Welcome to the Megadungeon…

Season 2!

The last time we did one of these dungeon things – a long, LONG time ago – we reached a major milestone. We had a big spreadsheet that laid out the entire plot, encounter, and experience progression for the whole honking dungeon.

Now, there’s a lot of details left to fill in and work out. There’s still a lot of open points. We have only a vague notion of who the antagonists are. And what the final goal is. We have a skeletal backstory with a lot of missing pieces. We haven’t figured out how all of our keys and gates work from a mechanical perspective. And so on. We also haven’t discussed treasure.

But – and it’s going to blow your mind when I say what I’m about to say – but, there is such a thing as OVERPLANNING. Especially when it comes to table-top RPGs. And that is why we’re going to step AWAY from video games for a little while. I know. I can’t believe I said that either. But here me out.

The thing with video games is that they are big, complicated beasts built by several or several dozen or even several hundred people. And Alice in scenario design can’t just start banging together the Frostfire Rift before Bob and Carol have worked out the basic jump physics for the rocket boots. Alice and Bob and Carol have to agree on how the rocket boots will work before Alice can start designing levels and Bob and Carol can start coding physics and Dave can start doing animations. And if Bob and Carol decide to change the physics, say because Elaine discovers a problem in some other scenario, Alice is going to have to go back and fix all of the Frostfire Rift to account for the new physics. And Dave might have to add a bunch of animation frames to make up for the longer jump cycle. And Fred – who just needs the damned thing to ship on time for once – is going to slap all of them.

Now, that s$&% does happen. Even with all of the planning, there’s a lot of feedback between different teams and things gradually get tweaked. And that can happen because you have bunches of people working in parallel.

BUT, I am one human being. I don’t have the advantage of teams working simultaneously on different parts, but I also don’t have the problem of dependencies – work that one team is doing relying on what another team does. And I can play things fast and loose. If I decide to tweak the physics on my Boots of Jumping, yeah, I might have to change the DCs on a bunch of jumps, but I really don’t need to decide on the physics UNTIL I’ve reached my first jumpable chasm. I can work in sequence.

But I can also bounce around. So, if I don’t feel like dealing with jump physics as reflected in DCs and Athletics checks today, I can just leave a bunch of blank spaces on my maps that say things like “need a hard jump here, figure it out later.” I won’t f$&% up the scenario design by suddenly having to move everything two grid units to the right because the jumps need to be wider.

On top of that, there’s also an issue of wobbliness. The whole design is kind of shaky. What do I mean? Well, first of all, I mean that – as much as we’re planning things – there’s a few combinations of abilities and mechanics that might break things. Or the players might – because of the freedom inherent in role-playing games – figure out how to f$&% up the sequence. The druid might decide to turn into a fish and swim all the way into the Crystalline Caverns. Maybe. In a video game, that s$&% breaks games. If you go into the wrong area before a certain flag is set, dialogue starts popping up out of order, maps don’t load properly, and the next thing you know, someone is on YouTube posting an any percent speedrun of your thirty-hour epic with a completion time of seven minutes. Or worse, someone is complaining that their game crashed and their save is unrecoverable because they saved in a place they should never have been and now can’t get out.

All of that s$&% can happen in role-playing games too. Except the speedruns. Could you imagine watching a speedrun of, say, Tomb of Horrors? But, it’s not actually a huge problem. First of all, the game will not crash because people ended up in the wrong place. They just end up in the wrong place. And probably get killed by an encounter too powerful for them to handle. And then they make new characters and start again. If the druid does pull that sequence-breaking s$&%, he’s probably strained through the digestive tracts of a swarm of vampiric dire quippers and the rest of the party never knows what happened to him. That’s part of the game. More importantly, the GM can make adjustments for the party doing really ridiculous things. If the party starts burrowing into the Fiery Abyss at third level, the GM can decide to let them die in there like Minecraft Steve digging straight down or she can decide to warn them not to or just disallow the action or slap the players and find a group of players that isn’t trying to break everything. Part of being a GM is learning to deal with that s$&%.

And, to some extent, the wobbliness, the shakiness, is a feature. It’s not a bug. We WANT the players to have between 11 and 13 levels of XP to face the final boss. But, if they rush the adventure and go in early, they might still be able to pull it off if they are smart. Or they might be lost forever. If they manage to somehow gain an extra level, the fight will be a little easier, but it’ll still feel like a cool victory that they earned.

