PDA

View Full Version : Advanced Dungeon and Ruins Design



Yora
2016-05-21, 07:36 AM
Given that people have been creating dungeons and ruins for RPGs for over 40 years, and that lots of people have been writing GM advice on the internet for at least 10, I find it quite interesting and strange that there is very little material around that deal with dungeon design. When you look over to videogames, there is quite a lot that has been produced about analyzing and understanding great level design.
So let's talk about this subject and see what kinds of wisdom we can discover.

While videogame level design has to account for the player interactions of videogames, and RPG dungeon design needs to be based on the interaction between players and GM, there's still a great amount of overlap where we can learn a good deal of methods and tricks. I've been watching two half-hour videos about the level design in Dark Souls (https://www.youtube.com/watch?v=ZZ9_RJ2EPo0) and Half-Life 2 (https://www.youtube.com/watch?v=FRBisIZvl24), which I think are full of great ideas how a ruin or a dungeon could be designed for an RPG. Stuff that I've never seen any GM manual or Homebrew designers talks about.

One big thing mentioned in both videos, is to provide spots from which the players can get a rough look over a good section of the area. Not good enough to make out any details, but good enough to get a general idea of the overall layout. In HL2, the idea behind this is that you show the players the position of their goal and they have to find a path to reach it. The game tends to be super linear with pretty much no branching paths, so you can't really get lost. But it still creates a sense of progress.
This is quite different from the usual dungeon design I see, where every door and corner is a comple surprise for the players. Which doesn't matter so much when it's expected that the players will have been to every room by the end anyway. But having information about what lies ahead enables the players to make meaningful informed choices. And meaningful choices are the very essence of agency.

Something mentioned in the HL2 video is the use of "fences". You can see through them, hear through them, and shot through them, but you can't move through them. Earlier on there's windows, which you can only see through, but not interact through. Both of them allow players to anticipate what lies ahead and prepare themselves for what they expect to encounter. This is completely different from "You open the door, there's 8 zombies, roll initiative." You don't have to use this method for every encounter, but I think just using it on some encounters could greatly enhance a dungeon or ruin.

The Dark Souls video talks about shortcuts. Which is something I've seen in a dungeon in an RPG only once, because the unspoken assumption is that the players will be through every room once, movement over longer distances happens instantaneously (because of the way the GM narrates the game), rooms that are cleared once will remain cleared forever, and there are no enemies moving around. Under those considerations shortcuts are pointless.
Now in Dark Souls shortcuts are great because you constantly die or rest, and all enemies respawn. As a GM, there is absolutely nothing preventing you from making cleared areas unsafe again. If you're doing a megadungeon (and the Dark Souls games are basically that) and there's a town or base camp where players return to for supplies, healing, and selling stuff; they will constantly be going back and forth and could run into annoying enemies that will take resources (spells, hp) to deal with. The one example I've seen was somewhat boring though, as it's basically a magic elevator shaft that only the players and a small number of NPCs can use. It's a fast travel system that can be accessed from the main entry of the dungeon and not an alternative route that lets the players avoid some of the dangers. It just skips past all the dangers.
Going into game design for a moment, this is most effective in games where there is very little gain from fighting random monsters. If they give only minimal or no XP and treasure, they are just a (good) annoyance to the players that makes them want to avoid these encounters. If fighting these enemies provides just as much reward as reaching new areas they might even be welcome sources of easy XP and gold. When the player say "Let's take the long route so we can collect some rewards along the way", something is going wrong. That's farming.
The way Dark Souls makes shortcuts possible is by using a kind of spiraling tower level design. As you progress through the level, you eventually find yourself just above or below a spot you've been to before, and from that location you're able to open a shortcut that you didn't even see before. Lower a rope, remove the bar from a door, stuff like that.

