation. There’s no exit, so the PCs realize they need to make their own.
It could also be divided into gizzard, stomach, and intestine. The gizzard smashes things together, like real gizzards; the stomach breaks things down, like stomachs and small intestines (yes, I’m butchering the biology terminology a little); the intestine extracts whatever the monster needs, before presumably expelling any corpses or cobblestones it didn’t absorb through a gate or something. This model lends itself more to the monster extracting something specific from its food, powering itself to keep going. It also provides a use for the “you’ll be leaving here shortly” nature of the third gut bit; it’s draining something out of you at a rate you’ll notice within an adventuring day, so you need to keep ducking between stomach and intestine as you try to get stuff done, until you can just use the plane shift to go freely between intestine and real world.

Let’s name these two possibilities after their unique dimensions. One is the Colon Option, the other the Gizzard Option. Both options have strengths and weaknesses, depending on the kind of story you want to tell. The Colon Option wraps up the initial gut exploration quickly, which works well if the final confrontation needs to be outside the monster or involve trekking back through all three dimensions. The Gizzard Option works better if you want the final confrontation to be in an unexplored part of the monster’s insides, which happens to be most easily reached from the most dangerous part of its gut. They both work well with certain options for what the monster is and what it gets out of eating everything, which has implications on everything from the sorts of encounters and dangers the party would face in each area to the sort of people who would summon the thing.
It sounds like this is a pretty important choice for the entire project. That’s why I’m not making it now. See, if I decide that I’m going to make it a god-lich that transforms the strongest souls inside it into fuel for its anti-miracles, that’s going to lock in a lot of decisions down the road for everything on the crunch-fluff spectrum, from themes to random encounters. I want to be able to decide those details freely and then figure out the big picture. You might be thinking that this is backwards, that I need to figure out What’s Going On to figure out the details. But the devil’s always in the details, and I want to root them out before locking anything in.
Let me give you an example from my past as a terrible fanfic writer. DBZA Kai 3 and Dragon Ball Fusions have gotten me on a Dragon Ball kick, so I’ve been considering redoing an old fanfic I wrote a few years ago. I decided that What’s Going On was that Goku got Worm superpowers instead of being able to go Super Saiyan, and then stuff would happen that would make that interesting. There were a lot of problems with this that I didn’t recognize until well after I stopped updating, like how basically nothing of importance would change for the first arc of the story, save for some details that would set up the part of the story where things got interesting. But I’d locked into the idea that I was telling the story of Goku the Parasaiyan, starting from the moment he got those powers, with all the neat twists and turns I could squeeze into one arc causing fun butterfly effects down the line, and that meant I’d have to slog through chapter after chapter of recounting events that were almost like canon but without Super Saiyan Goku. Bleh. If I had instead started with a vaguer idea, I could have fiddled with a loose timeline of events and consequences until I found one with a series of events which lead to an interesting story at some point which served as a usable introduction.
Different types of writing are pretty similar. They all rely on a bunch of things fitting together perfectly, or you’re gonna have a bad story. You can’t make a plot that’s going to mesh perfectly with everything else if you don’t know at least a bit about all the other things you need to mesh it with—characters, setting, themes, mechanics, etc. So now that we have loose outlines for some of them, let’s make loose outlines for other important ones. Then we can tighten them bit by bit until we have a good adventure! (Or until I run into a stretch where I don’t have time to work on this and forget to ever come back to it.)


The last goal that’s barely been mentioned so far is the survivaly, resource-managey aspect. This is an important point to cover early, because it’s supposed to be a bedrock of the party’s mid-scale gameplay loop. The smaller loops involve individual encounters and the larger ones involve completing quests and whatnot, but in between we’re supposed to have that survival stuff. But what is it?
The first answer is “goodberry ruins everything,” and there’s some truth to that. D&D has a lot of spells that ruin entire categories of challenges without providing many new and interesting options, but few come at 1st level, and goodberry is the worst of them. After all, with something like knock[i] or [i]dimension door, you at least need to waste a spell slot on trivializing a challenge. But goodberry isn’t a waste of a spell slot at all; aside from feeding your party, it provides efficient out-of-combat healing. There’s no reason a party who could cast goodberry wouldn’t do so as soon as hunger was considered a possibility.

Some would say “Screw it” and just skip to the bit where I talk about the consumable resources that the party would need to scrounge for in the monster’s gut, but I won’t. After all, this sort of mechanical theme needs to start somewhere, and the most natural way to introduce the idea is probably the best. Tell the players that they’ll need to worry about getting food, and they’ll be primed for the idea of looking for other things they need.
Remember the tutorial-ey dungeon in the material plane? I intend that to be a place to introduce all the critical concepts of the adventure with the safety net of nearby towns and villages. But how nearby is “nearby”? Maybe these ritual people made their ritual dungeon out in the middle of harsh wilderness, and it’s winter so the locals charge inflated prices for their food. The players should be encouraged to spend some time looking for food, hunting game or gathering roots or something. This provides the DM a change to introduce whatever scrounging mechanics we come up with eventually.
Of course, the party could cast goodberry and, like the idiot who skips the tutorial the first time they pick up a controller, shoot themselves in the foot on this critical concept. But there are ways around this, if you’re willing to tinker with the rules a bit. The first is simple; add a houserule (“adventure-rule”?) that goodberry infuses pre-existing berries rather than conjuring new ones out of nowhere. Sure, it’s a lot easier to find one berry per person per day than it is to find 2,000 calories of miscellaneous wild food, but it shouldn’t be easy in the winter.

The more interesting option is to mess with the rest mechanics. If you don’t get a long rest every night, it takes more than one spell slot per long rest to feed people with goodberry. (After all, the berries still only last a day.) A first-level druid who needs to cast two of her three spells per rest just to keep the party fed properly is going to view this as a sacrifice, not a trivial matter. After all, that leaves just one other spell to cast between those two days!
I don’t like the idea of making casters rest a full week to get their spells back, as per the gritty realism so I’ll introduce a less extreme system. Maybe a night’s rest counts as a short rest, and taking a whole day off counts as a long rest. This wouldn’t normally have much effect outside of time-sensitive scenarios...but when you have a limited supply of something you need every day, time is always sensitive. Regaining spells means spending a day’s worth of resources, without getting a day’s worth of scrounging done.
Of course, you need to do this sort of tinkering carefully. D&D is (vaguely) balanced on a diet of two short rests per day; too few short rests disadvantage short-rest-dependent classes like warlocks while generally empowering normal casters, while too many short rests do the opposite. But part of this is just adventure design; most of the published adventures my group has played through failed to give a good opportunity for the party to take a short rest before their first opportunity to take a long one. (Though we might just have a warped idea of when it’s fine to take an hour-long break in the middle of a hostile fortress.) We would need to balance the rate of gaining and expending resources such that they can generally only afford one day off for every three adventuring. (One before the short rest, one between rests, one after the short rest, and then you take your long rest.)
This would also slow down the pace of the game, relative to the calendar. Normally, entire dungeons blaze by in just a couple days; now, that amount of content will need to be stretched over a week. This means that tightly-packed segments of dungeon aren’t practical, but lonely ones with not so much as a rat in sight for hours are. That changes the atmosphere of the adventure a decent bit, which is yet another argument in its favor.
Still, much as I like the idea, I’m not locking it in until I’ve had a chance to tinker with all the other aspects and talk about how they’d fit (or not fit) with the big change.