In the end, our spreadsheet is, at best, a tight guideline. We don’t have to adhere to it exactly. And we’re accepting things will be off. Hell, even the encounter math in the game isn’t exactly precise. Trying giving 5 PCs EXACTLY the right amount of XP for a 2nd level encounter. I think you’ll find that it doesn’t line up as nicely as you hope.

Several people have pointed out a slight inconsistency in the encounter math based on whether you consider the numbers in the XP Per Encounter table to be floors, ceilings, or averages. I even had a discussion on Twitter with one of the WotC staff about it on Twitter a few months ago. The DMG actually doesn’t specify. The official answer is that those numbers are floors, thresholds. When it says: “Easy 100 Medium 200 Hard 350 Deadly 500” what they are saying is that anything above 100 is Easy, above 200 is Medium, above 350 is Hard, and above 500 is Deadly. And I was using them as middle-of-the-road averages. Targets.

That doesn’t matter for two reasons. First, because even though I started with those numbers, I actually calculated how much XP and therefore how many encounters should exist based on the much less ambiguous “XP per Adventuring Day” table and the overall progression is based on the “XP to Level” progression. In the end, the encounter building stuff became sort of a loose guideline. Second, it doesn’t matter because you can’t hit precisely those numbers anyway. Seriously. Even if you have four PCs and use only single monsters, you’ll find it’s hard to get exactly the right XP to match the thresholds. You’re always going over or under. So, again, treating those like targets in a game of Horseshoes & Handgrenades won’t get you far off the mark.

Anyway, that’s enough discussion of the old. It’s time to start the next phase of the journey to megadungeon!

Mapping the Whole Megadungeon

Our goal now is to map the entire Megadungeon. That’s what Season 2 is all about. We want a map. And we want it to be a good map. Let me tell you a story.

Once upon a time, I was a HUGE fan of Dungeon Magazine. I didn’t use many of the adventures from the magazine, but I read a lot of them and loved them. But what I really loved was the maps. I LOVE maps. And if you didn’t know that, you need to learn to pay better attention. In March of 1995, in Issue #52, an adventure appeared by Christopher Perkins called My Lady’s Mirror. It was an awesome adventure. See, there was this distant, creepy castle ruled by a woman who had been capturing souls for years in her magic mirror. She used them to power her immortality. Then one day, while she was on vacation, her butler knocked over the mirror and released the souls. So, a bunch of NPCs – her victims – were now running around different corners of the giant castle. And the castle had a really great, isometric map. Seriously. Look at this. I’m risking getting sued by WotC or Paizo or whoever owns the right to this s$&% just to show off this map, which is merely ONE PART of the ENTIRE PUBLICATION which I am providing CRITIQUE ON for ILLUSTRATIVE PURPOSES.

Look at it.

I LOVED this map. Until I noticed the mistake. I was devastated. The mistake ruined everything. If you want to play along, see if you can find the mistake. Study the maps closely. If you can’t find it, I’ll give you a hint. It has to do with how the stairs align between two different floors. Still need a hint? Check out the stairs between Area 30 and the hallway outside of Area 50.

The problem with passionate love is how quickly it can turn to rage when the thing we love fails us. Because love and anger are two sides of the same coin: passion. This map betrayed me. And so, I made a solemn vow. I would never bring a map into the world like that. If I drew a map without a grid, it could be haphazard and “close enough.” Fine. But if I stuck a map on a grid, it was going to line up. My world would not have collision issues. It would not have impossible rooms. It would not be like the one room in Portal 2 that is actually physically impossible given the space it occupies. It would not be like those damned stairs.

Stairs on B1 need to be in the same spot in B2. If there’s a ravine on B1 that’s 30 feet deep, it better be on B2 and B3 and B4 assuming each floor is about 10 feet high. At least, it has to bottom out on B4. I’m a psychotic stickler for this crap.

So, even though it would be impossible, if my entire Megadungeon was stuck onto one three-dimensional sheet of hyperpaper, it would work. You could do it. If you wanted to put the whole thing into 3D modeling software, it would all work. You could build the thing.