Almost the opposite of this is a cool thing done in many of the areas of Shadow of the Colossus. Often you come to places that turn out to be much bigger on the inside than their simple and unassuming entrances made you think. It looks like just a small door cut into a hillside, but when you explore it it turns out to be a tunnel leading to a huge hidden valley with a whole ruined city in it. I think this could make for great moments in an adventure. As a GM my experience might be different from that of people who almost always play, but with most dungeons I've seen, it seems that you almost always get what you were expecting when you reached the entrance. Either because the entrance implies it, or because the party is knowing what kind of place they are looking for.
Those moments when you come out of a small tunnel and into a huge valley or cavern are also great opportunities for letting the players get a look of the area they'll get to next.

Going from the very big small to the very small, one of the first points in the HL2 video is that every room has something to do in it. Even if it's tiny and seemingly pointless. Just something that makes the players stop and think. Say the door is blocked and it needs to be forced open. Or there's a hole in the floor. Or a stack of crates behind which something might be hiding. Even if the very first action the players attempt results in telling them that everything is clear, they still did something. There was a moment where they had to worry if there might be a danger. Even if it turns out there wasn't.
Movement can also be obstructed by tiny obstacles. Say a window is very high and the players have to push a table over to reach it, or giving each other a lift.

And another small thing that I was thinking off watching the HL2 video is how much more interesting a fight can be if the enemies are not all coming from one direction. Now a lot of players never consider the possibility of retreat (because most published adventures are written so it won't ever be necessary), but if the path through which they entered becomes blocked, the fight can get a lot more tension. It doesn't mean that they have to fight to the death and there could still be other paths to escape, but those would be paths the players have not cleared yet. (Or cleared recently, if you have enemies moving around in the area.)

OldTrees1
2016-05-21, 10:44 AM
I think it is important to always remember "X is a tool for Y objective". That way when a tool is less apt (D&D characters can climb fences) then one can focus on the objective rather than trying to force the tool.

Why is looking over the area valuable? It grants the PCs the option to get information to make informed choices.

Why are unlocking shortcut so nice? Partially the fast safe travel, but mostly as a way to make progress in a perpetually dangerous area.



Enabling the PCs the opportunity to make informed choices is quite important in Dungeon design.
1) How does the geometry of a Dark Temple, a Crypt, and a Vault differ?* If there is a meaningful difference then experienced PCs have a chance to not only identify what kind of Dungeon they are in, but also to have a general idea of what areas exist ahead and in roughly what arrangement.

2) Denizens leave traces. Depending on the direction of PC entry, they might travel through areas that have minor signs of the denizens. Part of these traces shape the geometry of the dungeon. So PCs can deduce information about one from information about the other.

3) While side rooms might have few entrances, large main rooms often have many entrances that have notable differences. By altering the relative elevation and view angle of the entrances you can allow the PCs information about the room before they commit to an entrance.

4) Don't be a slave to symmetry. Symmetry often takes a meaningful choice and reduces it to an arbitrary choice. This is fine since not every choice needs to be meaningful and the symmetry can convey information (see #1). But find opportunities to break that symmetry. For example: There are 2 doors into a large room one on either side of a T intersection. Why not have one of those go up to a balcony while the other descends to the main floor of the room?

*Different DM will have different answers but having a consistent answer is quite useful. My answers are:
Dark Temple: Larger regions having access to the next smaller region. The number of entrances decreases with each inner sanctum. However there are more intra-region connections than inter-region connections.
Crypt: Forking pathways with side chambers to minimize the average distance from entrance to tomb.
Vault: Each region-region connection only has 1 inter-region connection. Each region usually has the means for unlocking/revealing the next inter-region connection. But expect at least 1 twist (Ex: One inter-region connection is controlled by the previous 2 regions)

Yora
2016-05-21, 10:53 AM
Putting visible traces of a creature's presence is great. It's a bit like hearing noises in that it lets you notice a creature before it can do anything to you, but it's much more versatile.
If you tell the playes they hear a noise, they expect it to be important because it was mentioned. With visual clues you can just drop them into the description when players enter or search an area. They expect to be told what they see and that much of it won't be of any consequence. They might not even notice that they've already encountered a lot of clues until someone suddenly makes the realization.

nedz
2016-05-21, 11:27 AM
You can be less rigid in your design than a videogame, and the players can provide many of these effects.

