Happy Megadungeon … whatever the f$&% day it is. I think it’ll be Tuesday when this is done. And then the schedule should regulate again. But let’s not waste any words on that. We have more important stuff to cover.

Much as I want to jump right in, I want to mention a few things about this specific article. First of all, I’ve received some criticism for this series because it is about some very advanced adventure building stuff and not all of it is applicable to smaller adventures. Now, sometimes when I say things like that, I get a few folks in the comment who feel the need to respond to my critics and say things like “well, you TOLD everyone that in the introduction” and “why are people dumba$&es” and “don’t stop because I like this even if stupid people don’t.” So, first of all, calm your tits. Criticism is no big deal. I’m not going to stop or change the series. It’s just that every so often, new people stumble on my website and start reading in the middle and don’t know what’s going on. Or others miss the introduction. So, after this week, I’m going to start putting a little disclaimer on this series so folks know what’s what and link to the category page so people can start reading in order.

But THIS article is a little different because this one is going to discuss a NOT-SO-ADVANCED technique for building adventures of ANY size and we’re then going to use it to build something smallish that we’ll use in this adventure. We’re then going to keep building the same technique to build more complicated things in future articles. So, even if you ARE turned off by this complicated stuff, stick around for this one.

This article will also be different because we’re going to start off by discussing video games. Now, I know what you’re thinking. You’re thinking “that’s ALL you do. ALL you do, Angry, is try to turn your favorite Metroids… I mean video games… into a dungeon adventure.” Well, shut the f$&% up. Sometimes we talk about Dark Souls and Arkham Asylum. But this one is different because we’re not discussing any of the video games we’ve talked about as inspiring the overall feel of this series. Instead, we’re going to talk about an important OTHER lesson from SOMEWHERE ELSE. Whoooooooooo!

This article is going to be MOSTLY theory with just a little bit of mechanical building at the end to show off the basic idea. But we’ll keep building on this idea to build all sorts of stuff for our dungeon. So, it’s important.

So let’s get theorying!

Mega Man and Scenario Design

Let’s talk about a video game series that totally isn’t Metroid or Dark Souls for once. Let’s talk about Mega Man. The original Mega Man. The REAL Mega Man. Not X. Not Net Battler. Not that f$&%ing Mighty Nozzle Number bulls$&%. Mega. F$&%ing. Man.

Many many years ago, the Mega Man series appeared on the Nintendo Entertainment System. And you can replay the original games right now on Xbox and Playstation with the Mega Man Legacy Collection. It’s totally worth it. They are action platformers. 2D things. You are a robot. You move from left to right and sometimes up or down. You jump over pits and spikes. You shoot your gun. And you fight super robots and steal their weapons. It’s really cool.

But what sets it apart from other action platformer games with linear levels in which you jump and shoot and blow up evil robots is its level design. Also called scenario design or environment design, level design is actually a part of game development. If you’ve used Mario Maker, you know what I’m talking about.

Level design isn’t really about nuts and bolts game design. Instead, it’s about taking all of the various bits and pieces of a game – enemies, obstacles, game mechanics, and so on – and putting them together to create the actual parts of the game people want to play. In RPG terms, scenario design is what we’d call stocking the dungeon or building encounters or whatever.

The Mega Man games are notable because they have EXCELLENT level design. And part of what makes their level design so good is actually part of why most GMs don’t even realize level design is a thing. And it’s why even when GMs try to consciously avoid the “kitchen sink” approach to dungeon design, they still fail to create something that feels good.

If you’ve never heard the term, let me explain “kitchen sink” dungeon design. In the early, early days of D&D, most dungeon adventures were just endlessly sprawling complexes filled with stuff. This room had rats, that one had spiders, that other one had goblins, there’s an ogre here, and over there is where the gnolls hang out. This room has a pit trap, fire trap protecting that treasure chest, scything blades in the hall, poisoned darts in the foyer. The dungeon contained “everything but the kitchen sink.”

