This is a continuation of my post from last week Turning Dnd into a Roguelike. Last week I talked about how character creation has to change to better accommodate a system in which players will be constantly dying. This week, I will be covering a topic that applies to more than just roguelike Dnd, and should hopefully be of help for when players go off the beaten path.
Laying the Foundation
One of the first things you have to do when you decide to randomize a dungeon is to figure out the layout of the dungeon itself. There are many different methods of randomizing a dungeon, all of them effective at different tasks. One of the first options is to use the tables in the Dungeon Masters Guide found on page 290. I have used this option before, and have had decent success with it. Another option would be to use Donjon’s random dungeon generator, which can help you create a floor very quickly out of nowhere. Another option I have considered is a stack of cards with different rooms drawn on them which you can draw from a randomized pile of cards. Let’s go over the pros and cons of each option:
Option 1: Use the DMG Tables
- Only need books and dice
- Can be customized the best with different tables, or changing on the fly
- Nobody knows what is going to happen
- Is the slowest to do during the game
- Rolling on so many different tables gets overwhelming sometimes
- Takes up a lot of space
A week ago, when I was first testing this roguelike idea out, I decided to use the tables found in the Dungeon Masters Guide. I’ve used these tables before to help me create dungeons, but I’ve never used them at the table before. We started off with a very barebones dungeon as I struggled to roll on table after table after table. Filling in details was difficult as I was trying to figure out all of the details in the room such as door location, and whether there were any secrets or stairs. Eventually, however, I got into the groove of it, and we started having rooms such as one filled with diamonds and ores, and the next being filled with Lizardfolk chefs!
Getting into the swing of it was not easy, but here is what I would recommend to anyone who wants to do this style. First ignore tables when you can. The passage table is nice, but ultimately just adds time. When moving from one room to another draw a curvy passage and then roll on Table A (Chamber?) to quickly determine what’s behind the door. Other tables to ignore can include Door Type, Beyond a Door, Chamber Exits, Exit Location, and Exit Type. This way the only tables you are rolling on are Starting Area, Table A, Chamber, Table B (Monsters?), and Table C (Treasure?). This cuts down rolling significantly, while still allowing rooms to be connected in meaningful ways.
Table A (Chamber?)
1-3: Chamber (roll on the Chamber table)
4: Stairs (roll on the Stairs table)
Table B (Monster?)
1-3: Monster (Random Encounter)
Add +1 to Monster for each floor past 1. Reset back to 1-3 every five levels. (For example on level two the chance for an encounter increases to 1-4 out of 10, level three 1-5 out of 10 and so on).
Table C (Treasure?)
1-3: Treasure (Random Loot)
Add +1 to Treasure for each floor past 1. Reset back to 1-3 every five levels. (For example on level two the chance for an encounter increases to 1-4 out of 10, level three 1-5 out of 10 and so on).
Next, if you are doing a roguelike session, determine what the theme of the floor is. You can do this by using the next few tables located in Stocking the Dungeon on page 292. You can roll on Table D (Theme) if you want to see what options that might come up with. You can also add your own tables if you feel like, in order to introduce even more variety to your game.
Table D (Theme)
- Death Trap
- Planar Gate
- Temple or Shrine
- Treasure Vault
- General Dungeon Chambers
When rolling on tables in this fashion one of the most important things to remember is speed. If your players are sitting there waiting for you to roll on 7 different tables every time they step into a new room they might get impatient. You can mitigate this problem in two ways. First, you can tell the information as it comes. (Roll. You come across a 20 x 20 ft room. Roll. There are 3 doors. Roll. 1 is on the east side of the room, the other by the door you just came in . . . ) The other way to deal with this problem is to roll on fewer tables. By having only 3-4 tables you can quickly figure out the contents of the room without keeping the players waiting for too long. Combine these 2 methods and dungeon generation can feel surprisingly seamless to your players.
Option 2: Use Donjon
- Is very fast
- Preloads objects and monsters based on settings
- Can be set to your preferences
- Needs a computer/internet connection
- No way to fiddle for exactly what you want
- Might start to feel samey
Donjon is an excellent online resource for Dnd that provides lots of great information for organizing and cataloging all of Dnd’s information. What many people don’t know is that Donjon also comes pre-equipped with a random dungeon generator. This generator is incredibly useful when trying to come up with anything on the fly. It won’t be a perfect dungeon, but the amount of customizability Donjon presents allows you to come up with something really close. Donjon is not the only generator you can use, there are many others out there, but I am used to Donjon so that is what I will be presenting here.
If you want to use Donjon for a Roguelike session, here are the settings I would recommend.
Dungeon Level: Your party’s current level
Dungeon Details: Basic
Dungeon Motif: Your Preference/Random
Dungeon Layout: Your Preference/Random
Dungeon Size: Fine
Peripheral Egress?: No
Room Layout: Your Preference/Random
Room Size: Your Preference/Random
Corridors: Your Preference/Random
Remove Dead Ends?: All/Random
The settings I present here are this way for a reason. For the dungeons size, I have chosen the smallest option possible in Fine. This is because when trying to run a roguelike session, speed is king. A smaller level means the party will be able to go through it much quicker. Spend too long on each level, and you’ll never be able to finish the entire dungeon by the end of the session. This is also why I put the stairs setting on Many. Finding access to the next few floors is very important to maintain velocity, and if your players can’t find the single set of stairs on the level you’ll be losing valuable time. Most other settings can be fiddled with to fit what you like, but for the sake of a roguelike, Many stairs and a Fine dungeon allow for the best experience.