Windows / Fences: Scrying etc.
Shortcut: Teleport (Useful only when you know where you are going)

I suspect that you have seen these things more often than you realised - just in a different media.

NichG
2016-05-22, 01:48 AM
One thing I find problematic in tabletop dungeon design is that for tabletop gameplay, repetitive/pointless elements are really bad compared to in video games. It's not that they're good in video games, but it's possible to have something like a 10-second or 20-second reminder in the video game that 'oh yeah, this place refreshes', but in a tabletop game that reminder could take 10 minutes to play through.

So I tend to really try to avoid situations in which the players will repeat the same content.

But that means it can be hard to actually make players feel the utility of things like taming a region of the dungeon, finding shortcuts, being careful about blocking the way for reinforcements to arrive, being efficient with their time, etc. In general, anything involving doubling back or scouting, preparing, and dealing with it as separate stages tends to be tedious to play out - even if the party had someone who could become stealthy and scout out the next few encounters, its a non-trivial amount of time spent waiting out of character for that one player to go, do their scouting, report, etc. Which means that there's a strong metagame pressure to just kick in the door blindly - its more fun, but mostly because there isn't a fun way to handle the alternative.

Since it's a metagame problem, I suppose the answer could be a metagame solution - the first time you pass through an area, you play it out in detail, but subsequent times you pass through treat the area as some kind of fixed resource drain that can be easily calculated on the character sheets without playing through a dozen nearly-identical encounters. So that way, if you pacify a section of the dungeon or find/make a shortcut, there are concrete and clearly communicated benefits to the players without becoming tedious. For example, if there's a segment of the dungeon that refreshes with low-level threats, you could just say 'passing through this room again requires you to expend 3 spell levels and 250gp of resources'. If you can stop the source of the refresh or station a permanent guard there, you can make that room free to traverse.

I think this would work best with a sort of megadungeon setup, where the game is less about 'you're this group of heroes who deal with threats as they come up' and more about 'your this group of heroes who, over time, colonize a dangerous and unknown land (the dungeon) and work to bring order to it'. Creating stable outposts, constructing and protecting short-cuts, maintaining supply lines, etc wouldn't just be clever ideas, but they'd be strict prerequisites for even being able to make it to the deepest areas without totally running dry on resources.

Yora
2016-05-22, 05:02 AM
That sounds like respawning monsters. Which is indeed boring but also makes little sense.

The answer to that is again random encounters. When you pass through an area a second time, you don't know who you will encounter, how many of them, and in what situation. Effectively you got a new area to move through, just without drawing a new map.

nedz
2016-05-22, 05:26 AM
That sounds like respawning monsters. Which is indeed boring but also makes little sense.

It can be amusing - occasionally.

Either through a summoning trap or a portal which must be closed.

NichG
2016-05-22, 06:09 AM
That sounds like respawning monsters. Which is indeed boring but also makes little sense.

The answer to that is again random encounters. When you pass through an area a second time, you don't know who you will encounter, how many of them, and in what situation. Effectively you got a new area to move through, just without drawing a new map.

I consider random encounters to be basically the same as respawning monsters, reinforcing guards, etc. The area isn't different, just the names of the bags of hitpoints that the players have to fight through to get across it (I'm exaggerating a bit here of course). The point is, the players can't really discover anything new, do anything new, etc as a result of the fight - its a tax on their resources and time just to be able to get to the actual new content. Which is a problem, because if I'm running then I want to skip all that stuff and get to the new things, and if I'm playing I want to skip all that stuff and get to the new things, but in a sort of high-level abstract sense, there is a real game design-motivated reason for that tax to be there if you could figure out a way that didn't have more negatives than positives in the tabletop medium.