And while I’m at it, I have two more fixes to goodberry that could be applied once the party enters the monster.
The first would be to introduce a hazard that impedes healing spells. Maybe they automatically fail, maybe they just go wrong on a non-trivial casting ability check or something. Either way, you introduce both a unique hurdle and a hard stop on goodberry shenanigans. Needless to say, we’d need to design encounters and “adventuring days” more forgivingly to account for adventurers’ lower endurance, and maybe scatter around some healing potions (or at least the equivalents of healer’s kits). We’d also define “healing spell” quite narrowly; in particular, anything that gives temporary HP is fair game. Regardless, we’d want to introduce some kind of counter sooner rather than later.
The second goes the other direction. After all, not all parties are going to have a druid; we can’t assume that they’ll be able to trivialize food, because if they can’t, they starve. So we’ll throw in a magic item that can cast goodberry early in the adventure—just as soon as the party can be expected to realize how little use it is right then. (After all, it’s still affected by the anti-healing thing.) Make it something convenient to carry so they decide to keep it, just in case, and when they counteract the anti-healing thing, they have tons of food! Maybe it’s something upgradable like Dark Souls’s estus flasks, a slowly-growing pool of healing that restores itself every day.

But now, let’s consider other things the party would be scrounging for.
Before the party gets omnomnommed, they should have something other than food to scrounge for (both in case they’re clever enough to trivialize food-gathering and to acclimate them to scrounging for multiple things at a time). Maybe there’s some kind of poisonous or cursed miasma in some part of the initial dungeon, and they need to gather herbs for a remedy. It shouldn’t be something lethal, but something irritating; maybe 1d3 points of poison/necrotic/something damage every X period of time you hang around in the dungeon. Not enough to draw blood in any but the closest of shaves, but enough to make them interested in a remedy. Probably good to let the players know “There’s a weird miasma around the dungeon, but some fancy herbs or whatever counteract it for a while” beforehand; it’ll encourage them to look around while they scrounge for supplies.
We want to give them stuff they can actively use, too, though I’m not sure what. Survival games generally scatter crafting components around, for instance, but D&D doesn’t have much equipment which isn’t either cheap enough to get at 1st level or too expensive to be thrown around willy-nilly. I might be able to find some intermediate items to scatter about, low-level scrolls or powerful alchemical items or something, but if I want to do anything like that which didn’t suck I’d need to homebrew it. The only idea that comes to mind is letting them put together some cheap magic items of their own, e.g. a talisman that you cast a spell into during creation and it can cast that spell once per such-and-such time.
There’s also shelter. I’m imagining that one of the common hazards in the gut dimensions will be solid rain, either stuff the monster just ate or stuff that’s just been transferred from one dimension to the next. So until the party gets access to some kind of magical shelter, they might need to find shelter materials to either build new shelters or repair whatever shelter they’re in; otherwise, they can’t rest (long or short). Depending on how cold the dungeon’s neighborhood is, they might need to do something like that there to stay comfortable. (This also ensures that resting always has a resource cost.)

But maybe I should stop trying to invent new mechanics and instead toy with existing ones.
I’m imagining that for each new hazard, the party would initially go through a period of barely keeping up with their resource demand; they don’t get a lot of off days. By the end, though, they have things sorted out enough that they can safely take a long rest after most adventuring days. This does give the party a sense of mastery over the dimension, but it also screws with game balance. Clerics are going to be hurting for a lack of long rests at the beginning of the dimension, while warlocks won’t get a good chance to shine without short rests at the end.
How about we give the party artificial access to extra rests? Maybe we have some rare consumables that give the whole party the benefit of a short rest, and others that let them upgrade a short rest to a long one. If we sprinkle those through the adventure properly, we could balance out the party’s rests and give them something to look forward to finding, instead of just stuff that they need to get through the next day.
Sure, there’s nothing stopping the party from using these items at the “wrong” time. Maybe they blow through the last three encounters of an arc with all their short-rest abilities recharged, leaving them short-rest-deprived for a bit. That means that the short-rest classes get a chance to shine, which they pay for by not shining as brightly later. I’m fine with that. The players are going to throw off the balance, and that should be fine as long as the balance is within acceptable tolerances to begin with.


Alright, that seems like a decent overview of the ideas I’ve come up with so far for this adventure. I think I’ll be ready to start fiddling with spreadsheets next time.

[GreatWyrmGold]

Alright, let’s talk structure. Specifically, daily encounter structure.
D&D 5e is designed around a fairly standardized adventuring day. Several encounters, two short rests. I mentioned last time that I was planning to mess with the rest system, stretching this “adventuring day” over multiple days and adding items which grant bonus rests, but that only matters in so far as it affects supply systems and when we say encounters happen. Right now, we’re interested in what kind of pattern we want to put encounters in, and we’ll call that the “cycle”. Three sections delineated by short rests, separated from each other by long rests.

Alright, now for a big friggin’ spreadsheet.
http://oi68.tinypic.com/2rw1a48.jpg
Let’s go over the columns one by one.
Avg. XP: The average XP for one “encounter”. You should be aware that not all encounters should have the same XP value. To this I say: Meh, I’ll work out the details later. A day with one-quarter easy encounters, one-quarter hard encounters, the rest medium will have the same total XP value as if all were medium.
Rand. XP: The XP awarded for a random encounter. As per Angry DM’s megadungeon, I want to discourage grinding random encounters, so they give about 10% the XP of a “real” encounter.
Cycle: The number of long rests the party is expected to take, with three segments divided by short rests. Basically just a timekeeping column.
Min/Max Starting Level: If the party has completed everything they had to so far (including random encounters), they will be at the minimum starting level for that cycle. If they’ve also completed the optional encounters and had twice as many random encounters, they will be at the max starting level.
Encounters, Rand. Enc., Opt. Enc.: The number of required encounters, random encounters, and optional encounters, respectively. Most encounters are going to be combat, because that's the only pillar D&D supports natively, but I can include social encounters or special discoveries or whatever without any real issue. Heck, maybe I’ll ditch combat XP altogether and replace it with XP for discoveries or something, but I’ll skin the XP cat one way or another when I come to it.
Min XP, Max XP: The XP for completing all expected encounters (including random ones), and for doing that plus optional encounters and more random ones.
Min Cumul, Max Cumul: The minimum and maximum cumulative XP the party could get.
Notes: Other stuff I’ve thought of for that given bit. Will probably be split into more columns once this gets fleshed out more.

Keep in mind that these numbers are just first-draft junk that I’ll be changing later in the project. You might wonder why I’m making a spreadsheet like this so early in the project, when I know it’ll be changed later. The reason is simple: I'm trying to get a rough idea of what I'm working with, how much adventuring I have to spread out over this megadungeon thing. The spreadsheet tells me that (if I don’t change the levels where I want to put major story shifts) I can probably expect about three “adventure cycles” before entering the monster, seven in the first part, and six in the second. Maybe it’ll end up being four, six, and eight in the end; maybe it’ll be two, seven, and five; but by having a vague idea of what I’m working with, I can start to make vague plans.

Let’s start with identifying good places to put logistical items—that is, items that let characters overcome some supply challenge or another. You might wonder why I want to do this in a game centered around survival and gathering resources. The answer is simple; magic exists. D&D is chock-full of spells that negate basically any challenge that isn’t either combat or stuffed full of magic that negates whatever magic you were going to use.
If I don’t account for the magic, all of my challenges will be blandly ignored. But if I do account for the magic without accounting for parties that don’t have the magic, I’m going to murder any parties that didn’t try to work around the survival challenges via spell selection after the point I assume all parties will have access to the spells. Solution: Magic Items. Some will be primarily logistical (from a design perspective), but I’m going to also make sure they have alternate applications, whether obvious or obtuse.