Now, that’s pretty daunting. Because the map is already going to be complicated. Look at it like this. We’re planning on 179 encounters. 141 planned. 32 optional. And probably 6 on the last day. In addition to that, the dungeon will need empty space. Probably around 60 empty rooms. Those will serve for pacing reasons, to house discoveries, provide spaces for short rests, and provide room for random encounters. That’s not including hallways and transition spaces. All of that will be spread out over eight distinct geographical zones spread out over five floors. And it will all be contained within the shape of an arc because the dungeon is built through the wall of a volcanic crater. And we’re going to make the layout so exacting that someone could put the whole thing into a 3D modeler and render it without having to fix any issues. Because of a magazine that traumatized my irrational teenager brain over 21 years ago.

Wow, this article is making me feel old. And psychologically damaged.

How are we going to that?

Enhance! ENHANCE!

If there is one thing I’ve learned from TV, it’s that if you want to see more detail, you just have to yell enhance at photoshop until you can make out a license plate number reflected in the moustache wax of a key witness. And that’s exactly what we’re going to do. We’re going to yell enhance at our spreadsheet until it looks like a detailed dungeon.

When you get down to it, the spreadsheet is pretty much the broadest, most highest level view of our dungeon possible. It’s not even a satellite image. It’s a blob of vague data. We want to get down to the point where we can count the floor tiles in any room. And we’re going to do that in a couple of passes. Each one adding more fine detail and snapping the dungeon a little closer to the grid.

The first pass is still going to be pretty blurry and vague. And that’s what we’re going to do today when I get done talking about. The second pass will snap the dungeon to a large scale grid. The third pass will actually fix the individual rooms and hallways and architectural and geographical features.

By the way, that second pass is going to be what we spend a LOT of time on. If you’re wondering WHY Megadungeon was quiet for a few weeks, it has to do with figuring out how to present these design steps without literally taking a year of designing each individual adventuring day TWICE before we start designing each adventuring day for reals.

To be honest, I didn’t think this design blog thing through.

So, our goal today is to get a flowchart that vaguely shows us where stuff kind-of should be. Where is Day 3 taking place roughly compared to Day 1. And Day 7? Will Day 14 get in the way of Day 20? How does the party walk from Day 6 to Day 16?

By the way, it’s important to note that we’re not going to try to move from “days” to “rooms” yet. We’re still picturing a day as a sort of space in the dungeon that has around 6 to 10 rooms and connects up to other days via controlled transition points. Our second pass will expand each of the days into collections of physical spaces.

Now that you understand what we’re trying to do, let me show you how I did it.

The Template

First of all, I needed a template. A basic way of keeping track of roughly where stuff was in the dungeon in relation to other things. And, because of the Trauma of 1995, I was going to jump right in and make sure everything sort of fit together vertically and horizontally.

First of all, my dungeon has five floors and seven zones. On Level 1, we have the Source of the Flow. On Level 2, we have the Desiccated Sanctuary, the Sacred Halls, and the Great Tree. On Level 3, we have the Crypt of the Ageless and the Flooded Underhalls. On Level 4, we have the Crystalline Caverns. And on Level 5, we have the Fiery Abyss.

Right off the bat, I decided to cheat a little. See, there is only ONE way into the Fiery Abyss. It’s the sealed tunnel that the party has to open to get to the demon queen. If we assume the passage is a long, sloping tunnel that stretches away from the Crystalline Caverns, then the Fiery Abyss isn’t directly below the Crystalline Caverns. So, we can put Level 4 and Level 5 side-by-side.

So, I took a piece of graph paper and divided it into four columns of equal width. Each represented a vertical level of the dungeon. Think of those columns as four sheets of graph paper laid side-by-side. And each of those is a slice through a layer of the dungeon. If you cut those four strips up and laid them on top of each other, they would show you the four floors of the dungeon. And that way, you can use the graph paper to keep track of how things align. Basically, the squares in the lower left hand corners of each column of the paper are directly above and below the others. If you go two squares up and two squares right from the lower left corner of the Level 1 column, and you dig straight down, you’ll get to the square two squares up and two squares right from the lower left corner of the Level 2 column.

Follow that?

Next, I divided each of the columns into zones to conform with the zones of the dungeon. I did it based on a vague notion of what zones are above and below what zones and how big, relatively the zones were. As you’ll see later, I kind of screwed this up pretty bad.

And that’s it.

Plotting the Critical Path by Day

Remember that the critical path is basically the most efficient path through the dungeon that carries it through all of the required encounters. With my spreadsheet in hand and a vague sense of how I wanted the story to proceed, I started drawing a flowchart. I did it in pencil with an eraser in hand. This is the end result. But I’ll tell you how I made the decisions I did.