Yora
2016-05-22, 06:17 AM
Propper random encounters are somewhat of a lost art. The big majority of games released in the last decades don't mention them at all and when they do they don't explain them. And generic random encounter tables for different outdoor environments rarely make much sense either. And those for dungeons are usually generic and not integrated into the background and purpose of the dungeon.
Though admitedly the old D&D editions of the 70s and early 80s didn't do a good job explaining them either. You either follow the rules blindly and discover some interesting things to do with the results over time, or you need someone who already knows them teach it to you.

Kaveman26
2016-05-22, 09:57 AM
Dungeon Design Stuff That Makes me Want to Bash My Head Against A Wall:


"Good news you have just defeated the goblin chieftain, his meager club falls the to the ground besides his lifeless body. In a chest behind his throne is a razor sharp glowing sword that ignites with flame"

"Well why the heck wasn't he using it?"

"The sound of your warhammer caving in the helm of the gnoll underboss echoes off the walls and with a final gasp he falls to the floor with a crash of armor and flesh. Fortunately the walls must be soundproof because not ten feet away the barracks housing his friends remains undisturbed"

So many times there are little things that seem glossed over or flat out ignored. Whenever I build a structure or design any kind of location I try to ask a few basic questions.

Why is the force or entity inside this particular location?

How do they feed and maintain their lair?

Is there treasure inside that they should be using?

Is there treasure inside they would in all likelihood destroy? I.E. Would a troll keep a flaming weapon or break it? Would a Fire Giant be keeping a Frostbrand in some random chest inside his castle?

Malimar
2016-05-22, 10:19 AM
The Alexandrian did a series awhile back on "Jaquaying the dungeon (http://thealexandrian.net/wordpress/13085/roleplaying-games/jaquaying-the-dungeon)", which contains some techniques (http://thealexandrian.net/wordpress/13103/roleplaying-games/jaquaying-the-dungeon-part-2-the-jaquays-techniques) that I sometimes try to use in dungeon design.

Templarkommando
2016-05-22, 01:02 PM
A lot of medieval buildings have something of a formula for how they're built. This is especially true of religious buildings, but it's also true of buildings with a more secular purpose. The dungeon that I am working up presently involves the party going to a monastery which has a structure based roughly off of the structure of a medieval monastery. This being the case, I used this as a starting point: http://www.timeref.com/life/abbey5.htm

This made it easier to know what sorts of rooms could be expected throughout the course of the monastery, and incidentally made it about a lot more than "You walk into a room it's empty except for a couple of random bits of refuse."

Instead, even in the case of a functionally empty room, I'm able to give a description for a room that explains why the room is there rather than "just another room." Here's an example:
"Scriptorium: You walk through the door here, and the smell of the room immediately assaults your senses. You can smell the sweet smell of books here. A high, vaulted ceiling tops the room and several stone pillars support it. You see Several long tables that have bits of parchment lying atop them. Scattered along the tables are ink wells and quill pens. Several of the pages have writing on them, but they all seem to be in some foreign language that you don't understand. One of them has a pretty picture drawn in it - a portrait of some monk. There are several high shelves with books lining them. There are several sconces in which rest lit clay lamps. In addition there seems to be a device used to bind the books. Near the device are several large leather covers."

You can follow this method of planning for a lot of different buildings by basing them off of the floor plans of actual medieval buildings. If you feel so inclined, you can also take pictures off of the net - actual photographs even of such places - and use them to help your party imagine the places that they are adventuring. You can borrow from the floor plans of castles, court houses, palaces, manors, estates, and I'm sure there are even ruins and dungeons that have been mapped by archaeologists. For example this: http://tinyurl.com/hk9jfjw
is a map of actual Roman Catacombs. It's a little bit complex, but you can incorporate the idea of that complex maze-like structure if you feel that would add to the adventure.