This is probably a good time to discuss how I’m planning to hand out the critical magic items, which means I need to talk about world structure. Angry DM kept comparing his dungeon to games like Super Metroid, which functionally take place in one big dungeon. I’m planning this to be more like an open-world game, Breath of the Wild or something. But I’m using the same tools, and to explain why (and how) I need to talk about Dark Souls.
...Well, I don’t need to do that, and I don’t really need to explain why I’m using the same tools in the first place, but I’m gonna do both of those things regardless. Because I ramble.
Dark Souls and most of the “soulslikes” it inspired aren’t quite open-world games, because the worlds they create (while broad and deeply interconnected) are constrained. In many ways, they are built like dungeons. Look at this map of the Dark Souls world:
https://img.fireden.net/vg/image/1441/40/1441404170667.png
It breaks down the world into discrete areas with pathways between, not unlike a dungeon. Yet those discrete areas are often massive, sprawling, and may have only limited barriers to stop you from crossing them easily in any direction. In short, while it’s more open than dungeon game worlds, it’s more dungeon-ey than open game worlds.
“Dungeon” games and open-world games are not the same thing, but their differences come more from their weaknesses and drawbacks than their strengths. A well-designed game world for either kind of game guides the player along an engaging path while still giving them a sense of choice (and other things you want the player to experience).

I can’t make this whole gut dimension some giant, desolate wasteland. Not only is that bland, it doesn’t give me any tools to work with to guide the players. They have total freedom, but what good is the freedom to wander aimlessly? But I don’t want to make it a linear corridor, lined by impassable obstacles, with goodies scattered about.
My plan is to make the world a hostile but mostly travelable wasteland, dotted with obstacles (which may or may not be traversable by higher-level magic) and obvious landmarks. These need to form a path to travel which provides a good experience for the players, without being an obvious “only path available”. I’ll be putting the bulk of rewards and whatnot in these landmarks, making them almost impossible for players to miss on accident.

But before I can decide when to put various items, I need to decide what items I put where. That requires figuring out what is going to disrupt things, and what I’m going to add to supplement that disruption.
Let’s start with goodberry (which I will be abbreviating as GB). Last post, I mentioned the possibility of restricting GB in the early part of the gut dimension by restricting healing spells in general. This gives us two items we need to hand out. The first is pretty simple, an item that casts GB every day. This gives you food, but also healing, and that’s enough utility built right into the spell.
The second is an item that wards off whatever energy-draining dealie is doing that. I’m imagining a fancy talisman or what-have-you being used by some positive-energy-wielding maniac boss. It needs a secondary function; the most sensible would be either protecting against necrotic damage or enhancing healing spells somehow, but it might also be something used by the aforementioned maniac during their boss fight.
I checked whether GB allows fulfills water requirements, and apparently a single magic berry doesn’t contain enough water to sustain a living creature for very long. So we can still use water as a leash, until the cleric decides that casting create water daily is fine. So we can’t use it for very long; maybe it just ties the characters to a big, static source of clean water until they find the water-giving item. I’m torn on whether to make it a decanter of endless water with extra benefits, or a water-magic item which lets the players realize “Hey, this makes water we can drink!” The latter lets the players feel like they cheated the adventure, the former ensures that players won’t get stuck.
I mentioned shelter, and Leomund’s tiny hut short-circuits that the way GB short-circuits food concerns. So the players will come across some kind of shelter-forming item around the time they might consider spending spell slots on tiny huts. I’m thinking either some kind of collapsible hut or an earthbending tool, with either one working as battlefield control in addition to their logistical function.
I might be able to find some way to make consumables that helps characters survive extreme temperatures, and then a fire/ice item that lets them ignore that, but there are a lot of holes in that idea and too many potential ways it could be either trivialized or accidentally lethal if we’re not careful. Eh, let’s stick a pin in it.
Aaand...that’s about all that I can add without inventing the supernatural environmental hazards that I know are going to show up.
The anti-anti-healing item needs to come in fairly quickly, because going too long without letting the party heal (even with reduced encounters per ‘day’) is just asking for something to Go Wrong. But I don’t want it to come up immediately, because that makes the whole thing feel pointless. I put it right at the point around which I expect players to reach 5th level; that lets them lose a major restriction at the same time as their first big jump in power, which feels about right.
The shelter item replaces a spell first available at 5th level, and it’s a ritual spell, so no slots need to be expended for it. But thanks to pacing concerns, I can’t give the shelter item too soon after the anti-anti-healing one. So let’s wait a level or so while the wizards feel clever for learning that spell. In fact, let’s ensure the shelter item doesn’t completely eclipse Leomund to reward players who spend spells known on this spell.

At this point, I have two magic items and one good space afterwards to put them in. The other just kinda has to be squeezed in somewhere. That would make the first section of my table look like this:
https://i.imgur.com/6N4HqYI.png
Those spaces are important. For proper pacing, you need to have valleys between the peaks, with the breadth of the valleys corresponding to the heights of the peaks. (And the valleys need their own smaller peaks within to avoid dragging, creating smaller valleys that need tinier peaks, etc, and each peak needs to generally higher than the last, but that’s not as important right now.) If you stuff too many big magic rewards into the game too quickly, they feel less like the peaks and more like just another patch of slope.
So I fiddle with things, and realize that I’ve been making a silly assumption. I’ve been assuming that the loop of going to each landmark and scavenging it for supplies, new tools, information, etc would take about one “cycle” each. That seems a bit short (especially when I’m trying to keep the rest:encounter rate high) and distinctly monotonous. So let’s not do that! Instead, let’s try something like this:
https://i.imgur.com/yn87acr.png
There’s still some uncertainty there, of course, but there’s also a clear enough outline to start with. There’s a relatively small starting area, and the party shows up within a few days’ travel of a landmark that clearly includes a spring of fresh water. They spend a few days scrounging for surviving rain barrels from the dungeon or something while they move towards the landmark, then they fill waterskins and stuff with water and start exploring nearby landmarks. One has an anti-anti-healing item wielded by some maniac who abuses positive energy, the other is a water-creating item wielded by someone who presumably has a water theme. Freed from the constraints of staying close to a water source, the players wander further afield, investigate a really important-looking landmark, and discover a way to get to the next area of the gullet. One that has bigger threats, all-new dangers, and...something that we’ll need to hint to the players is deeper in the monster, but we can handle that when we’ve got a better understanding of what’s going on.

New world, new challenges; second verse same as the first. But since I’ve already solved all the normal survival problems that PCs can solve through magic (ie, all of them), I need to find new problems for new items to solve.
Let’s start with mass mobility. Since this adventure is using an “overworld” map with exotic terrain, it’s easy enough to lock adventure areas off until the players have access to spells like wind walk or teleport (6th and 7th level, respectively). There’s also teleportation circle as a 5th-level spell, but unless there are a bunch of teleportation circles in the monster’s gullet and a decent supply of rare chalk it won’t do much. (Hm…[i]scribble scribble[i])
The players should enter the second gullet dimension at or near 8th level, and they get access to wind walk at 11th, which happens to be right at the end of the segment. On the other hand, fly (a 3rd-level spell) should be available for tactical use well before they show up, so giving the PCs flight relatively early wouldn’t be too disruptive. Probably makes more sense to give it about halfway through, though.

By this point, it’s obvious that there’s an obvious potential elemental theme going on here. We have a water-giving magic item, an earth-manipulating one, a flying one, and one which could easily be associated with “light” or “positive energy” or something. Also possibly a magical berry bush for “wood,” which is a Chinese classical element.
An idea I’ve been fond of is combining the five Chinese elements with the four Classical elements (plus æther, which was sometimes included in the latter list, and perhaps void, which basically just serves the same purpose as æther in different occult frameworks but has different implications to a modern audience, and it’s probably not hard to link æther to positive energy and void to negative).
I’m not sure what I’m going to make of this probably-between-six-and-eight-elemental motif, but I might as well put a pin in it and see if there’s anything interesting to be done. If not, we can maybe shift some of the artifacts to not have the elemental associations; no harm no foul. (Side note: If we put wood and metal in opposition, as well as æther and void, the four elements in the second half of the dungeon are all opposed to those in the first half. That was 100% unintentional, even though I started leaning into the elemental thing about when I started fiddling with tables and the sizes of these areas.)

