When players understand the underlying systems they are working with, and can manipulate those systems to get a desired outcome, they are playing the game, and playing the game is where the fun is. What isn’t fun is feeling like you are being deliberately left in the dark and don’t know what to do. Hence the perpetual frustration with dungeon crawls where the party has to choose between exploring a left or right corridor. With no other information you might as well flip a coin to decide.

I’ve seen a lot of suggestions on various OSR blogs and other parts of the internet that a GM should be providing more information and details about these corridors that give a clue about what awaits the party. That’s fine for cases where it makes sense that there are some distinguishing features about a path forward, but as a standard procedure of dungeon design I find that to be a bit of a contrivance. How many buildings can you think of that have completely unique rooms where you can tell what the room will be even with the door closed? But even lets say, okay, this is a fantastical space where everything is unique and has features that signpost what lies ahead, the other problem with this advice is it is at least doubling the amount of prep now required of a GM. And I’ve never seen a good accounting of how one is supposed to come up with and supply these details.

A better approach, in my mind, is to follow Chris McDowall’s advice in Electrric Bastionland and just give the players the maps. Now they have the information to make informed decisions. The map you give them doesn’t have to completely spoil the dungeon, but it should supply them with sufficient information that they can now make informed and interesting decisions. They are now playing the game, instead of merely attempting to play.

Why do the players have a map of the dungeon? If the part of the roleplaying game that is more important for you is the game part, this doesn’t necessarily matter. But if the roleplaying part is more important for your table you probably want some fictionally satisfying reason that the players know what they know. Luckily there are no shortage of potential solutions to this problem.

  • If your game is in a contemporay setting they just download a map from the internet. Even if its for some secret military base or other thing that isn’t readily available they can know someone that pulls the info from the dark web, but now you owe them a favor. So you even have a future plot hook.
  • If your game is in a fantasy setting then magic is a thing, so you can have magic spells, enchantments, artifacts, etc. that reveal the layouts of areas.
  • The players find a map in the first room from dead adventurers that came before.
  • If the players are the type to do any sort of preparations beforehand, you tell them they come across a map. You can give them some bonus XP for this if youwant to incentivize that sort of thing.
  • The players are navigating a space they are already familiar with, but now contains a threat. Think of the film Alien. The drama here is a place that used to be relatively safe, but now isn’t, and has to be navigated under these new conditions.
  • If the players are hired for a job, their benefactor provides a map as part of the standard arrangement.
  • There’s an old folk song that describes the place.

Maps don’t need to be perfect. They can have inaccuracies that hopefully lead to interesting complications. They can be unclear in ways that provide interesting instead of frustrating ambiguity. Getting better maps can be part of the adventure. You can even make a mechanic out of this, rolling a d6 (or whatever is appropriate for the size of your map) to see how many inaccuracies it contains. Depending on the sort of preparations the players take beforehand they can account for these various errors. Although at this point you should probably give some clear indication about what is the correct piece of information, otherwise you land back at the do you arbitrarily pick a left or right corridor problem. We want to enrich the problem space, not reduce it back down to random choices.