There are also videos that you can find that describe these things. I didn't know what a sacristy was, so I was able to find this: https://www.youtube.com/watch?v=VcaMJotrOhA

Even though it was short, it at least gave me an idea of what was being talked about.

Storm_Of_Snow
2016-05-23, 09:08 AM
"The sound of your warhammer caving in the helm of the gnoll underboss echoes off the walls and with a final gasp he falls to the floor with a crash of armor and flesh. Fortunately the walls must be soundproof because not ten feet away the barracks housing his friends remains undisturbed"

I ran an adventure at university years ago that had the caves the PCs were clearing out behind a waterfall, so the sound of the water would obscure all combat noise for that exact reason (me as DM and 5 players with only one other experienced player, so I took it a little easy on them).



So many times there are little things that seem glossed over or flat out ignored. Whenever I build a structure or design any kind of location I try to ask a few basic questions.

Why is the force or entity inside this particular location?

How do they feed and maintain their lair?

Is there treasure inside that they should be using?

Is there treasure inside they would in all likelihood destroy? I.E. Would a troll keep a flaming weapon or break it? Would a Fire Giant be keeping a Frostbrand in some random chest inside his castle?
My first question would be "Why is it there?", and work from there, even if it's no longer in active use by the original builders or their inheritors.

As for the treasure, maybe the troll could have a flaming weapon in their lair, if they ambushed the original owner, killed them before they could draw it and it needs to be activated in some way. Perhaps the Fire Giant has the Frostbrand hidden away so that he can make a move against his chief and take over the clan, just as soon as he finds a pair of gloves or something similar that will allow him to hold it.

Yora
2016-05-23, 10:41 AM
I found this interesting article (http://udan-adan.blogspot.de/2016/03/the-noble-art-of-adventure-design.html) on the topic on Against the Wicked City. (Relatively new site that I think should get better known.)
It also goes into the same idea as the HL2 video that there needs to be things to interact with, but takes it much further to make use of the potential of RPGs. Putting little things for 30 second activities into each room makes the place feel more interesting and creates a better sense of pacing, but as a GM you can give the players a lot of stuff to play this that they might use or not and even take to completely different areas of the dungeon. In a videogame you usually want to avoid backtracking as it can take a good amount of time during which nothing happens, but in an RPG you can just instantly get to a place without sitting through the boring legwork.

N810
2016-05-23, 11:47 AM
Some interesting links:

http://journal.stuffwithstuff.com/2014/12/21/rooms-and-mazes/ <-- procedural dungeon creator (well rooms and corridors that is)

Chauncymancer
2016-05-27, 07:17 AM
TheAngryDM's blog contains a great deal of content on the topic of dungeon design, inspired greatly by the metroidvania school of video game dungeon. Right now he's working on a Mega Dungeon Mondays project that's a step-by-step breakdown of how to design a long-scale dungeon exploration campaign.

Yora
2016-05-27, 07:40 AM
Angry made me start playing Super Metroid. And it really is a very interesting game level-wise. (But also in general.)

Florian
2016-05-27, 07:57 AM
@Yura:

Consider one thing first: In video-games, level design equals story design. The things you mention are based on visuals and equal to prequel, foreshadowing and meta-plot when it comes to story design.

The question is what you want to achieve and that should inform you on what design tools to use to get to that result.

Traditionally, a "dungeon" is site-based an either created on the basis of a simulation or it is challenge-based, potentially including back- and side-tracking to reflect that challenges and capabilities improve along the way.

In this way, story is created in retrospection by looking what has been achieved and overcome along the way.

The other thing is "dungeon as story". This too can be challenge-based but it is mainly a framing device for the overall story to be told. In this mode, if you´re not going meta all too hard, the individual rooms of the dungeon don´t really matter unless they help drive the story onward.

In this way, a story is pre-planned and just take a break when players do anything that doesn´t help taking the next step into it.

N810
2016-05-27, 09:09 AM
Oh found a better one:
http://www.myth-weavers.com/generate_dungeon.php