Back to the obstacles, but with that in mind. Our elements remaining from our potential eight-element framework are void, metal, and (everyone’s favorite) fire. Fire is generally associated with destruction, but we have no reason to be tied to that, so let’s invert it. I touched on the possibility of temperature dangers earlier; perhaps there’s something the party could claim in the first area that protects them against the chilling cold in the second dimension, because the monster is draining the heat from the area for some reason. (I’m working on it.)
The fire item would also need some backup protective uses; fire/cold resistance is a bit lame alone and a damaging fire aura goes against the “protection” motif we’re going for. My two thoughts are either a mirage-ey thing where it makes people seem to flicker like flame, effectively giving a dodge action to one or a couple targets in exchange for your action, or some kind of non-damaging firewalls. Ideally, whichever one we went with would be connected to the cold protection offered. Something to mull over.
That leaves metal and void. Metal is traditionally a defensive element, but it’s also what we make weapons out of, so it shouldn’t be hard to make something effective at offense. Ideally, it would compliment the fire item, since fire and metal have a thematic connection (what with fire being critical to forging stuff). Maybe it creates some kind of damaging area-control effect that compliments an area-control effect the fire thing has? Oh, and it should solve some kind of logistical challenge or travel challenge or something, presumably related to one of the weird magical features of this dimension. Another thing to mull over.
Finally, we have void. This should probably be the last one, since it’s thematically linked to most things this monster could be. Also, it should absolutely be destructive (which incidentally means metal can’t be the second-to-last one, because I don’t want to have two fancy weapons in a row). Maybe its only use is being destructive, and it breaks the seal that lets the party escape or something.
But we don’t want the same five-part structure we had last time (four items plus escape). Maybe we could add a side section where the party fights some remnant of the bad guys who originally got them here, hopped up on space monster juice or something that makes it challenging eight levels down the line. That would give them a drive now that they realize escaping the belly of the beast might not be such a simple prospect, and also provide an opportunity to piece together the lore tidbits they’ve been missing or ignoring.
...It doesn’t take me long to realize this doesn’t fit well into the adventuring structure we’ve got here, so I decide to fold that idea into an item quest. Or maybe it’s the last thing they need to get to before they can punch a hole out and leave.
And now the second section looks something like this:
https://i.imgur.com/5UfdZMQ.png

That seems like enough for now. Next time, I’ll start filling in the holes I left in the key items.

A few quick notes, mostly for my own use:
I came across a third-level spell called catnap that gives the benefits of a short rest, quickly. This shouldn’t disrupt the alternate rest model, but it could give the “short rest in a can” consumable more mechanical grounding.
I also discovered AeronDrake’s Wilderness Survival Guide. Its resource-gathering system is more focused on alchemy than survival, and doesn’t allow much room for expansion (I’m imagining a random table that could spit out multiple results per PC searching), but there are some ideas in it I might tweak for my own uses.
Finally, I need to work in other magic items. this article (https://dmdavid.com/tag/what-is-the-typical-amout-of-treasure-awarded-in-a-fifth-edition-dungeons-dragons-campaign/) has a useful table for estimating how many magic items a character should have at any given level, and helpfully links to this (http://www.enworld.org/forum/showthread.php?437937-Magic-Item-Math-of-5e) analysis. They come to different conclusions about what a given character “should” end up with at a given level, but have broadly similar suggestions about how often magic items should be handed out overall.

[GreatWyrmGold]

Let’s discuss those Key Magic Items, since they’re going to be important for the themes and environment of our megadungeon. What we’ve figured out thus far:
Æther: Something that lets people heal despite anti-healy magic stuff.
Air: Some kind of flying thing. Probably a big flying carpet.
Earth: An earth-shaping tool that lets the party build shelter.
Fire: Some kind of protective thing that keeps the whole party warm.
Metal: Something destructive that synergizes with the above.
Void: Something destructive without much utility.
Water: Something that makes water.
Wood: Something that lets you cast goodberry.
Very detailed, clearly. Let’s work through some additional constraints and see if anything comes to mind.
First off, since the opposites of each element happen to be in separate halves of the megadungeon, we should do something with those contrasts. Fire and water should be opposites in some regard, as should air and earth, wood and metal, and æther and void.
Second, perhaps more importantly, we want these key items to have multiple uses (with the exception of the empty, hungry Void artifact). The flying carpet makes travel easier, but flight has uses in combat, and we might throw in a couple of neat tidbits besides. The water-making artifact makes water to drink, but also to attack with or something. And so on.
Third, the items should relate in some way to classic elemental themes. It might play them straight, or it might intentionally subvert them, but it shouldn’t do something completely random.

Let’s start with a pretty simple synergy between Fire and Metal I’ve thought of. Fire creates barriers that stop ranged attacks from passing through, Metal lets you make attacks at a distance despite this. Maybe have the firewalls block ranged attacks, and the metal attacks are chains or something that are just melee attacks with a long reach? Which could also help with climbing or something?
So we have a Hearthfire Shroud, which can create protective barriers of flame and also keep those near you nice and comfortable, and perhaps a Cloak of Chains which smashes people at your command like you’re Doc Ock or something. Cool, basic ideas settled.

The Cloak of Chains is a very physical item, and the Wood item casts spells. Goodberry at the very least, but probably others so the party doesn’t chuck it out once they realize it’s completely useless in this no-heal zone. Presumably, the other spells would similarly be ways to conjure plant material to indirectly affect things. Entangle, say. It’ll probably be a staff or a wand or something, with charges that recharge daily and can be used to cast spells. Let’s call it the Holly Staff for now.

Meanwhile, the water item needs to contrast the Hearthfire Shroud. Where the Shroud is defensive, the water item is offensive, some kind of weapon. Perhaps some kind of powerful torrent of water, smashing stuff? It also provides water, but I feel like it needs just a little something...extra to make it special. Maybe forced movement? Hm.

The Earth and Air items are mostly done. We’ve figured out that the former is a thing that makes earthen walls (perfect for improvised shelter or mid-combat cover), while the latter is a flying carpet that grants extended mobility and 3D maneuvering potential in combat. One is good for maintaining static positions, one is good for moving about.
Is there more I can do? Well, let’s start by considering the Earth item. I haven’t made anything that’s a classical weapon, so let’s make the earth item a nice maul. It does all this earth-moving by literally hitting the ground and making it do stuff we want it to do. And it’s limited to moving earth, not stone or metal or brick, so we can ensure the party has plenty of stuff to do with it without letting them wreck critical adventure locations too much. And maybe throw in a ranged attack option of some kind; maybe casting catapult or something as a Str/Con-based spell? Let’s call it the Earthshaper, a descriptive name that sounds moderately cool.
The carpet needs a little work. First off, we should probably try to make it easy on the DM; maybe make the flight reliable but a bit clumsy. It can’t rise or fall too quickly, for instance, meaning it has to stay at a roughly constant elevation above the battlefield (reducing, if not eliminating, the headaches that 3D combat brings). Second, we should give it extra goodies. The Earthshaper is primarily a defensive item; maybe the flying carpet has some offensive abilities? Actually, let’s drop the “carpet” thing and have it be a big cloud you ride around on. One that fires short-range lightning bolts, encouraging the user to stick close to the battlefield. There we go. Call it the Tempest or something and let’s move on.

The æther item is...vague. It does something to lift the anti-healing aura, cool. It also does something that lends itself to being used by some kind of madman, cool. Um…
Perhaps it just strengthens nearby magic in general, and it being used by a mage who’s using it to make all of their spells more terrifying. Simple. As part of this, it also lifts or resists effects that stop spells from working. I should look up what those are.
I’m imagining some kind of wizard’s hat, because there aren’t enough cool magic hats in this game. Maybe call it the Cap of Unfettered Thaumaturgy?