Day 1 begins in the Desiccated Sanctuary. It’s the entrance to the dungeon and therefore. It’s also the southernmost point of the dungeon because that’s where the mountain is. So, given that, why did I start it in the middle of the space for the Desiccated Sanctuary. Well, because it’s an old habit. When you start drawing a dungeon map, you start in the middle of the paper to give yourself the most number of degrees of freedom. It was a dumb move. But it didn’t matter.

According to our story, in Day 2, the PCs become trapped in the crypt below the sanctuary and have to escape. In escaping, they open a path forward to Day 3 and also back to Day 1. How? Well, right now I’m envisioning something like a cave in that takes the floor out from under them and buries them alive in the crypt. They find another exit from the crypt that allows them to open a barred and locked door from the other side to escape back to Day 1. You can see how I represented this on the map. Based on the grid coordinates, Day 2 is directly below Day 1. The dotted line arrow represents a transition between levels. The dotted line arrow coming BACK from Day 2 connects to a line that goes between Day 1 and Day 3.

Day 3 gives the PCs their first gate key. So I decided that Day 4 doesn’t connect directly to Day 3. I wanted the PCs to learn quickly about walking away from paths they can’t open, discovering keys, and returning to them later to open them. On Day 1, the party will encounter the door that leads to Day 4. On Day 3, they find the key. Then, they can return to open it.

Day 5 takes place in the Great Tree. And that requires a long transition and it was my first sign that I had given myself TOO MUCH north-south space. I will probably shorten that up. It also illustrates that the distinction between Bubble and Arrow is going to vanish. What do I mean? Well, the GOAL for Day 5 is in the Great Tree area. But the encounters and rooms that make it up will end up smeared between the Sanctuary, the Sacred Halls, and the Great Tree. The arrows that represent the transitions will start to get gobbled up a bit and become part of the days of adventure. But that will come later.

Day 5 gives another gate key. So, again, we’d prefer to make the party backtrack and explore new regions in the zones they’ve visited before we send them back to the tree for the boss fight of Day 7. Incidentally, you can see how Day 3 and Day 6 together make up the heart of the dungeon. They become focal points. And that makes sense. Those spaces are probably big spaces connected to a lot of others because they are represent the major temples and spiritual structures of this place.

Day 7 removes the poisonous death vine roots from the dungeon and allows the party to explore new places. So we send them back to the Desiccated Sanctuary. We want to keep sending them back here because they are getting close to changing that area in a major way when they open the Floodgates.

And then we enter the Kobold arc. The party defeats the bulk of the kobolds, finds a way to breathe water, penetrates the lair of the dragon, destroys the dragon, and finds their way up to the floodgate that keeps the Underhalls flooded and the Sanctuary dried. These events are closely related so we keep Day 9, Day 10, Day 11, and Day 12 close together.

Day 13 is an interesting one. It’s a side day. One of those exploration days without a fixed goal. And it’s there to showcase how, with the canals filled and the floodgates opened, new areas of the Desiccated Sanctuary are accessible. But right now, it seems sort of vestigial, doesn’t it? Stuck off of Day 8 like that. Well, wait a bit and it’s real purpose will become clear when we start filling in more details.

Day 14 and Day 15 represent the party’s exploration of the drained Flooded Underhalls. And ultimately, they find the teleportation key. And that allows them to access Day 16 and Day 17. Day 16 is stuck off in the Crypt. It’s a place they can access with the teleporter. But they haven’t been back to the Crypt since Day 2. It’s HIGHLY unlikely that they will remember to go back there. So it might seem like I’ve lost my mind. But again, we’re going to fix that when we do something else.

With the teleportation key, the party can go down to the Crystalline Caverns and there they will find the Frostburned Runekey and also be stopped by the seal on the passage down to the Abyss. About two thirds of the way through the game is a good place to show them the final bosses lair. Or at least, show them the door.

Using the frost key, they can open a couple of sealed vaults back in the Sacred Halls. One of which grants them the ability to fly.

The reason we used Day 16 to remind them the Crypt exists (and we’ll talk about how in a minute) is because, with the ability to fly, the party can reach the corrupted spirit’s lair and destroy it. That’s Day 20. And that sets in motion the endgame.

Day 21, Day 22, Day 23, and Day 24 involve appeasing the four elven spirits to gain the power to break the seal on Day 25. They have to do this from the main temple. So we stick that right off the central Day 3/Day 6 nexus.

