Can anyone point me to an excellent, visual, article on dungeon design? (or the lost images of a certain enworld thread xD)

Hand of Evil

Hero
Epic
There are a number of older threads on the subject in DM Advice. As for visuals, try https://www.pinterest.com/ or https://www.deviantart.com/

The best one line: What is the dungeon purpose? It is a simple question but one that is often forgot about. Is it a mine, a prison, the home of a powerful wizard, a living creature, a gateway, a city, a fort, a lab, etc.... Once you know that, you can design around it.
 

log in or register to remove this ad

Ilbranteloth

Explorer
[MENTION=50895]gamerprinter[/MENTION] - at first glance that's a pretty-looking map but on second look the dungeon it's depicting is very linear. The only loop is provided by area 3, giving a bypass to the trap at area 2; unless I'm misreading and the gatehouse (area 4) is a second entrance to the dungeon. Either way, areas 5 and onward are a straight line other than one two-room dead-end offshoot (areas 10-11).

This is great if the design intent is to funnel the PCs through the encounters in a mostly-predictable order - tournament dungeons are almost always designed this way for just this reason - but for non-tournament use linear layouts like this aren't often much fun to play through as ultimately the only meaningful choice is to press on or turn back. That said, there's limits to how intricate you can make a 13-area dungeon. :)

I often have strict linear dungeons because that’s what makes sense. I don’t design a dungeon to be interesting as a place for gamers to explore. It exists for a reason.

For example, most tombs are a simple linear design ( and often use maps of actual tombs). Sometimes there are a couple of choices, but for the most part they are just simple tombs.

That’s not to say you can’t Jaquay it. But in that case it’s usually because of natural causes (erosion, weather, earthquakes, etc) or creatures such as burrowing monsters, or prior expeditions that have altered the dungeon.

One of the advantages of this approach is I can do what I always do and design it for an appropriate level. That might be a much higher level that the PCs, but either the dangers have been lessened or defeated, or original dangers are intact and the evidence of other original dangers alerts them to when they are getting in over their head. Usually they are smart enough to take the hint...

If there is a dungeon holding great treasure, I expect that to hold up to setting integrity. Why hasn’t this great treasure been plundered in 3000 years? It’s usually because it is unknown, or so dangerous that even the highest level adventurers that have tried, failed.

To me, interesting dungeon design is more about its lore. The placement in the setting, or the importance to the PC or NPC. What makes it interesting is the circumstance. Sometimes that circumstance is long expanses if nothing.

For example, I’ve seen several maps made for Moria. In most cases, especially those mass for RPGs, it amounts to maybe a hundred rooms. Similar abandoned dwarves cities are around the same size.

I love the ICE MERP version because it’s miles upon miles of passages. Regions where thousands of dwarves once lived. Many levels, where most of the dungeon is left for the DM to fill out (including placement of rooms, homes, etc.) with examples given. In dungeons like this, along with places like catacombs, the players have literally spent months. They aren’t trying to “clear” them or explore every room, of course. They have s purpose, and they move on. But it gives a scale that is much more expansive, and basically forces them to not treat it like a video game and try to kill every monster and get every last bit of treasure.

The point is, a dungeon is linear when it makes sense to be linear. It becomes largely irrelevant the more the dungeon is interesting because of its existence, it’s purpose, and integration into the campaign, whether that tie is through setting or plot. It’s interesting because the players/characters find it interesting. It tells its own story, in addition to being part of this story.

A recent tomb in my campaign used a map from one of the tombs in the Valley of Kings and was entirely linear. The opening was found in the back of a displaced beast lair, long since covered by a rockslide and vegetation. The tomb had clearly been plundered, and now lay in an advanced state of decay due to water and rot. A stream ran down the passage where several traps that had already been triggered or defeated could’ve found, and a few that still worked, but the poison was no longer lethal, and mechanics were degraded or locked by rust. Small creatures used the old tomb as their lair.

They found the remains of many that had tried, and failed, before one group succeeded in breaching the tomb. An slightly open (but hard to reach) formerly concealed door told them that the tomb that was found was a false one. This one revealed the history of more that had failed, their remains cursed to rise in defense. This part of the tomb was subjected primarily to dry rot. It too was plundered, but some treasures remained.

The tomb was also a false tomb, though, and the crumbling walls revealed a chamber that the treasure seekers had not found. They approached it with much more caution, only to find that it, and the treasure room, had been plundered from below. Drow, using dwarven slaves, had tunneled in to steal its riches, and had slain the mummy. They had not taken all of the treasure yet, and those tunnels led to others, and other tombs.

The treasures and information they found among the (un)dead adventurers provided lore, history and hooks. What they are learning about the interred did too. Some useable treasures were found, and of course they are deciding what to do about the drow.