While D&D tried very consciously to move away from that approach because of the realization that it’s actually s$&% design, it’s actually really hard to get away from the “kitchen sink” feeling in D&D without making things very small and self-contained. The first attempts to move away from “kitchen sink” dungeons were “kitchen sinks with an excuse.” For example, the famous adventure B1 Keep on the Borderlands features a pile of caves called The Caves of Chaos that are very much a “kitchen sink” dungeon. BUT, the adventure very clearly says that it isn’t a “kitchen sink” because a few lines of text talk about why the monsters all live near each other and how they react to each other. The Village of Homlett and the other adventures in the Temple of Elemental series begin with the premise that a powerful, evil force literally created temples to gather all different types of monsters in one place, and that is why gnolls live a few doors away from a f$&%ing cockatrice.

Modern adventures still OFTEN feel a little “kitchen sinky” even when they aren’t. Especially homebrew dungeons. And the reason for that is “variety.”

If you look at any level from the first six Mega Man games, what you will find is that there isn’t ACTUALLY a lot of variety inside the level. There’s only a few different types of enemies, one or two gimmick traps or obstacles, and that’s it. But it never FEELS that way because the game takes those few obstacles and enemies and plays mix and match.

In the first room, you fight a robot bats. In the next room, you jump some pits. In the next room, you jump some pits while robot bats are fighting you. In the next room, there’s frogs that shoot smaller frogs. Then, there’s frogs that shoot smaller frogs while a bat or two come at you. Then frogs with pits. Bats with frogs with pits. Then the boss. And when you combine those things with environment factors like large rooms, small rooms, flat rooms, labyrinthine rooms, and so on, you get a huge amount of variety from a very small number of individual elements.

Now, most GMs don’t design dungeons like that. Instead, we tend to think of dungeons as rooms to fill with stuff. We have ten rooms, we need ten things to drop in those rooms. This is the bat room. This is the frog room. And this is the pit room. And then… well, after that, we don’t want ANOTHER bat room. Or another frog room. Maybe we can do one room with bats and frogs. But that’s it. We’re still six rooms short of filling a dungeon. And so, instead, we start looking for things that we can put into other rooms. Maybe variants. This is the room of the one giant frog. This is the room with the fire bats. If we’re going to allow fire bats, we can allow a fire trap. And suddenly, the dungeon starts to feel “kitchen sinky” even though it really isn’t. Except it is.

As GMs, we tend to value variety. We don’t believe the players want TWO goblin rooms. Or even THREE or FOUR. And we tend to forget how much of encounter design depends on the environment. So we can’t even fathom that two rooms with the exact same goblins can FEEL very different depending on the terrain and the tactics. And, as GMs, we don’t have any limitations. Ten rooms with ten different types of enemies isn’t any harder to pull off than ten rooms with identical enemies. We have giant books FILLED with monsters and traps. We can easily create more. There’s very lost cost for variety.

Video games aren’t like that. Old video games had aggressively tiny limitations on space. You were limited to a certain number of enemies because the hardware would literally explode if you tried to cram one more. So many old cartridge games were designed to trick their way around hardware limitations. And even without severe hardware limitations, it takes a lot to add a new enemy or trap or whatever to a game. You can bang out a new monster in 30 minutes in D&D. But to create a new enemy in a modern video game takes a LOT of resources. Design, programming, animation, testing.

But so what? Isn’t it a great strength of our medium that we can do whatever we want with very little cost. We CAN have variety. We DON’T HAVE TO repeat stuff. Why SHOULD we?

Why Mega Man is Better than Keep on the Borderlands

It is ALWAYS better to build a level from a small collection of assets mixed and match together than it is to build it from anything you want. ALWAYS. The trouble is, the reasons are not always obvious.

First of all, remember that D&D is a tactical game. One of the things that makes combat interesting is that the players can make lots of choices about how to engage. Do we close with those enemies, do we let them come to us, do we use ranged combat, melee combat, do I spend resources on a burning hands spell or just throw a cantrip, and so on and so on. Honestly, if you build combats that are tactically interesting, you’ll find most players respond really well. I’m watching a group of complete neophytes to D&D RIGHT NOW who are mainly into story and character nonetheless get excited by exploiting the tactics in the combat sections because they are learning how to exploit their tactics.