Of special note, notice Day 24 is not accessible from the Crypt even though it’s IN the Crypt. It’s actually reachable through the Crystalline Caverns. The reason is because the Crystalline Caverns need a little bit more adventure in them (and we’ll fix this a little more later). But we can use this to create an interesting puzzle. The party will discover that one of the spirits they need to appease is in the crypt. But when they find her crypt, the doorway is buried or collapsed or something. They can’t reach it. But environmental clues will show that the crystals from the cavern are spreading into this part of the crypt. That will lead them to find a path to come in from below. And note that puzzle has been designed by the fact that our flowchart highlighted how little we’re doing with the Crystalline Caverns and by showing how easily we can connect Caverns with the Crypt.

And then they can march into Day 26 and kick some demon ass.

Coloring the Lines

I probably don’t have to explain this next step.

Basically, I took out colored markers and highlighted the lines in bubbles in colors to show the gate/key dependencies. When we get into actually designing the dungeon in detail, this will get a little more complicated. But let me add a little detail about gating right now.

When a day is colored purple, that means the frostkey is the most powerful object needed to do that day. It doesn’t mean that ONLY the frostkey is used in there. The party might still need to teleport in spots or use the arcane key. There might be areas that were swallowed by poisoned roots that are now accessible. We never want the PCs to go too long without using any of their tools or without being reminded of the transformations they’ve visited on the dungeon.

Just keep that in mind. Replacing one key with another or having keys that are used precisely once is bad design. It isn’t fun.

Pinning Optional Encounters

Now, you might notice, at this point, some of the areas are a little sparse. But the critical path is not the entire adventure, is it. We planned for a whole bunch of optional encounters. And while we could just drop them into our days haphazardly, we can do better with them.

For example, we want to drive exploration and we want the party to constantly be exploring old areas with new tools. And there’s a few areas we want to add a bit more adventure to. So, we can take the optional encounters from the spreadsheet and scatter them around. Basically, attach them to specific days. And we can even put them behind gates.

For example, on Day, we have an optional encounter slated to appear. And we know that, by this point, the party has access to Days 1, 2, and 3 and it has access to the Skeleton Key. So, we can stick a locked door in Day 2 that has the optional encounter behind it.

That means the party has to work a little harder to find the optional encounters. Some of them aren’t “just off the path.” They require time and energy to hunt down. That makes sense because remember that the optional encounters and discoveries are what will spell the difference between a party being barely powerful enough to win and a party kicking the demon’s ass with two extra levels under their belt.

So, we take each of these optional encounters and we stick them into days like little pushpins. Just to tell us whereabouts they are in the dungeon. Take a look at the flowchart to see what I mean.

Notice the little red and gray pushpins sticking out of Day 2 in the Crypt of the Ageless. Those are optional encounters. One can be accessed right away. The other one is gated behind locked door.

I’m not going to talk about every pushpin I stuck into the flowchart. I really did stick some of them haphazardly. But let’s discuss a few of the more interesting ones.

First of all, there’s two keyed to the Floodgates in the Desiccated Sanctuary. I want the Floodgate to be a major transformation for that area. That will emphasize it.

I keyed three optional encounters to Water Breathing because Water Breathing got kind of screwed. You only ever need it once on the critical path. So, we’ll let it give access to some nifty treasures.

On both Day 17 and Day 18, I stuck three optional encounters each attached to defeating the corrupted spirit. And that’s for several reasons. First of all, I want the party to spend more time in the Flooded Underhalls and the Crystalline Caverns. Second, remember that the corrupting spirit died in this area when it was flooded. So it stands to reason that this a focus for the necrotic energy that her destruction banishes. Third, we want the party to discover the last details of the story of the fall of this place during this period. And so we’re going to need to hide some discoveries that can only be made here. These are the details that will tie the whole story together and fill in any missing pieces. In the end, instead of spreading those optional encounters around, I tacked them on to Day 17 and Day 18 as optional half-days of adventure. Basically, small segments of the dungeon that have several encounters and important treasures and discoveries inside of them. That can also help gear up a party for the final encounter with optional, useful treasures. Consider it the final, ultimate reward for scouring everything.

Traffic Control and Shortcuts

We’re not quite done yet. Right now, we have a critical path. But notice that it’s a pretty long walk from some parts of the dungeon to the others. We want to build a few shortcuts that open up between different sections as the party explores. Not too many. Just a couple.