That just leaves the Void item. I want it to be dangerous but powerful, something you don’t want to use recklessly yet which is still too powerful not to use. I’m imagining a core dynamic of “drain energy from creatures, use scraps to power other stuff”. Something like throwing stuff into a black hole to get power from its accretion disk.
Maybe it fires off cone-shaped bursts of dark energy, giving all those creatures levels of exhaustion or necrotic max-HP damage or something. (Oh, and the initial vertex is the one to the user’s back, not their front; they’re always included in the blast.) For every level of exhaustion/point of damage/whatever, a certain amount of charge builds up; this charge bleeds away rapidly if not used, but can be used to power mighty abilities.
Cones are hard to make selective, so you’ll either need to plan carefully or accept that you’ll hit your friends. But the powers that the artifact grants...they’re destructive as well, but arguably worth it.
Let’s call it the Maw.


Alright, so we have our key items loosely designed. What the heck are they doing here?
Our ragtag band of heroes aren’t the first to come through here. Many people have been eaten, most unwillingly, but some intentionally. Perhaps some of these items come from adventurers who previously came here to destroy the beast. But where they failed, perhaps our band can succeed, by drawing on the knowledge and tools of not only those who would slay the beast, but those who would control it.
So these magic items largely come from others who came to destroy the undefined monster in the first place. Maybe this requires a ritual that consumes one item of each of the eight elements. That explains why so many are kicking around, and means the players will need to sacrifice eight of their items to progress to the Endgame. Of course, we won’t force them to give up their key items. A bit of diligence and luck will turn up some other elemental items that will serve just as well; the players just have to look for them.

It seems like a good idea to define the groups that brought these items into the megadungeon in the first place, so we can get a feel for the shape of this story.
The Cloak of Chains stands out. The symbolism of chains to represent bondage and enslavement is about as subtle as dressing your hero in white and gold. That’s not necessarily a bad thing, especially in a medium where most of the audience (and, let’s be honest, most of the writers) rarely thinks deeper than surface-level motifs. So one of the groups wasn’t trying to kill the space monster, they were trying to use the space monster. Maybe that’s the group that the party interrupted early on, or maybe it’s an earlier group that the modern antagonists stole notes from. Let’s throw the Earthshaper in, too; it feels loosely fitting, and we don’t want to leave them with just one item. We’ll call this group the Binders.
The Hearthfire Shroud, Holly Wand, and Cap of Unfettered Thaumaturgy feel like they all belong together. They’re all items that “feel” warm and helpful, more for supporting friends than tearing down foes. Perhaps there was a group of hedge wizards and village druids that came together to discuss natural philosophy and protect travelers and whatnot, who gathered their might against the space monster when it seemed a threat to their world. (Wizards and druids aren’t people you think of coming together often, but it’s not hard to think of ways to give them interests in common.) Let’s call this group the Scholars.
The Tempest feels like a raw chunk of elemental something. Maybe the water item (let’s call it the Torrent) could be the same, and both could come from a cabal of elementalists. Perhaps they were kindly, perhaps they were sinister, but either way they saw something involving this giant monster that would threaten their designs. This group is the Cabal.
There we go; three groups with interests in this monster. Binders, Scholars, Cabal. One seeks to bind its power, one seeks to protect the world from the threat it poses, one seeks to protect their own interests. Evil, Good, Neutral. And the elemental themes showing up through the artifacts and the Cabal make it clear that this needs to be some kind of vaguely natural monster. Perverted or cancerous nature, perhaps, but not some wholly artificial creation or godling. Something to think about.

This is probably a good time to start thinking about themes, too. The Binders and the actual antagonists (who may or may not be the same group) have a pretty clear villainous goal; take this majestic natural thing and use it for their own ends, regardless of the consequences for that thing or the world at large. Sort of like a high-fantasy Captain Planet villain. Meanwhile, we have a group of people who wants to use natural elements for their own ends in a different way, and a group of people who like nature.
So it looks like the big thematic challenge I’m presented with is “How do I work with this material without making it come off like a piece of schlocky save-the-rainforest crap from the 80’s?” Well, luckily for me, I happen to have a more nuanced 21st-century view of environmental issues than most people making green schlock in the 80’s, which acknowledges things like “you don’t need to be pure evil to hurt the planet”. So the first step should probably be defining motivations and some sort of worldbuilding thing.
Let’s start with a couple of sentences about the biggest real-world environmental issue, global warming. We don’t not quit fossil fuels because Big Oil controls the government or Big Auto is sitting on a patent for cars running on water that could easily earn them billions; we stick with them because they fulfil a big need in our world that no mature technology can replace. By analogy, perhaps all of this elemental stuff is critical to something that the D&D world needs to function.

Unfortunately, most D&D worlds are stubbornly resistant to making them reliant on magical MacGuffins; they usually cling closely to European feudalism with a tablespoon of Greco-Roman and a dash of “exotic” culture for flavor...even when the powerful monsters and incredible magic should logically change that. Either we need to set this in a unique campaign setting, or...I actually can’t think of a good alternative. So yeah, unique campaign setting. Set the actual action deep enough in the boonies that the game can be cut and pasted into the boonies of any setting, but it needs the bones of its own setting.
So we have a setting that’s become reliant on elemental magic for all the cool stuff it can do, and we need a sub-setting which shows off why the things that elemental magic can do are not inherently bad even if the ways we power them are harmful. How about a polar region with Wood-elemental fertilizer, Fire-elemental “sunlamps,” and so on making small areas of fertile farmland? That way, we can show the wonders of our elementalism before introducing the party to the dangers and drawbacks it has.
This does mean we have to add an additional element or two to justify the PCs needing to live off the land, but not too much. The farmland provided by elementalist devices is hardly cheap, so it’s hard to rapidly increase agricultural capacity. The PCs arrive the winter after some disaster down south sent refugees fleeing to the new oases up north; their storehouses are enough to keep everyone fed for the winter, but it’s close, so food prices are inflated accordingly. (On the bright side, prices for other things have dropped, and there’s no shortage of hunters and trappers who can give the PCs some tips in exchange for a couple days’ labor.)

Now we need to figure out how elementalism works, in broad terms, and what the three big factions want to do.
Elementalism draws on raw elemental energies, distills them into eight components, and uses them to power various processes. The obvious consequence of this would be “Taking the energy makes things die,” but that seems a bit...cheesy to me. How about something a bit more air-pollution-ey: Processing and using the elemental energy is an inefficient process which leads to bits of element escaping the machines, which clump together in elementals. These elementals aren’t good; they’re not giant rampaging monstrosities, but they’re more than enough to burn down sheds or get grain damp enough to rot or whatever.
The Cabal’s motivations are simple; they run the elemental extraction processes, and want to keep their wealth and influence. But if the space monster is a bundle of elemental energy, their desires should be basically the same as the Binders’. So what if the space monster is more akin to nuclear energy—a viable alternative, if handled carefully? (Maybe a godling or something would be a better choice.) Then the Cabal would want to shut down anything involving the space monster, the Binders would want to use space monster juice to fuel their evil schemes, and the Scholars would want to keep either group from screwing the proverbial pooch.

Where does the party come in? They show up to investigate some weird goblin activity or whatever, which turns out to just be the tip of the iceberg. Already, the antagonists are using slivers of space monster magic, but their plans are not quite ready. They have some big ritual which requires bringing the whole thing into the Material Plane for but a moment, giving it a sacrifice, and taking its gift. But the PCs have learned the bad guys’ plans, and accidentally disrupt them just enough to let the monster wander a bit in the Material Plane.
Luckily, they’re in the middle of nowhere and the monster isn’t as actively insatiable as most. It crawls through the skies, devouring only the occasional village, and the ruins inside the creature’s gullet dimensions help them piece together what it is. So once they escape, they can consult with scholars to figure out what to do, do The Ritual, drain eight elemental artifacts of their power, and dive into the breach for a final encounter.
But what do the PCs do then? I’ll give them a last-second choice, seeded ahead of time so they can ponder it. Do they destroy the power source at the space monster’s heart, like the Cabal wants? Do they take the monster’s power for themselves, like the Binders wish they could? Or will they send the power to sleep and trust the Scholars to use it to make the world a better place?