But to allow for that kind of interesting tactical play, you have to build for it. Let me give you an example. In a recent game I ran, I included these dinosaur-like raptor enemies. They were fast, but they were really good jumpers. And if they could cover enough ground before their attack (by running or jumping), they did hefty extra damage. Now, tactically speaking, that isn’t too complex to deal with. You don’t want to fight them at range because they will close quickly and do extra damage in the process. Instead, you want to get right in their faces as quickly as possible and pin them down. Or slow them down with spells or other effects.

The problem is, players generally won’t figure that s$&% out in the first fight. Or they will figure it out late in the fight, after they’ve been screwed over by the charge and jump. It wasn’t until the SECOND fight with those bastards that the players really exploited the tactics. And they felt really good about winning that second fight. Because they took far less damage.

You really can’t make any combat tactically interesting until the players start to understand how their enemies actually work. The first fight with a new opponent is never the most interesting one. The same is true for traps and other obstacles. The first pit trap is a surprise. Either it hurts the players or it doesn’t. When the party spots the second pit trap that completely blocks the hallway they need to go down, that’s the interesting one. That’s the one where they have to figure out how to cross 15 feet of hallway without touching the ground.

Speaking of pits, the third fight with those lizard monsters happened in a room that was divided in half by a ravine. The monsters could easily retreat across the ravine with their prodigious jumping skills. They started disengaging, jumping across the ravine, and then leaping back into the fray the next round to do massive damage. Suddenly, the ranged combatants became more important again with those enforced hit and run tactics.

With one ravine and one lizard monster, I created three fights that got progressively more interesting. Hell, imagine those lizard monsters in a room with pit traps.

Now, imagine I have two different lizard monsters. One spits acid. One is a jumping pouncer. Do you charge the acid spitter and leave the pouncers free to run around? Do you pin down the pouncers and suffer the acid attacks? What if those creatures have a ravine. How many different encounters can I build with acid spitters, pouncers, and ravines? I could fill a ten room dungeon with tactically interesting challenges with three elements.

And it isn’t always about adding challenge. Imagine the satisfaction when the PCs enter a small, constrained room with a couple of pouncers they can easily pin down. Suddenly, they feel powerful.

You don’t get that if room one is pouncers, room two is spitters, room three is a pit, room four is kobolds, room five is a momma dinosaur, and room six is a kobold sorcerer.

And you might say “well, it’s hard to design tactically interesting encounters, I’m not good at it, so variety is easier.” Sure, it is. But if you impose a hard-and-fast rule that you only have two monsters and one environment hazard to use and you have to fill eight rooms, you GET good at it. Because you have to. Limitations and handicaps force you to design better things.

And it isn’t just about combat. Or traps. Imagine a hazard like green slime. Green slime hides on the ceiling and plops on you and sticks to you and dissolves your skin. The first time, it’ll surprise the hell out of someone. After that, the party will start looking up. They will see it every time. How many different things can you do with green slime assuming the party is never going to be surprised by green slime again?

Try it right now. Think of how many different ways to use green slime if the party is already looking up.

On the ceiling above the door the party wants to go through. On the ceiling inside the door so the party can’t see it from outside the room. Several patches above a narrow walkway over a drop. On the ceiling above a treasure chest. On the ceiling above the pedestal that has the book that is easily dissolved by acid. Blocking a narrow hallway. See what I mean? If I told you had to use green slime four times in a dungeon, you’d HAVE TO get creative.

Essentially, the Mega Man approach is all about empowerment. It empowers you, the GM, to be creative by forcing you to be. And it empowers the players by training them to deal with obstacles and challenges and then forcing them to adapt that training to new situations.

The Asset Library Approach

You may or may not know that – for funsies – I’ve been learning how to build games in Unity3D. Unity3D is a game engine. It’s not quite building a video game from scratch. Instead, Unity3D comes with a bunch of stuff already handled. Things like getting input from the keyboard, mouse, or game pad, how to display graphics, and the other deep, deep, basic programming that makes games run at the core level. Instead, you focus on building game components and designing a game. It’s powerful and versatile, but you don’t have to understand how a computer communicates with a controller or how a video card works. It’s a lot of fun.

I bring it up because there’s an organizing principle that helps exemplify exactly what I’m talking about: the concept of an Asset Library.

