Wednesday, October 03, 2012

Building a Random Dungeon with Central Casting

At some point in the past, I ended up with a copy of a Task Force Games product called Central Casting: Dungeons, a random dungeon construction module from the early 90's. Despite having had it in among my RPG gear for many years, I had never really used it in the way that it was intended- to generate a complete random dungeon- but had instead just browsed through it, maybe occasionally using a floor plan for a room to fit into an existing game. It always appeared needlessly complex- rolling for corridor lengths, rolling for random rooms, rolling for room sizes and treasure and encounters- and likely to produce nonsense results that, even for the RPG world, seemed implausible in a real structure.

Still, having recently gotten back more into dungeon crawl-style gaming, I decided to pull it out and give it a whirl. The results: it's just as much of a pain as I imagined. It creates results that are just as ridiculous as I imagined. I gave up on using it after an hour or so. And I still came out thinking it's a great product.

Here's the map I created using GridMapper:

Central Casting: Dungeons is essentially a big book of d100 tables. There are tables to determine how often you roll to see if a corridor changes direction or sprouts a door, tables to determine the type of room behind a door, whether the door is open, locked, stuck, or all three, etc. By changing the 'spacing' of the dungeon (how many blocks of corridor between rolls to determine if a feature is present), you can create a dungeon with long, winding corridors, a dense, house-like dungeon with few hallways, or anything in between. You can also roll up details like how many people occupied a chamber when it was in use, how many bodies are buried in a crypt, whether there is a natural water source in a bathroom, etc. Each room or feature type has a few associated tables to determine the room size, contents, and other details.

Rolling all this out can be a little exhausting. I found myself fudging rolls to reduce the number of 'special' rooms or features- my dice seemed to have a real yearning for high numbers tonight, and results in the 90's are likely to produce 'special' results that require consulting yet another table or performing multiple rolls. Keeping track of what I had determined and hadn't involved a fair bit of book keeping- I ended up writing myself notes to indicate where I needed to go back and determine an encounter or treasure using yet another set of tables, rather than trying to fill those details in as I went.

Pleasantly, encounters and treasure are fully system agnostic- you get Class A-E encounters, ranging from trivial Class A to 'Run Away!' Class E. You can scale what, say, a Class C encounter means for your campaign in terms of difficulty, effective level, amount of treasure, etc. using guidelines for your specific rule set.

Early on, I started to get some weird results. My first room in the dungeon was an 'Exhibit Room'- essentially a trophy room. My next room was a deadly trap. After that was a bath (without installed water supply) and just across the hall from the bath, a Burial Chamber containing one non-undead body. I was starting to suspect that my dungeon was going to amount to random nonsense. A servant's chamber that allegedly had room to sleep 58 people inside a 40'x30' box prompted some contemplation of the slightly weird sense of scale that the standard 10'x10' square sometimes gives you- rooms that can comfortably hold a half-dozen people having dinner look like cubby holes on a standard grid map.

After a couple more odds-n-ends rolls (a bedroom, a trap or two) I hit a real bonanza of new rooms- a 15 room bedroom suite! I started rolling up random rooms for the suite, ending up with 2 libraries, 2 sitting rooms, 2 bedrooms for special retainers, a wardrobe, bath, storage room, and treasure room, as well as a waiting room and a guard room.

It was at this point I finally started to get a vision of what my dungeon had been before it became a loot-or-die opportunity for bold adventurers. Someone important had lived here- someone who regularly received guests, liked books, had some trophies to show off, and kept one dead VIP near the front of the house. I started to come up with a scenario.

The dungeon had been one floor of the home of a powerful noble. He had gone abroad to wage war, returning home with treasure and riches to build a new home and stock a trophy room. But all was not well; soon after returning from abroad, his beloved wife died unexpectedly. He turned the manor's chapel into a shrine for his wife. Troubled by her death and his own mortality, he began to delve into the necromantic arts, ultimately becoming a powerful wizard.

He continued to publicly receive guests, as befit his station. The two sitting rooms? One was a 'public' room where visitors were received, another a 'private' room for entertaining important visitors and intimates. The two retainers' bedrooms belonged to his majordomo and his mistress (he missed his wife, but life does go on). The front library was primarily for show- history books and genealogies for visiting scholar to flip through- while the second library, accessible only from his bed chamber, contained his arcane writings and necromantic works.

At this point, I stopped rolling randomly and just started thinking about how I could build out other levels of this dungeon. What kind of dwelling was it part of? Given the nature of its occupants, a fortified manor house seemed like the best fit. I could add some outbuildings, a courtyard, stables, etc. Other levels could include workrooms and alchemical laboratories, as well as more room for servants and men-at-arms. I started considering some campaign options: where was the nobleman now? Was he, too, dead, and his ghost haunting the manor, waiting to be re-united with his dead wife? Had he become a lich and moved on to other pursuits? Why was the dungeon abandoned? (There are tables to cover these sorts of things in Central Casting as well, but I was more interested in working things out for myself by this point)

At this point, I'll probably go back and re-build this dungeon from scratch, discarding a number of the random results- but I now have a clear idea of what I want to build, how I can create an adventure around it and even how to develop it into a campaign. By and large, the details of traps and corridor lengths didn't matter too much- a few random rolls (the Exhibit Room, the awkwardly positioned burial chamber, and the massive bedroom suites with multiple libraries) started to suggest an idea, and at that point I could just run with it. I can now pick and choose what I need from the stock rooms in the book and build them into something that works for the game.

Odds are I won't be using Central Casting: Dungeons exactly as intended by the publisher- to create a complete dungeon. Instead, I can see myself rolling up a few rooms at random, and then trying to connect them with some kind of plot or back story- was part of the dungeon abandoned, and then converted to a different function? Did two different groups of humanoids live here and modify different sections to meet their needs? Why do you need a music room and five bath chambers?  (obviously a naiad Partridge Family!)

The Central Casting books have been out of print for some time, but there are a few available used on Amazon. I'll certainly be hanging on to mine, but in the meantime I wonder if anyone has created a comparable product- it seems inevitable that during the OGL era someone didn't churn out a similar book covering stock dungeon features. There are definitely some programs on the web that will build a whole random dungeon for you, but one of the nice features of CC:D is that you can pick and choose from the tables- if you have a dungeon, but want some reason for its existence, you can roll or look through that. If you have a good idea of the type of dungeon you want to create (a crypt, a wizard's tower, a fortress) but just want floorplans or guidelines for the appropriate type of room, you can look them up. It turns out to be a pretty nice toolkit- somewhere in between a full generation program and a pre-made and fully fleshed out setting.

No comments:

Post a Comment