Meanwhile, the party comes to realize the consequences of elementalism. It brings great prosperty and happiness, but it comes with great danger. Elementals show up in the wilderness, which is weird, and also in the gullet dimensions. Maybe there are some big elementalist mechanisms left behind in one or both, still running off of some source of power or another, and shutting them down means elementals stop showing up in random encounters.The players will hopefully put two and two together and realize that all the elemental mischief that’s been mentioned (hopefully) since day 1 has been caused by the elementalism all around them. This should, hopefully, contextualize the Cabal and the consequences of either replacing the current wonderful power source or letting it stand.

Alright, I think that’s enough thematic plotty stuff for now. I should probably get back to spreadsheets and proto-maps next time.

[Squire Doodad]

Didn't quite read all of that but the Aether sounded interesting.
Aether is sometimes portrayed as pure magic (when Thauma is not a thing on its own). Alternatively, it is spirits and thoughts: this means it can allow the player to make their thoughts manifest, possibly allowing them to cast magic with their intention instead of just magical power. Let's say you want to hurl a fireball at some goblins to spook them. Normally, you hurl the fireball, it hits the area behind the goblins, they get scared. But an item of pure Aether, they can hurl a fireball which explodes with a blaze that has the same effect as a normal fireball in terms of raw numbers, but for the goblins it is more likely to scare them. Basically it would translate to "+(#) to relevant checks concerning magic". If you want to dazzle the crowd, or purify water so it tastes like fresh glacial meltwater, or heal something such that it heals a wound, you can make use of the fancy doodad to get a bonus.

If you want to make it be raw magic, I'd suggest having it be that the player can, once per extended rest, regain some spells at any point they want.
A possible permutation:
The player rolls 1d6 (maybe 1d6-1 so its a range of cantrips to 5th level), and then rolls a d4. The d6 determines the spell level; the d4 determines the amount of spells they recover. So if I have a Wizard using the doodad, rolls a 3 and then a 2. The Wizard will be able to choose 2 of their 3rd level spells and be able to use them again. You might want to put in some clauses to nerf it a bit, but eh.

[GreatWyrmGold]

Alright, so. Mapping. It’s a funny thing. Judging by the published adventures I’ve read, most TRPG designers seem to think you need to start with a carefully-designed schematic of what a place should look like before you consider how it’ll play in-game. For a while, this made perfect sense, but it doesn’t hold up that well after decades of video games showing that you can get the same thematic effect even if you start by designing how it’s supposed to work in-game. So let’s start with that.
But before we decide how everything’s going to fit together, we need to decide what those things are and what parts are fitting together. To decide that, or even explain what I’m talking about, we need to discuss how I imagine the inter-dungeon gameplay loop is going to work.

Remember when I mentioned survival elements? I’m imagining it playing out a little like a board game. The players have this region map, probably a hex-map, and they have a marker for their group and maybe for each character who’s splitting off to do their own thing. They move about, roll to see what they can scavenge, and move on. Sometimes they roll a random encounter. After some in-game days of this (which hopefully goes by quickly), they reach one of the obvious landmark dungeons scattered around the map and go into normal-dungeon-delving mode long enough to plunder its supplies and tools before moving on. Maybe throw in some microdungeons of one or two encounters plus some precious supplies around the place.
I’ll need to balance the scavenging system so they get about the right amount of supplies per day. Ideally, a party will be finding almost enough supplies to survive a day for each day they just scavenge random tiles, making up the deficit by scavenging the dungeons and whatnot scattered throughout the land.
First question, what size do we make the hexes? The overland travel rules give hourly paces of 2-4 miles per hour, or half of that in difficult terrain. If we measure movement hourly, we would pretty much require 1-mile hexes. This would mean that, indecent terrain and at a moderate pace, the party could cover 24 hexes every day, or more than 70 per adventuring cycle (assuming short rests are nightly rests and long rests are days off). The hex maps in the Traveller book I happen to have on my bookshelf has pages eight by ten hexes in size; this means that a cycle’s travel with single-mile hexes would either require impractically large maps or impractically small hexes. This will never do.
Alright, let’s consider potential daily paces. 18, 24, and 30 miles per day, and potentially half that (9, 12, and 15 mpd) through difficult terrain (or increased with vehicles/mounts/Tempest, but that’s less important). The only common denominator of those six figures is three, so let’s consider what three-mile hexes would imply. At a slow pace, the party would go through three or six hexes per day; fast paces through good terrain would cover 10 hexes per day. This means that a day’s travel would always fit on a Traveller-page-sized map, and an adventuring cycle’s could under good conditions.
It might be a good idea to add a rule that you can only scavenge at slow speeds. That makes sense, forces the party to decide between speed and sustainability (until they have supplies to burn), and helps us keep maps smaller.
Why did all of that matter? Considering the constraints of our map will let us design that map without violating those constraints. Our map design won’t jump straight to hexes, but some of the stuff we discussed is going to be relevant (I think), and I wanted to get it all out of the way at once.

Alright, let’s discuss maps and proto-maps. The way I see it, “maps” are the finished product, one which contains all the information for every tile in a neatly organized format. But that includes a lot of information that doesn’t really matter, like “How far is it from this arbitrary point nobody cares about to this other arbitrary point nobody cares about?” What matters is the arbitrary points with things people care about them, how far it is between those points, and any obstacles between them. Proto-maps let you figure those things out without being bound by things like “fitting into tiles” or “figuring out what everything is”.
Let’s start with the first gut dimension. It’s divided into four or five regions, with a big feature or two each. The first region or two is centered around a big, obvious spring where the PCs can get water. Actually, maybe it’s the remains of some big water-making artifact, from some previous attempts to kill the space monster. Anyways, it’s obvious and the PCs should make their way there once they realize they have limited water supplies. Then it serves as a hub until they find the artifact that lets them make their own water.
How far can the spokes be? Well, most adventurers can be expected to own a waterskin, which holds two quarts of liquid. Adventurers require one gallon of water per day, or two in hot weather. (I’m inclined to make the first gut dimension hot, both to contrast the chilly second one.) So it’s clear we’ll have to provide adventurers with extra water containers, which in turn means we can control the length of the spokes. To an extent.
This is a good time to discuss my fairly laissez-faire attitude towards the limitations we’re setting up. I’m deliberately making them soft limits rather than hard. It’s possible that players will have chosen the Tanning skill and start making waterskins out of monsters, or that they’ll find a way to turn half-broken water barrels into portable water containers, or that they’ll gather supplies near the spring for days until they have enough for a fast-paced expedition to the most distant fortresses. That’s fine; if the players go that route, more power to ‘em. It’s not going to throw encounter balance off too much; that’s one of the benefits of the ratcheting difficulty I implemented. I’ll need to throw in something that stops them from casually hopping into the second gut dimension after just a few encounters in the first, though, or they’ll get murdered. But that’s a design challenge for Future!GWG.

Present!GWG, on the other hand, needs to make a proto-map figuring out the approximate distances between the big dungeons and the spring-hub. And for the first draft of said map, we have...
https://imgur.com/zWXPsPb
What, are you not impressed?
In all seriousness, this is a pretty small step forward. It’s just the first step in an iterative process, but I’ve realized that it’s as far as I can go until I break down the areas a bit more. Whoops!

We have a certain number of short-rest cycles per segment. Some of these are going to be bundled into minidungeons, some as separate one-off microdungeons. Let’s see how many short rests we have per big reward:
Holly Staff/Spring: 4 Cap of Unfettered Thaumaturgy: 3 Earthshaper: 5 Torrent: 4 Exit or Dimension Key or whatever: 5
So let’s divide them up as:
Holly Staff/Spring: Two μDungeons, two in minidungeon C.U.T: Three in minidungeon Earthshaper: Three μD, two in mini Torrent: Two μD, two in mini Exit: Two μD, three in mini
Alright, now that I know how many minidungeons I’ll be scattering around, I can put a slightly-less-rough draft together.
https://imgur.com/WfCtiNh
Alrighty then. Next, we need to decide what’s going on at all these minidungeons and microdungeons.