Your Unity3D game has a bunch of folders or boxes or piles of different things called assets. And as you start to build, you tend to think in terms of building up a game as a collection of assets rather than thinking of it in terms of drawing a map and then adding stuff to it. For example, early in the course I’m taking, I built a standard brick breaker game. You know the type. Slide a paddle back and forth, bounce a ball, use it to break bricks, break all the bricks to win.

In D&D, I might design the game encounter by encounter. Here’s the room with three bricks. Here’s the room with six bricks. Here’s the room with unbreakable bricks. In this room I’ll make bricks that shoot lasers at your paddle. In that one, I’ll… and so on.

Instead, you tend to think in terms of building all of the components. I made a ball. A paddle. A basic brick. Bricks that took different numbers of hits to break. Bricks that were unbreakable. And so on. And I put all of those things in a folder. And THEN, I started building different levels or encounters or whatever by mixing and matching my components.

And THAT is an approach I advise you to take with ANY adventure. Don’t design encounter by encounter. Instead, come up with a limited list of assets and then jumble them up. Now, that isn’t to say that unique things don’t have their place. Games have bosses and minibosses and unique puzzles from time to time. But those are the rare exception, not the rule.

And, again, this doesn’t just apply to traps and monsters. You can use it in ANY type of adventure. For example, imagine a stealth infiltration. The PCs are sneaking into an enemy stronghold. They might do it by stealth, by trickery, or by quietly assassinating foes one at a time. Whatever their approach, you don’t need that many challenges to build a good scenario. You have guards and you know how perceptive they are, how prone to trickery they are, how easy they are to kill in one shot, and so on. And maybe you have dogs that are extremely perceptive. One guard in the middle of the room alone and looking in the wrong direction? Sneak by. Dog in the middle of the room? Kill it. Dog and guard together? Well, the guard is more likely to raise an alarm and get everyone’s attention. The dog will bark a lot and then jump into the fight. So maybe the kill the guard first, then the dog and hope no one hears the dog.

How many different scenarios can you build with just guards and dogs? Probably lots. And each one is tactically interesting.

And that is what we’re going to focus a lot of attention on in the near future: constructing asset libraries for our megadungeon. Now, our dungeon is HUGE. And it’s going to need lots of different assets. But, in the end, when we start our detail mapping and encounter design, we’re just going to be combining components from different types of obstacles and traps and rosters of enemies. And, as time goes on, we’re going to discuss some of the other reasons for this powerful approach.

But, for today, so this article isn’t JUST about theory, we’re going to build a very simple library. A boring library. But a necessary one. We’re going to build some doors.

Of Doors and DCs

Let’s talk about doors. Doors and gates. Openable, closable thingies that keep people from passing through openings in walls. Our dungeon is going to need some doors.

Recall that two of our gate keys are literal keys. On Day 3, our heroes will discover the Skeleton Key which will allow them to open certain doors in the dungeon. On Day 5, they will discover the Arcane Runekey which allows them to open certain different doors in the dungeon. Now, we COULD just say we have these two types of doors – the locked doors and the arcane locked doors – and assume they can’t be bypassed except by those keys. But that would kind of suck.

Way, way, way long ago when we first started discussing the concept of gating off our dungeon, we agreed that doors and locks were a s$&% way of gating by themselves because players are actually used to finding ways to bypass locked doors. PCs are always picking locks and bashing down doors or hacking them apart or whatever. And there’s a really good reason to let the PCs do that. Smashing doors or picking locks is BREAKING THE RULES. Doors tell you where you can and can’t go. But D&D is about freedom. And part of making a game feel like it’s based on exploration is making the players feel as if they are free to explore anywhere. And that means allowing them to break the rules.

The solution we came up with was to make doors they could get through and doors they couldn’t get through. But that solution can look contrived. In fact, it strips away the whole illusion of rule-breaking. If we designate some doors as “you are allowed to break these” and others as “you can’t break these ones,” we’re just creating a different set of rules.

