So christmas vacation is over and I'm looking at my scribbled notes from last game, realizing that:
1.) I need to start writing neater, some of this stuff is incomprihensible - seriously; what is that supposed to say? is that an "E" or a "T"?!
2.) I don't know what I'm doing with this plot thread anymore: there are some corrupted dryads to the northeast, and some bandits with their pet zombies to the south but hell if I know what I was thinking.
And now that I'm doing some top grade ass-pulling and beating this stuff into some semblance of a plot, I can see that I'm going to do another labyrinth.
Last time I did one I made an incredible detailed map with twists and turns, I populated it's hallway with bizarre undead, I played creepy music in the background and described the haunted ruins of a crashed sky-city with as much anachronistic panache as I could muster. As the players navigated the winding halls I kept meticolous track of where they were, but didn't draw them a single line on a paper - all map-making was done by them without my interference. I though it was a pretty good setup before we played, I was even a bit proud.
And it was s#!!, utter waste of time.
I could have railroaded them through the whole sequence with some good descriptions and a couple of wandering monsters and the result would have been the same.
After the game I asked for feedback and we all agreed that while the mood of the labyrinth was cool, the execution was a bit of a waste of time and it didn't add anything to the game (except giving me more stuff to keep track of).
So now I'm seeing that the corrupted dryad that I'm using as a boss for the end (who might be a fungus queen in the end) is going to make her lair in a old overgrovn druid grove surrounded by a thorny hedge maze.
In the final encounter I'm going to sprinkle treants, shambling mounds, maybe some effed-up redcaps and other fey/sylvan stuff - all of it in a hazy fog of fungal spores and sharp thorns. I might throw in some spiders, because, hey: why not? and I just saw the hobbit movies again so steal where you can.
thanks for reading all that nonsense, here is the actual question: how do you handle mazes and labyrinths?
do you make it a series of skill checks with matching descriptions? (climb to get to There, stealth to go over There, acrobatics to leap over that to end up There etc)
Do you make a map of the whole thing? either by hand or by using a generator?
Do you put the whole map on the table as they enter it? do you cover the map with post-it-notes and remove them one-by-one as some sort of geographical strip-tease? or maybe just describe the turns and let them make their own map?
Or how about something like the chase-cards? just slightly refluffed for the game ...
I'm stumped on how to do it this time, I have untill next week to come up with some cool way to describe being lost in a maze to my players - and this time I'd like it to be more fun.
| 1 person marked this as a favorite. |
The main problem is that if "fun" and "lost in a maze" aren't mutually exclusive...they're pretty darn close.
Mazes make the game slow to a standstill, especially if you enforce "lighting" (players can only see the bits of the map their character can see at any given time).
Eventually you'll see one of two things: Frustrated players listlessly trying passage after passage until they find the right one, bored out of their skulls, or players begging you to please just get on with it and move them to the next thing that matters.
How I do labyrinths: I don't. I hate them as a player, and try to avoid them as a GM.
You can keep the hedge maze flavor without devolving into a full blown labyrinth. Make it more of a compound, with things to explore.
New things to uncover is nice, and makes slow progress more rewarding. Especially when the PLAYERS know where they need to go, even if their characters do not. They can have fun exploring, knowing there's something worthwhile somewhere, but always with the knowledge that when they really want to, they can simply navigate the maze to get where they're going.
Basically, all exploration is by choice, instead of forced, and all frustration is in character, not out.
You can still DRAW it as a maze of course, but running it as a more standard dungeon but with PCs being able to see the whole map at once (sans some secret bits) will likely be more fun.
I did one successfully a couple of years ago by pre-drawing the entire thing on grid paper. Then, I made a handful of separate map sections on flat circular discs. They were designed to fit into specific places on the main map, and "rotate" so that the walls of the maze could actually shift as the PCs moved through. When the first "rotation" was triggered, I lifted any minis on those sections, placed the discs so the map exactly matched what was on the main map sheet below, re-positioned the minis, then did the rotates - and then I left the discs in place and could just rotate them every time a rotate was triggered.
This was a pretty high-level adventure, so I also included some teleport gates at various dead-ends in the maze, and a rotating key for where the portals led - that is, the portals were labeled as #s 1 through 10, and to start with, they connected to their portal # + 1 (portal 1 led to 2, 2 led to 3). But each time someone passed through a portal, the "add" number incremented by 1. (It was the same add number for all portals.) So if three people in a row ran into portal 1, the first would emerge from (+1=) 2, the second would emerge from (+2=) 3, and the third would emerge from (+3=) 4. And if the guy who emerged from 4 then jumped back into portal 4, he came out of (+4=) 8.
The combination of predicatable rotation (although they had to figure out the rotation trigger) and destination-variable portals made it so that even though I displayed the complete map right from the start of the encounter, it was still a cool and challenging encounter area.
|
| 1 person marked this as a favorite. |
I have a player who likes to hit the sauce when he plays, and was playing a PC who was supposed to be dumb as dirt, but with a habit of becoming a strategic autistic savant if his PC was threatened.
In a maze where they'd been separated, with monsters who could mimic the PCs voices, I knew he'd metagame to avoid being fooled, so I waited till he went for a smoke break, and moved all their individual sections of map around the table (to the other players' amusement). Then they agreed to shout out the 'PCs' calls for help, when given my cues.