First, let’s consider the intended route. The Cap’s minidungeon is intended to be the first thing players gravitate towards after they get to the spring, so they can get their healing magic back. But on our map, it’s far away from all the other structures. Why should they go for the big, isolated structure instead of the chain of closer microdungeons?
I mentioned earlier that the Cap would be used by some kind of positive-energy-wielding maniac. What if that maniac is basically a cult leader, and their minidungeon is a small town? That should catch player attention! They trek out into the wilderness towards the one sign of active civilization in this place, only to discover that it’s as hostile as the wilderness. After fiddling with the XP and encounters, the party looks like they should arrive at the cusp of level 5; hit them with a tough encounter as they enter town, watch them board up a convenient building, and let them retaliate against these people with newfound power.

Alright, with that out of the way, we can go back to the very beginning. At this point, I’m imagining the first minidungeon (the one with the spring in it) to be some kind of elemental machinery gone wrong. Maybe some enterprising elementalists saw this big source of pure energy and experimented with draining part of it for use elsewhere. Unfortunately, it turns out that the inside of a space monster is not a hospitable work environment, and they had to pull out after establishing some basic facilities in the first dimension.
The most impressive of these facilities was intended to provide food and water to the on-site workers, harnessing water and wood elemental energies to make rice patties or something; over time, these devolved into algae-covered lakes and rivers flowing to the edge of the floating island thing, and the systems for siphoning the other six elemental energies off for whatever have broken down, releasing a variety of elementals over the region. (Which reminds me, I’ll need to find or make æther, metal, void, and wood elementals at some point. But that’s down the road.)
From the distance, the party sees a distant ruin surrounded by lakes and rivers and pond scum, with some inexplicable streams of elemental energy (constant firestorms, piles of dirt and rust, strange weather, etc). It looks dangerous, but it’s nearby and it shouldn’t take the party long to recognize the value of the only visible bodies of water around.

The area by the Earthshaper has a lot of microdungeons. Remember how I was talking about societies evolving to cope with the environment inside the space monster’s gut dimensions? That seems like a good place to put them. Maybe there’s some feature of land, a something-rich ridge or a river valley or something, that makes living there more practical than elsewhere. Maybe it could just be some kind of defensible terrain, if there’s enough of an ambient threat from æther-cultists or elementals or space-monster-antibodies or something. (Which there probably is, judging by the sheer number of expected random encounters.)
What’s the minidungeon with the Earthshaper, then? The details are going to need to wait until I figure out what kinds of society exist inside the gut dimensions, which will need to wait until I finally decide what the space monster is, which I will get to sooner or later. But it’s going to be some kind of center for the locals, a citadel or temple or royal manor or something.

The Torrent is associated with the Elemental Cabal, so what’s it doing at the opposite end of the adventure-space from their base? Perhaps when the Elementalists pulled out, some stayed behind—a mixture of those who too passionately believed in the project and those without the means to escape once the elementalists got everyone they cared about out.
Some stayed by the ruins of their old refining/food-making facility, while others were kicked out for some reason or another. Some joined other communities or became hermits or raiders, but a bunch gathered around a few powerful mages (druids? sorcerers? awesome bards? dunno) who were gathering the elements themselves as weapons and tools. The Torrent is being held by one of those.
I imagine that the Tempest got to the second gut dimension when a powerful elemental cultist set out on an expedition there—perhaps a rebel bringing people unhappy with the state of things to what they thought was outside not long after they were stranded, perhaps a zealot generations down the line making a pilgrimage to the heart of the beast. Either way, they never returned.
I imagine some of the microdungeons scattered about are hostile villages, but some would be ruins of villages and fortresses. Heck, one or two might be active fortresses. Maybe the descendants of the elementalists holed up with the Torrent are at war with the natives (well, here-longers, if the distinction matters) that hold the Earthshaper?
Some might also be from the modern antagonists, whose dungeon-bits crash-landed all around.

The Exit is presumably near the edge of the floating island all of this is on, and it seems like a good place for the antagonists to set up shop. They need to get out, too, just like the PCs. (Also, they’re the best climactic encounter, but hush.) Maybe their minidungeon is a chunk of the first dungeon on its side or upside-down or something, collapsed and refurbished to make it work better as a dungeon headquarters? An idea to keep in mind when mapping things, well down the road.
The microdungeons around the Exit could be more chunks of dungeon repurposed into huts or whatever. This whole section of the megadungeon is going to have a bit of a “scavenger world” (https://tvtropes.org/pmwiki/pmwiki.php/Main/ScavengerWorld) vibe, at least in its architecture, but the dungeon bits that the antagonists had to recently and rapidly convert into living quarters will be the most obvious.


Next is the second gut dimension. We had a decent idea of what we wanted out of the first, but let’s nail down some details about the second.

Again, we have five sections, four with magi items at the end and one distant one which has only the way out. The big environmental hazard that’s overcome fairly quickly with a special item is the biting cold, which requires fuel. This means there can’t be structures full of wooden support beams and furniture, nor forests of flammable flora. But I want this realm to come across as nature reclaiming runs, to contrast the first dimension (with its buildings actively destroyed by violence) while still being thematically connected.
I’m imagining something like a huge mountain range, a jagged lump of stone and other nigh-indigestible stuff that the space monster is squeezing every drop of energy out of. Like real-world mountains, it’s covered in hardy plant life (lichens and mosses, some grasses, the occasional shrub or small tree). Unlike real-world mountains, many of the peaks are oddly square and some boulders are recognizable as heavily-eroded statues or keystones or whatever, and while no buildings remain intact, a keen eye notes what looks like petrified villages razed to their foundations, tilted, and overgrown. Empty spaces in patterns like lost houses, granaries, churches, roads, etc.
Maybe not “petrified,” that’s too associated with elemental earth, and we want this to be elementally agnostic. Maybe ash or dust? Something crumbling. And the whole place is haunted by the incarnation of the space monster’s digestive tract, specters that presumably transform inorganic matter to whatever the mountain’s made of and do something bad to living things. The energy they get, they bring up to some other part of the space monster’s biology, which looks like a cold sun hanging in the not-sky.
Or maybe living things are supposed to die before coming here, and took root as parasites. So add some kind of immune response to the random encounters list. A different kind of spirit, blazing bright like the first gut dimension, draining energy from life which doesn’t somehow avoid it.