The way to get around that feeling of contrivance is to create a logical system. Something that feels natural, like it makes sense in the game space. For example, imagine you are exploring a dungeon and you encounter several different doors. You find some old, rotten wooden doors. And then you find some other doors made of stronger wood. And some other ones are iron-banded. And others still are reinforced iron. And some others are just solid iron doors. You’d figure that the designers put different doors in different places depending on how secure they wanted the areas to be. And you’d expect the doors to be easier or harder to break through. If you’re able to attempt to break down every type of door and the GM lets you roll die rolls and there is some sort of logic to the difficulty, you would actually be okay with discovering that some of the heftiest doors are just beyond you at first level. In fact, unless you roll the maximum you possibly can, you’re more likely to blame crappy die rolls than you are to assume the doors were designed to be unbreakable until a higher level.

Moreover, whenever you encounter a door you can’t open, that door becomes a goal. It becomes something you want to go back to later. So, when you finally DO discover a key – which you probably aren’t even expecting because keys are oddly rare in D&D – you’re likely to start exploring doors. Meanwhile, you’re waiting for the day when you can cast knock or improve your proficiency bonus or find a ring of the ram or something.

Now, the thing is, our logical system has to be logical. Yes. You read that right. A logical system must have an actual logic. And that logic needs to be visible to the players. If the doors all look identical but some doors are impossible to break and others are easy, we don’t have a logical system. We have an arbitrary system.

Doors in D&D

First of all, let’s look at opening doors in D&D. If we’re going to design a system for doors, we need to understand the rules as they exist. And the thing is, finding all the rules for doors is ludicrously complicated because they are scattered all over the place in the books and the indexing is so crappy that you really have to hunt to find it.

The basic rules for doors run like this (DMG 103). Doors can be stuck or locked. Stuck or locked doors can be forced open with a Strength check. A character who is proficient with Thieves’ Tools may also use those tools to pick the lock by rolling a Dexterity check and adding their proficiency bonus. The door may also be broken down by dealing enough damage to it. The rules for Strength and Dexterity checks are pretty easy (PHB 174). Roll 1d20, add the appropriate ability modifier and a proficiency bonus if it applies, and compare the result to a specific DC. If the result is equal to or greater than that DC, the action succeeds. Otherwise, the action either results in no progress or results in progress with a setback. Dealing damage to a door uses the rules for damaging objects (DMG 246-247). That comes down to rolling an attack against a specific AC and then deducting the damage from the object’s hit points. When the object runs out of hit points, the object is destroyed. Objects are automatically immune to poison and psychic damage and can be immune to other types of damage. And some large objects have a damage threshold. If an attack does not deal at least as much damage as the object’s threshold, the object takes no damage at all.

And that’s it. That’s what we’ve got to work with. And there are a lot of vagaries. First of all, stats like DCs, AC, and HP are all determined by the DM. Sure, the rules give some specific guidance, but we can go ahead and assume that we can set whatever numbers we want as long as we tell the DM what the numbers are. Second of all, though, there’s some questions the rules just don’t answer.

For example, if you fail to force a door or pick a lock, what happens? Can you try again? Can you just keep trying until you get it? The game is very unclear about that. It’s actually kind of vague on some pretty important questions (like retries) overall.

In those cases, we’re going to have to give the GM guidance. And, because the game is unclear, we can give whatever answers we want as long as they don’t contradict the rules already given. Remember, we don’t want to break any rule without a really good reason. So, let’s start by figuring out our rules for doors and then try to assign some specific numbers.

Megadungeon Doors

We know that there’s three approaches to doors. Doors that are stuck can be forced or broken. Doors that are locked can be forced or picked or broken. To force a door, roll a Strength check against a specific DC. To pick a lock, if and only if you have thieves’ tools and are proficient with thieves’ tools, roll a Dexterity check plus proficiency bonus. To break a door down, roll an attack against a specific AC and deal damage. Doors may be immune to or resistant to specific types of damage and may have damage thresholds.

Let’s think about each of these things in turn and see if we can come up with a logic for all of it.

First of all, we can assume that all of the doors in the dungeon are close enough to the same size that we don’t have to think too hard about size variations. In my mind, a double door and a single door are close enough in size. All doors are Large objects in the parlance of D&D.