All of this was a linear dungeon until they got to the drow tunnels. Yet it has led, through potential story hooks (to as yet unwritten stories) to more than a half-dozen additional directions that are being actively pursued by several adventuring groups now. And only one death (and three combats...well, really two since one was twice since the undead guardians rise at dawn each day).
 

gamerprinter

Mapper/Publisher
In my previous post, I mention a sci-fi one-shot space horror module, Rude Awakening. While it's supposed to be located aboard a gargantuan starship, only a part of the deck plan is necessary for the adventure, and like the previous posted map, it is very much a linear dungeon in design - in fact the module plays very much like a standard D&D/PF dungeon, with sci-fi clothes and dressing.

Here's the map for that one-shot module...
 

Attachments

  • rude-awakening-full-map-grid.jpg
    rude-awakening-full-map-grid.jpg
    3.3 MB · Views: 87

Lanefan

Victoria Rules
While my published Curse of the Golden Spear trilogy (for my Kaidan setting of Japanese Horror PFRPG) starts a bit linear - the PCs are escorting a merchant to delivery a gift to a powerful lord on some exotic land they take by boat, so it's getting from point A to point B. But the second and third modules start having ramifications if they do or don't do things, and the adventures are designed so the players could choose a different path and that's okay. Some things will get missed, but somethings you pick up and move to where the PCs are. Situations come up that create new incentives to do this and that like removing a seriously detrimental curse the party picks up, and the final goal is to get the hell out this insane exotic land. They have multiple options on how to do it.
This sounds cool!

So a linear adventure can be written in such a way that it doesn't feel linear. It's only a railroad, if the players can see the tracks - so just hide it well, they'll never know.
Agreed.

Problem is, there's players out there (as evidenced by these forums) who don't just look for the tracks, they excavate vast swathes of ground looking for any sign that tracks may have ever passed through here; and only on finding none do they reluctantly admit that maybe - just maybe - there's no railroad this time.

Multi-site linear adventures such as yours above are certainly easier to disguise than linear dungeons; if for no other reason than as the players map a dungeon their map will soon enough show the overall layout and how it's developing.

Lanefan
 

Lanefan

Victoria Rules
I often have strict linear dungeons because that’s what makes sense. I don’t design a dungeon to be interesting as a place for gamers to explore. It exists for a reason.

For example, most tombs are a simple linear design ( and often use maps of actual tombs). Sometimes there are a couple of choices, but for the most part they are just simple tombs.

That’s not to say you can’t Jaquay it. But in that case it’s usually because of natural causes (erosion, weather, earthquakes, etc) or creatures such as burrowing monsters, or prior expeditions that have altered the dungeon.
I also try to design as if living breathing people actually once lived there...but even then it's not that hard to come up with reasons for loops and interweaving: trap bypasses, secret exits, servants hallways vs. nobles hallways/stairs, etc.

A good example of this - and not to pick on [MENTION=50895]gamerprinter[/MENTION] but if he will insist on graciously putting his maps on here for us to analyze... :) - is in the map in post 9 there would logically be a secret door right by the entrance at area 1, going north into the laboratory (area 11), so the owner or inhabitant of the place wouldn't have to go through the whole dungeon in order to go outside or return to his quarters.

One of the advantages of this approach is I can do what I always do and design it for an appropriate level. That might be a much higher level that the PCs, but either the dangers have been lessened or defeated, or original dangers are intact and the evidence of other original dangers alerts them to when they are getting in over their head. Usually they are smart enough to take the hint...

If there is a dungeon holding great treasure, I expect that to hold up to setting integrity. Why hasn’t this great treasure been plundered in 3000 years? It’s usually because it is unknown, or so dangerous that even the highest level adventurers that have tried, failed.
Agreed. Never mind that the great treasure has only become greater over the years by the posthumous contributions from all those fallen adventurers. :)

To me, interesting dungeon design is more about its lore. The placement in the setting, or the importance to the PC or NPC. What makes it interesting is the circumstance. Sometimes that circumstance is long expanses if nothing.

For example, I’ve seen several maps made for Moria. In most cases, especially those mass for RPGs, it amounts to maybe a hundred rooms. Similar abandoned dwarves cities are around the same size.

I love the ICE MERP version because it’s miles upon miles of passages. Regions where thousands of dwarves once lived. Many levels, where most of the dungeon is left for the DM to fill out (including placement of rooms, homes, etc.) with examples given. In dungeons like this, along with places like catacombs, the players have literally spent months. They aren’t trying to “clear” them or explore every room, of course. They have s purpose, and they move on. But it gives a scale that is much more expansive, and basically forces them to not treat it like a video game and try to kill every monster and get every last bit of treasure.
Moria - the original megadungeon. :)

The only time I've ever tried running a party into an abandoned Dwarven city was also the only time I've ever resorted to quasi-random dungeon design on the fly. The place was so big - and the odds of the party seeing much of it so small (they were there on a specific mission) - that there was no point in my trying to map it all out ahead of time. So instead I designed the secret passage through which they'd be entering (long story behind this) and other than a few key locations left the city to design itself as they went along.

End result: playable, but not the best by any means.