More importantly, we have two weird things going on with our critical path progression. The first is the fact that Day 13 is not a direction the party will logically go after Day 12. They are more likely to head to the Flooded Underhalls and Day 14 and Day 15. But suppose a path opened from Day 12 to Day 13 when the party opened the Floodgates. And that path also connected Day 13 to Day 8 and Day 9. That way, the Desiccated Sanctuary becomes a sort of intersection and shortcut down to the Flooded Underhalls by way of Day 9. That will also immediately send the party to the Desiccated Sanctuary to see the results of their work.

By the same token, what keeps the party from jumping right from Day 14 and Day 15 to Day 17 as soon as they can teleport. Why would they go to Day 2 and open Day 16 first? Well, what if they discover a connection from Day 14 to Day 2 before they find the teleporter. One that conveniently takes them right near the teleporter gate to Day 16.

Again, drawing this out not only highlights places where our critical path might not work, but it also suggests solutions. And sometimes, all it takes to control the party is to show them an easy and obvious path. By giving PCs easy ways to get to the things we want them to see, we can make them give themselves the guided tour we want to give them.

Finally, once the party can fly, we can give them a major bonus. From the beginning, whenever the party wants to get into the dungeon, they have to use the door. The ONLY door. Once they can fly, we can allow them some other entrances. The most obvious is through the Great Tree because that area is outdoors. The party can just fly into the crater and land wherever they want in the Great Tree region. We’ll represent those by brown entrances keyed to flying.

BUT, we also want to let the party to get deeper. Specifically, when they are ready for the last day, we don’t want to bog them with random encounters and long wandering and other bulls$&%. In our first sketches of the dungeon, we had a waterfall that fell all the way down from the Great Tree into the Fiery Abyss. What if the party can fly down into this ravine and skip to the Crystalline Caverns NEAR the entrance to the last day. Just for funsies, we will also allow the party to enter the Sacred Halls via open shrines to the sky and to access the Source of the Flow by way of the waterfall. Again, they have to be able to fly.

Well, actually, “just for funsies” is a little bit of a lie. We’re actually letting the party fly pretty close to where three of the four spirits need to be appeased for Day 21, Day 22, Day 23, and Day 24. If a smart party thinks of it, they can take a lot of pain out of the scavenger hunt that precedes the finale by using flying shortcuts. Because, remember, as we get close to the end, we want to build momentum. But we also want to reward exploration. So, in the last five days, the spirit of exploration gives its final rewards in the form of secret areas filled with lore and treasure and shortcuts to make the final days much easier.

Final Analysis

You can look over the final flowchart and get a good sense of how the adventure flows. But there’s clearly a few issues that we will need to fix.

First of all, the dungeon is stretched out along the north-south axis. The Great Tree is too far north. And the Source of the Flow is also a bit too far north. Both the Desiccated Sanctuary and the Crypt of the Ageless could also move a little more north. Basically, the whole thing could be tightened up. We’ll fix that when we start our second pass map.

Second of all, we’ve got an issue with arrows. Specifically, the transitions between Day 12, Day 13, and Day 9 are facing the wrong direction. We want to move directly from Day 12 to Day 13 and then back to Day 8.

Third of all, we’ve got to remember when we’re designing the Crypt and the Flooded Underhalls that we’ve got to incorporate a few details to drive the traffic back to Day 2 before driving it forward to Day 17. And we’ve also got to allow the party to see the inaccessible Day 24 and add environment details that will send them to the Crystalline Caverns.

But all of those problems are problems for another day.

Edit: I’m NOT Crazy!

Well, I am. But for different reasons. After several private messages on social media and at least one comment below, I had to take the Castle Freitstein map and put it on graph paper to confirm the error. It’s really there. I’m not crazy. YOU ARE.

The problem is the straight stairs from the corner room, Room 30. If you line up the circular stairs and the walls of the castle, those stairs don’t align along the north-south direction. They are disjointed.

I had to check because enough people missed it. And, the thing is, even though I am never wrong, I am open to the possibility that someday I might actually be wrong about something. Open enough to make sure I’m really not wrong. And I’m not.

The thing, too, is that spatial reasoning – the ability to manipulate objects in your mind or to create a three-dimensional image from a two dimension diagram or to judge two spatial options to determine if there’s a contradiction – it’s a skill that varies from person to person. Just like any other skill. Some people genuinely have a harder time than others translating a map to a reality in their head. That’s something you should be cognizant of as a GM.