First of all, we have stuck doors. A door that is stuck is one that is swollen in its frame or its hinges or rusty or the latch is jammed or something. Forcing the door is simply a matter of hitting it hard enough in the right place to unstick it to overcome whatever age has done to it. Besides that, a stuck door is just an open door with a Strength check in the way of it. Nothing more than that. And frankly, that isn’t too interesting by itself. If you just have to roll a die to open the door or not open the door, there’s nothing interesting happening in the game. And that’s why we have to think about what happens when you fail.

First of all, can you keep trying to unstick a stuck door? The answer is yes. Obviously. Just keep slamming away at it until you force it open. But that’s even less interesting. After all, if you can just keep rolling and rolling until you succeed, the die roll is just busy work.

What happens when you fail to open a stuck door? Well, the attempt itself is noisy. Anything on the other side of the door is going to know that something is trying to get in. And it’s going to prepare. So, whenever there are creatures near a stuck door, we need to tell the GM how the creatures prepare for an attack if the first attempt to force open a stuck door fails. We can write specifics into each room with a stuck door about how the inhabitants prepare for attack. Maybe they flee to another room, maybe they hide, or maybe they prepare actions to unleash when the door opens. We can work those out on a case-by-case basis.

We could stop there. We could make our stuck door rule look like this:

To open a stuck door, roll a Strength check. If it succeeds, the door bursts open in one shot. If it fails, it takes several tries to force the door. Creatures on the other side of the door (if any) are ‘on alert’ and will take different actions which will be described in the encounter description.

And that means that, no matter what, there’s only one die roll. Either the door opens or the door opens and stuff on the other side gets the drop on the players. That’s actually a pretty good general rule. But, we can do even better. Because if there’s nothing on the other side of the door, there’s no point in rolling. Stuck doors become boring again.

We know our dungeon is going to have wandering monsters in it. Noisy attempts to force open a door are going to attract attention. And the more attempts it takes, the more likely something is to wander up before the party has the door open. So, each failed attempt to force a door can force a wandering monster roll. That creates a sense of rising tension. If a door is being stubborn, the party faces a “press your luck” situation.

So, our rule is:

To open a stuck door, roll a Strength check. If it succeeds, the door bursts open in one shot. If it fails, the door remains stuck. Creatures on the other side of the door are now ‘on alert’ as described in the encounter description. Use the wandering monster rules to determine if wandering monsters arrive while the PCs are trying to force the door. PCs may attempt to force a stuck door as often as they’d like.

We’ll come back to setting the DCs in a moment. But for now, just understand how this works. If the PCs succeed on their first attempt to open a stuck door, there are no consequences. They burst into the room. If it takes more than one try, they’ve given the foes on the other side enough time to ready themselves for attack and they might attract an encounter on their own side of the door as well. That’s a nice, simple way to handle things.

Locked doors can be forced and there’s no reason to use a different rule than the one above. The only thing to keep in mind is that locked doors are much harder to force open because the thing keeping them closed isn’t just incidental damage from age or whatever, it’s the actual lock mechanism. And that mechanism is designed to resist being forced open. And once a locked door has been forced open, the lock is broken and the door can never be locked again.

But when it comes to picking locks, things are a little bit different. We have the same basic question: can the party keep trying over and over until they succeed? And the answer comes down to whether we can make the attempt interesting or not. EXCEPT, with locked doors, we absolutely KNOW there are doors that CANNOT be picked or forced. We’re going to design some doors to be so sturdy as to be impossible to get through without the key. It’s bad enough PCs can keep smashing and smashing and smashing a door forever with no hope of ever getting through. But picking locks lacks that noisy consequence. By its nature, it’s a quiet activity. In fact, one could argue that the reward for picking a lock is maintaining the element of surprise however long it takes.

Unlike forcing a door which – if it doesn’t work the first time – it might not attract any attention at all. We could argue it is time consuming enough to warrant wandering monster checks, but I don’t want to do that. When we discuss wandering monsters, we’re going to discuss some very good reasons to leave timekeeping out of it.

So, we can make lockpicking a one-shot thing. That is, either the lock is pickable and the PCs pick it, or they discover it’s too complex for them to ever pick. Done and done. Admittedly, that’s kind of a pain in the a$&. Because it requires the GM to keep track of what locks the party has and hasn’t failed to pick. And that’s why we’re not going with the more traditional rule about being able to retry attempts to pick locks after you gain a level or something like that. Because that creates too much recordkeeping.