Yet life found a way, as did some sapient beings—at bare minimum, enough to bring four artifacts into this dimension. Who were they?
The Cloak of Chains was central to deciding that there would be Binders, so the place where the party finds it should be important to learning about what they want. Maybe they were some past group paralleling the modern antagonists, who want similar enough things that their records were valuable to the antagonists. Some of the more powerful ones (including the big boss) slipped between dimensions, and some immediately gravitated towards this...place.
What is it? Well, it used to be some kind of laboratory or wizard’s tower or ritual site or something like that, with the Cloak of Chains intended as part of some ritual to do whatever it is they wanted to do. Probably smash the world back to the Stone Age or something. But they failed; whether this was due to adventurers, logistics, or the space monster’s inner monsters should be kept ambiguous. The true story was wiped away with the passage of time.
But there are still enough tools, books, etc for the prime antagonists (who really need their own name) to make use of. Combining these archaic tools with modern magicks, they’re preparing to hijack the ritual locations and complete their own plans. So the Cloak is at what was once the prime ritual location, near the center of the array, with the other ritual locations being minidungeons scattered around the megadungeon. But as I decided about a sentence ago, the Binders came and went before anyone else of note entered the space monster, and the other ritual locations were inhabited by later interlopers.
The PCs will crash near the Hearthfire Shroud’s resting place. Likely, this ritual site was converted into some kind of shelter by the Scholars. Maybe they were planning to evacuate settlements in the first Gut Dimension, or followed victims of the space monster into its gullet. Either way, they had to abandon the place quickly, and left behind some crucial gear...including the Hearthfire Shroud.
The Tempest came from the Elemental Cabal. It’s a tool designed for moving around rapidly; perhaps some scouting team trying to figure out if this monster was useful met an unfortunate end. Perhaps there was a traitor, who saw more value in trying to forge a new empire than rising slowly in an old one. This traitor took control of their device of elemental air, and sought to exploit the local resources to gather power for some unknown end—likely just power’s own sake, like the Cabal before them. But there wasn’t much power available. Maybe the Cabal guy became some kind of storm-lich and still rules over the descendants of their expedition and/or locals, or maybe the traitor’s descendants continued to claim right to the elemental air and use that to rule over whoever lives there.
The dungeon exit is not located at what was originally the ritual heart, but the current antagonists repurposed the rituals to something which made a different ritual site the nexus of the whatever. The means for their ends is different, I’ll figure out the details once I’ve figured out those ends and just what this space monster is. It’s where the big boss has been hanging out, managing something or other important.

That just leaves the Maw, which is tied into the ideas I’ve had about the “natives”.
Their ancestors came from the first gut dimension, whose inhabitants were already tough. But this world is worse. You can’t grow wheat on a mountainside of eldritch dusty stuff, and the monsters here are far more actively malicious. I imagine they mostly survive by herding some kind of hardy livestock (probably a goat breed) that can survive off of the meager plant life in the area, wandering from slope to slope in search of something to supplement their diets of goat milk and meat.
They see two common classes of supernatural creature. One has no interest in living things, the other is actively hostile to them. They came to see the specters draining energy from dead matter and “petrifying” it as angels from a desperate god, who was protecting mortals from bogeymonsters living in the ground. Their burial rites involve leaving the body of the deceased in the open for the specters to claim, both to strengthen their god and to ensure the deceased’s spirit will not feed the ground-monsters. (This may come after a bit of ritual cannibalism, given the poverty of the region.)
The Maw is a chunk of the space monster’s hunger. To the locals, it is a shard of sunlight and holy weapon against the monsters. Perhaps they’ve even found a way to use the energy produced by draining energy from said monsters to feed their people, letting a small holy city-state establish itself among the goatherds and hermits of the dimension. If the PCs are of a roleplaying breed, they might try to convince the locals that they deserve to take up the holy weapon against enemies of all humanoidkind; otherwise, they can engage in the standard adventurer diplomatic procedure.

Alright, now for the actual proto-map.
https://imgur.com/s6PjZCv
I left off travel times this time, because there’s no gating based on travel times. Instead, we’re gating by travel abilities! I’m only putting hard locks on a couple of adventure areas, to make the area feel more like one unified area than a bunch of discrete instances, but there will be smaller obstacles that movement abilities will make it much easier to get around.

We’ve already figured out what the minidungeons around the major items would be, but what about the microdungeons scattered along the way? I imagine most would be minor ritual nodes being manned by a few antagonist goons, but there should be some variety. Maybe some are ruins from previous groups who got stuck down here. That’s especially true of the second one by the Tempest, which is a secondary site used by the Cabal-traitor offshoot. Presumably, the primary antagonists were working on kicking them out of their home and the natives out of their temple-city when the PCs stumbled into the dimension. Also, the line of microdungeons leading to the Hearthfire Shroud might just be along a reasonable path down a major peak that the PCs wound up on top of.

Alright, that should do for this post. Not sure what the next step is; maybe “Reread the megadungeon series to check for other critical groundwork”. Or maybe it’s “Fill in some of the critical story details you’ve been saying you’d figure out later”.

[Segev]

Not read all of this yet, but wanted to comment because I want to keep my eye on it. This is interesting.

Of the first 1/3 or so that I read, I will comment that you may want to rethink the obvious railroady nature of the transition from ritual to skydungeon to dreadnaught-devouring. This kind of thing works in a video game, because players of video games have come to expect that the plot advances at a pace set by their physical advancement through the dungeon, and that there are set pieces they will see but do nothing about. It gets ... cynical reactions even there, but in TTRPGs, it gets downright nasty.

I recently had a thread in the 5e forum on how disappointed I was in a DM's Guild module for Hisari, and one of the reasons was the no-save-you're-just-screwed cutscene that forces the PCs into the latter half or so of the module.

One way to create at least a sense of agency, even if they're being guided to these choices, is to have the rising into the sky be something they trigger, perhaps to thwart the ritual. Maybe the Dreadnought was buried in a sealed vault beneath the dungeon, and the ritual was designed to call it up and keep it chained. Separating the dungeon from the ground seemed like it would disrupt the ritual, but also freed the creature. Maybe the dungeon was originally an earthmote used as a magical seal to hold it in place.

Luring the creature up to the earthmote might keep it from devouring the kingdom surrounding the dungeon site, or something.

Also, "the party can't actually stop the ritual" is a guarantor of frustration, I think.

...this feels pretty contrived, still, sorry. I'm just trying to brainstorm ideas to make it feel less like a videogame sequence where the PCs don't have any agency in actually winding up in the megadungeon.

You CAN get away with denying agency, but it works best if you do it in a "you start the game in this situation" way, rather than starting them NOT where you need them and then forcing them down a path to get them to where you want them.

Even Angry's megadungeon has a massive flaw in that it assumes the PCs will want to come back and explore, and will use exactly the path he's laid out for "day 2." That said, he put a lot of effort into concealing the rails, making the trap that drops them into the catacombs feel like something they "fell for" (pun totally intended) rather than something he forced on them.

On themes and concepts, you might also look at the 3.5e MMV and a monster called "Dalmosh," which has a "stomach dimension" and thus may serve as some sort of inspiration or counterpoint to the modified Astral Dreadnought.



Anyway, I'll come back and read the rest of this in a bit. Definitely going to follow this. Design projects like this are fun. Good luck!

[GreatWyrmGold]

Of the first 1/3 or so that I read, I will comment that you may want to rethink the obvious railroady nature of the transition from ritual to skydungeon to dreadnaught-devouring. This kind of thing works in a video game, because players of video games have come to expect that the plot advances at a pace set by their physical advancement through the dungeon, and that there are set pieces they will see but do nothing about. It gets ... cynical reactions even there, but in TTRPGs, it gets downright nasty.

-snip-
Yeah, I'm not 100% happy with how that plays out. I do need the players to start in the regular world (because dumping them into a survival scenario without any practice or a support system in the nearby village to fall back on is just asking for a TPK), and I want to guide them through the transition slowly (with the sky-dungeon being intended as a way to get players used to camping out in hostile territory before getting dumped into supernaturally hostile territory). I also need them to end up on the other side, because that's where all the "figure out how to defeat the monster and gather the power to do so" stuff is lying.
The thing about railroading is that it's also connective tissue. I've played and DM'd adventure paths which had so little "railroading" that it was almost impossible to guide the players to the important places and events.
It's another one of those areas that I hope to flesh out in a way that dampens the pain of those bones. Which is why I'm keeping that section to the barest possible bones for now and focusing on the parts less likely to interfere with that particular flesh.

The idea of having the levitating dungeon be something they trigger to interfere with the ritual is...something I can work with. An idea I was trying to convey was that the PCs try to solve the problems in a straightforward manner, without any understanding of the situation, made things worse in a way that they can only solve by learning about the situation at hand. Having their initial solution (disrupting the ritual) accidentally release the monster that eats them, rather than just make it impossible to control the monster, is probably a better way to convey that, and one that doesn't stomp on player agency quite as badly.
Still a narrow tightrope to tread, but it seems doable.