Another thing to keep in mind with Donjon is that you can always change aspects about the dungeon. If it presents you with a dead end even though it touches up to another room, you can totally add a door there. (Or even better a secret door). Don’t like the monster that Donjon gave you? Feel free to change it. You only have to follow Donjon as much as you need to.
Option 3: Deck of Many Rooms
- Also really fast in the game
- You can put exactly the rooms you want to see down
- Can have multiple decks if you so choose
- Longest time to set up before the game (Or you can buy dungeon decks)
- Getting specific details for rooms can be difficult
- Needs multiple decks in order to have granularity
This method is one I really like, but having the time to implement this method seems kind of difficult. The idea is simple. Get a stack of 3×5 cards (preferably ones with grids already drawn on) draw a bunch of different rooms on all of your 3×5 cards, randomize the deck, and then reveal each room one by one. One thing that you may struggle with if you choose this method is determining whether rooms have stairs or not. I can think of 2 different ways to fix this problem. Either 1.) draw stairs on every 4th card or 2.) roll a d4 whenever you encounter a new room to determine whether stairs are there.
Ultimately, this system for dungeon design is one of the broadest, but also requires the most set up time. If you don’t have a deck of cards to draw from, you can’t really perform this method. But, if you do take the time to set up a deck of cards, this system can be as detailed or as simple as you need it to be. Multiple decks can also allow you to customize for the situation while still keeping everything random. This has the advantage of giving you the experience of rolling on lots of different tables, with half the time needed to reference that many tables. It does take up a lot of space but offers a lot of great advantages that other methods can’t match.
What’s in the Room?
The next part about creating a random dungeon is determining what is inside each room. The tables found on page 292 onward are very helpful for figuring out the theme and random items in the room, but I want to talk about monsters and treasure. When determining these things you have to go through 2 steps 1.) is ____ in the room? and 2.) what is ____? We can already determine the first part using tables B and C found above, which scale based on how far into the dungeon you are. The second part is a little more difficult. Previously, I used Goblinist, which created random encounters for me to equal the challenge rating needed to challenge the party. Goblinist works really well when you need to come up with a single encounter for the party. The problem I soon realized with Goblinist is that in the roguelike setting, you have way too many encounters to justify more difficult battles.
The DMG recommends that you have 6-8 encounters in a day. For most campaigns, this is a near impossible statistic to meet, usually because most DM’s only plan for 1-2 meaningful encounters in any given session. Thus, I learned to throw much more difficult challenges at my players with the expectation that they would have the tools necessary to deal with it. In this roguelike dungeon, they run out of resources exceptionally fast. 6-8 encounters are to be expected in this kind of environment, and because of this your encounters should be much weaker. Instead of fighting an entire bugbear tribe, you fight them 1 at a time. The players are less likely to die due to being overwhelmed, but they will still get that experience of feeling like they are running out of resources.
With that out of the way, determining what monsters to fit into the table is as simple as just equaling the CR (Challenge Rating) of the party. 1 Bugbear is equal to 1 CR which is good for a level 1 party. 8 Stirges equals 1 CR, 4 Skeletons equals 1 CR etc. These tables don’t have to be very detailed either. If the players are only going to be on a floor for 1-4 rooms, a table with 10 options is more than sufficient.
The next thing that we have to cover is random items. Again there are 2 different ways we can take this. First is using the roll tables in the DMG found on page 184. The second method is to create our own tables. Method 1 is effective because it doesn’t require very much work to implement. The only issue is flipping through the book constantly to get whatever you need. The second option has the advantage of simplicity, which is something I value highly. Also the tables in the DMG don’t hand out special magic items that might be necessary in the depths of this dungeon.
Last week, I talked about how to handle dungeoneering and leveling at such a high velocity. One of the changes that happened was you wouldn’t get to take a long rest until you had traveled through 5 levels. This made some classes such as spellcasters feel worse because they would run out of spell slots fairly quickly. Introducing items that restore features such as spell slots, or even have the effect of a long rest, can be really valuable to the party and isn’t available with the base tables. This combined with simpler an more focused tables can result in something like this:
Treasure (Levels 1-5)
1-4. Healing Potion
7. +1 Weapon
8. +1 Armor
9. Spell Slot Pearl
10. Long Rest Potion
That’s just a quick example to demonstrate how you can adjust the table to better fit you and your parties needs, and introduce magic items that wouldn’ work in normal Dnd. If you have the time to do this method, I would highly recommend it.
Doing Dnd in this way is not perfect. But, it can provide a unique experience and gaming session that is unexpected and memorable. I think there is still a lot of work that needs to be done in order to get this kind of system working, and it might even be correct to throw all of Dnd’s rules out the window and start with a brand new system. Whatever it takes to get it to work, ultimately, what I want to see is a team based roguelike game, with roleplay, action, and is fun for everyone. And this is only the beginning. Thank you all for reading, I hope you have a great week and an amazing Tuesday!
If you would like to support me and my content, be sure to get all your Dnd and Magic related needs from Game Grid Lehi in Utah.