There are two reasons we can get away with the “one attempt ever” thing. And they have to do with the fact that we’ve been calling our Day 3 key a “skeleton key.” We can assume that key will open all mundane door locks in the dungeon. Once the party finds that key, picking door locks will never be an issue again. That means they won’t be locked out of anyplace forever by failing to pick the lock. And also that the GM only really has to track locked doors for three days. After that, door locks just aren’t an obstacle anymore.

We can have locked chests or special doors that the skeleton key can’t open if we want. And we can call out those exceptions as they happen. And we want those things so rogues have something to do. And so do wizards with knock spells. But the general rule is locked doors can be opened with the skeleton key.

What about the last option? Fighting doors? Well, fighting a door is as noisy is as noisy as forcing a door. So, the same rule applies to attacking a door as trying to force it open. If you attack a door, anything on the other side becomes alerted and we check for wandering monsters.

Setting Stats

So now it comes down to this: how do we set the stats on the doors? And to do that, we need to understand some of the basic math of the system.

First of all, we can always assume that the best PC for the job will be attempting a task. The strongest PC will try to force a door. The rogue will try to pick the lock. Forcing the door is a Strength check. So, at 1st level, the party will likely have a +3 on any roll to force a door open. Picking the lock is a Dexterity check with proficiency bonus. The party will likely have +5 to +7 on any attempt to pick a lock. Why up to +7? Because thanks to the rogue ability “Expertise” a rogue might be able to double their proficiency bonus with thieves’ tools.

Fortunately, the system is designed such that we generally don’t have to worry about other bonuses. If several party members team up or the barbarian flies into a rage or whatever, they have advantage on the roll. Advantage makes success more likely, but it doesn’t change the highest possible roll. If you have +3 Strength and roll a Strength check, the highest you can ever roll is 23, with or without Advantage. That means, if we want a door that CAN’T be forced open, we need a DC greater than 23.

Picking locks is a bit trickier. The highest roll is conceivably about 27. So, we might want a door that’s locked with a DC of 30 to guarantee a rogue isn’t getting through there at low levels.

What factors go into the DC? Well, forcing a stuck door is pretty much just a matter of overcoming the age of the door. The door’s materials, workmanship, whatever, those things don’t make it easier or harder to force a stuck door. The door is stuck because it has failed in some way. So, all stuck doors might carry a DC of 15. That means that there’s about a 50% chance any first level party will force a stuck door on the first try.

Forcing a locked door is always going to be hard because locked doors are designed not to be forced open. And it’s going to vary based on the quality of the lock mechanism. You might think it has to do with the material of the door. But when you’re trying to force open a locked door, you’re trying to break the lock mechanism. So, the DC for forcing a locked door is going to vary based on the lock quality. The DC to pick the lock is also going to vary based on the lock quality. So, we can actually come up with different qualities of lock. And we can make a very simple table. Look.

	DC to Force/Pick
Stuck*	15
Locked, Simple	20
Locked, Moderate	25
Locked, Complex	30
* Stuck doors must be Forced

And that table is based entirely on the quality of the locking mechanism.

As for breaking doors, we can vary it based on the materials. We can imagine that the elves might use wood for their doors, but elves are long-lived and their structures have to endure. So their simplest doors would be wooden doors that have been strongly lacquered to resist age and the elements. Stronger doors might be banded with mithril silver or might be completely covered in a layer of mithril. And they might also make mithril silver gates.

The DMG (p. 246) gives us some guidelines for ACs and HP based on size and materials. We can assume, first of all, that all doors are resilient and large, so they all possess 27 (5d10) hit points. Wooden objects have an AC of 15, but we can assume that the elves resistant lacquer improves that to 17. Mithril bands on the door improve it to 19. Mitrhil covered doors and gates have an AC of 21. So, that makes doors pretty mighty to begin with. We can also decide that mithril plating gives the doors a resilience such that any attack that doesn’t deal, say 5 damage, merely does cosmetic damage. Mitrhil gates are more resilient. They have a damage threshold of 10. Finally, we can think about damage types. All objects are immune to poison and psychic damage. That makes sense. And this is where we face a particular problem. We can think through each damage type in turn and argue whether weather-resistant lacquer or mithril silver should be resistant or immune to this, that, or the other. And then we can include a complex list of resistances, immunities, and vulnerabilities for both sets of materials. Or we can do things a little more simply by inventing a general rule for doors that specifies only a few, proper ways to break down a door.