The point is, a dungeon is linear when it makes sense to be linear. It becomes largely irrelevant the more the dungeon is interesting because of its existence, it’s purpose, and integration into the campaign, whether that tie is through setting or plot. It’s interesting because the players/characters find it interesting. It tells its own story, in addition to being part of this story.
You can do both.

Take L1 Secret of Bone Hill. It's an old castle, partly ruined, with a dungeon complex beneath. Because the castle is partly ruined there's about 4 or 5 different ways of getting into it; and there's also several different ways to access the dungeon complex below including from within the castle and from elsewhere, some hidden, others obvious. And while parts of the dungeon are linear the various accesses provide vertical loops - the party doesn't have to come up where it went down.

End result: I've played or run this adventure at least 4 different times, and it's never been the same twice. And that's what I want from a module. :)

Lan-"your tomb example reminded me a bit of the old text-based Advent game: the stream, the dry stream bed, the grate..."-efan
 
Last edited:

gamerprinter

Mapper/Publisher
Well, Lanefan, the owner of that first dungeon map I posted is a 10th level+ (probably closer to 18th level) Cleric Lich of immense power and he had this dungeon and pontificate ecclesiastic site built to his intent. So I don't think he needs a side door to his lab. He doesn't need to negotiate the dungeon to reach his residence - a teleport or plane shift will do just fine... ;)
 

Ilbranteloth

Explorer
I also try to design as if living breathing people actually once lived there...but even then it's not that hard to come up with reasons for loops and interweaving: trap bypasses, secret exits, servants hallways vs. nobles hallways/stairs, etc.

A good example of this - and not to pick on [MENTION=50895]gamerprinter[/MENTION] but if he will insist on graciously putting his maps on here for us to analyze... :) - is in the map in post 9 there would logically be a secret door right by the entrance at area 1, going north into the laboratory (area 11), so the owner or inhabitant of the place wouldn't have to go through the whole dungeon in order to go outside or return to his quarters.

Agreed. Never mind that the great treasure has only become greater over the years by the posthumous contributions from all those fallen adventurers. :)

Moria - the original megadungeon. :)

The only time I've ever tried running a party into an abandoned Dwarven city was also the only time I've ever resorted to quasi-random dungeon design on the fly. The place was so big - and the odds of the party seeing much of it so small (they were there on a specific mission) - that there was no point in my trying to map it all out ahead of time. So instead I designed the secret passage through which they'd be entering (long story behind this) and other than a few key locations left the city to design itself as they went along.

End result: playable, but not the best by any means.

You can do both.

Take L1 Secret of Bone Hill. It's an old castle, partly ruined, with a dungeon complex beneath. Because the castle is partly ruined there's about 4 or 5 different ways of getting into it; and there's also several different ways to access the dungeon complex below including from within the castle and from elsewhere, some hidden, others obvious. And while parts of the dungeon are linear the various accesses provide vertical loops - the party doesn't have to come up where it went down.

End result: I've played or run this adventure at least 4 different times, and it's never been the same twice. And that's what I want from a module. :)

Lan-"your tomb example reminded me a bit of the old text-based Advent game: the stream, the dry stream bed, the grate..."-efan

Agreed, although I'd just reiterate that while it's easy to find ways to add loops, etc., I don't do it just for the sake of adding them. In other words, I reject the idea that linear = bad. In the context of the game as a whole, yes. Within a single dungeon, even one that lasts for a number of sessions, not so much.

--

As a side note, in terms of your experience with the way you ran Moria - I'm finding I do that sort of thing more often than not for large complexes. I utilize the map that I have (like Moria, or the catacombs of Paris) as a guideline, with the "standard" locales roughly detailed so I can place them when needed, and do a semi-random generation. Sometimes I'll place them myself, sometimes randomly. Usually the random aspect is in determining what type of area we're in (residential, business, government, etc.) and then flesh it out from there.

The approach has grown over time, and I wasn't entirely happy with how things flowed initially, but tried to identify where I needed to improve, and worked on those. The main challenge with the semi-random generation of the dungeon was to ensure that it had the same immersive feel of the rest of the campaign. But by treating the random rolls in a manner similar to my notes, I could freely decide whether or not something was a appropriate and move ahead. In the meantime, it provided a point of inpspiration to my thinking that didn't exist before.

In most cases, areas where I always question myself is where I think I improvised poorly ("Darn, I should have done this at that point, or not done that." That's why I like to have lots of notes, ideas, and such to draw from, it helps me avoid or reduce those number of issues. However, I'll also point out that the times I think I've done a poor job, my players almost always disagree. They don't notice the things I struggle with, and generally just keep talking about the cool stuff they enjoyed.
 

Lanefan

Victoria Rules
Well, Lanefan, the owner of that first dungeon map I posted is a 10th level+ (probably closer to 18th level) Cleric Lich of immense power and he had this dungeon and pontificate ecclesiastic site built to his intent. So I don't think he needs a side door to his lab. He doesn't need to negotiate the dungeon to reach his residence - a teleport or plane shift will do just fine... ;)
Ah - this makes sense. :)
 

Remove ads

Top