For example, we might argue that breaking down a door requires bludgeoning, force, slashing, or thunder damage. Basically, you can pound a door apart with blunt force (and force and thunder are blunt force) or hack through it.

In the end, then, we get this table.

	AC	HP*	Threshold
Lacquered Wood	17	27 (5d10)
Mithril-Banded Wood	19	27 (5d10)
Mithril-Plated Wood	21	27 (5d10)	5
Mithril Gate	21	27 (5d10)	10
* Doors are immune to all damage except Bludgeoning, Force, Slashing, or Thunder

So, now we have our basic rules. By specifying the type of door and the complexity of the lock, we can set stats for any door in the dungeon. Mithril-Plated Wood and Mithril Gates with Complex Locks are impossible to pick or force and very difficult to batter down. The damage threshold on the higher-tier materials makes it likely for the party to have several attacks that fail to deal damage, but which might still attract wandering monsters. Suddenly, it becomes very tiresome to try to destroy doors.

What About Magic

And now that leaves us with just the question of magically protected doors and magical ways to circumvent them. We can assume that magically locked doors have a variant of Arcane Lock cast on them. This takes the form of a golden sigil inscribed in the door (because the material component of Arcane Lock is gold powder). Arcane lock increases the DC to pick or force the door by 10. So, we don’t need separate stats for magically locked doors. We just specify that the Arcane Lock adds 10 to the DCs to force or pick them.

The Arcane Runekey then is nothing more than a magical object that suppresses the Arcane Lock on any door, thus allowing it to be unlocked normally. Then, the Skeleton Key can be used to open the door.

This actually creates an interesting parallel with the Knock spell. The Knock spell suppresses the Arcane Lock spell for 10 minutes, allowing the door to be locked, unlocked, or opened normally. So, the Arcane Locks and the Arcane Runekey make sense in the game world. This does create a slight risk because the PCs could, theoretically, possess the Knock spell and the Skeleton Key BEFORE they get the Arcane Runekey and, thus, they could break the sequence. Hold on to that thought for a moment. We’ll come back to it.

It is interesting to note that the Knock spell makes a loud, audible noise akin to forcing or breaking a door. So, we should also make a note that the Knock spell forces a wandering monster roll and alerts monsters on the other side of the door. Obviously, that is what the designers intended when they made that very specific note.

Breaking the Sequence

In the end, we’re left two ways that the party can break the sequence we’ve built into our door logic. First, they can theoretically deal enough damage to destroy any door. It’s time consuming and they might end up fighting a lot of vermin while doing it, but it can be done. Second, they can end up with the Knock spell and the Skeleton Key before they have the Arcane Runekey. In theory, they could use the Knock spell to suppress the Arcane Lock on a door, then use the Skeleton Key to open it.

How can we prevent these things?

Well, honestly, we can’t and we shouldn’t. First of all, there comes a point when trying to protect the adventure from the players starts to seem like a screwjob. So far, our locks and keys all have an internal logic that makes sense in the game world. As the players interact with different doors and different locks, they can make logical connections about materials and lock complexity and failed or successful rolls. But the more arbitrary rules we attach to this stuff, the more it feels like we’re just screwing them. It’s a fine line between good design and contrivance.

Second of all, it’s not as if the party can get TOO deep into the dungeon by breaking the door sequence. At first and second level, destroying a door by dealing damage is going to be time consuming, especially with the damage thresholds and the very limited list of damage types that work. And random encounters are going to chew up resources the party doesn’t have. In short, they should quickly get discouraged. Third level is the first level where they can do real sequence breaking, if they are willing to blow a second-level spell slot to bypass a door. And because of the tiered structure of the dungeon difficulty, they aren’t going to be confronting anything TOO dangerous for them. At most, they will be skipping two days of adventure between third and fourth level temporarily.

Simply put, if they are really dedicated to breaking down these doors, they are probably going to do it, but they won’t be able to do too much damage when they do.