Even though I left marks on each tile to show what direction they'd travelled from, the change in relative table position was enough to throw him off, and we just let his own blatant attempts to metagame send him charging deeper into danger, as his character would have done.
"Come on guys. Where are you? I keep getting sneak attacked by these stupid birdmen. I thought you were round that corner?"
| 1 person marked this as a favorite. |
I find it depends on your audience. I had a buddy in my game for a while that was a true old-skooler. He genuinely enjoyed strategy, resource management and meticulous gaming, so for him scouring a labyrinth was a lot of fun. For my current set of gamers, some of whom have the attention span of a 12 year old raised exclusively on Vine videos, the experience of delving a megadungeon is pure torture.
For something in the middle, I try to break up the delves.
Take any section of dungeon. Now connect up, say, 12 rooms and draw a circle around that. Now ignore the entire rest of the dungeon. Add in cave-ins, impassable magic fields, or teleporters; whatever you need to explain this section separate from the rest.
As for running this delve, keep things light. Plan, say, 5 rooms. These are your set pieces. Each of these have some connection to the others. For example you might have room 1 with trio of skeletons bearing a particular rune on their forehead. Then room #3 has a necromantic spell trap guarding a treasure with the same rune. In room #7 you have some cultists preparing a ceremonial knife who boast that the blade's sister is already primed and ready. Room #10 ends up being a sacrifice chained to an altar with the blasphemous priest ready to plunge said sister-blade into the heart of the victim and finally upon the defeat of the cult the blade disappears as well as it's twin, only to reappear in room #12 in the hands of some epic final boss guarding the way out of this section.
The other rooms in the section then become random. Make some random charts and have an idea of how to run the stuff on them. These can be anything; a monster, a trick/trap, some treasure, or just empty space. These "filler" chambers don't have to make sense with the this section. If in the above example you threw in a wide hall for room #4 and in that you randomly rolled a quartet of kobolds, you might rationalize them as interlopers from some other section. Maybe they have a hidden way in that is Tiny sized; they squeeze the whole way and certainly someone in the party isn't going to fit through that path.
By dividing the delve you give folks just enough to chew and swallow in a session. Within each should be a small story with a beginning, middle and end. This way your players always feel progress.
A note about detail: I find less is more. Pick out a couple of senses and tease them in each new section. It can be any of the senses though. Imagine if you had zombies in a room for weeks what that room would SMELL like. Alternately if you had moisture in a chamber for decades as well as benign molds, how would it FEEL to the PC brushing against the wall? The most impacting impression your players will get is the one they form in their own heads, not the one you or the flavor text puts there.
The main thing is to keep it interesting. If you have corrupt dryads and a hedge maze, throw in some fey creatures. Since fey are weak physically but good with spells and masterminding/charming others, pair these creatures with others in interesting ways. A sprite with its cantrips swapped out for some kind of necromantic ability and then it's riding inside of a bloody skeleton; the silouhette of a satyr playing his pipes while a raven swarm attacks (he's actually invisible somewhere nearby); a cockroach swarm reskinned as common forest beetles but within it's mass are lumpy shapes - these are mites with levels in witch and they're using some kind of homebrewed hex to move in the swarm w/out being injured/distracted.
Lastly about HOW to run it: I like your idea of letting the players map it. Yeah, it slows things down but I think this directly involves the players in what's going on. If they don't map, they don't know where they are, period. Beyond this and the less-is-more detailing, the individual challenges are no different from other encounters.
I hope this helps.
The main problem is that if "fun" and "lost in a maze" aren't mutually exclusive...they're pretty darn close.
Mazes make the game slow to a standstill, especially if you enforce "lighting" (players can only see the bits of the map their character can see at any given time).
Eventually you'll see one of two things: Frustrated players listlessly trying passage after passage until they find the right one, bored out of their skulls, or players begging you to please just get on with it and move them to the next thing that matters.
How I do labyrinths: I don't. I hate them as a player, and try to avoid them as a GM.
You can keep the hedge maze flavor without devolving into a full blown labyrinth. Make it more of a compound, with things to explore.
New things to uncover is nice, and makes slow progress more rewarding. Especially when the PLAYERS know where they need to go, even if their characters do not. They can have fun exploring, knowing there's something worthwhile somewhere, but always with the knowledge that when they really want to, they can simply navigate the maze to get where they're going.
Basically, all exploration is by choice, instead of forced, and all frustration is in character, not out.
You can still DRAW it as a maze of course, but running it as a more standard dungeon but with PCs being able to see the whole map at once (sans some secret bits) will likely be more fun.
This. This is golden advice.
Mazes that are focused on being difficult to navigate are borderline impossible to keep interesting. I did have success with a floating dungeon that changed paths every 5 minutes, but it was a splitting headache to keep track of all the moving walls.
In my experience, I think what makes a good dungeon (besides interesting fights) are the puzzles. A maze is far less entertaining and rewarding than a good puzzle that relies more upon the player's intuition than a die roll. I once made a room that was balanced on a single point at the center, so the PCs had to continue shifting their center of weight around to check all the side doors or risk falling down. I like to sprinkle some riddles too, and I often reference this page for ideas.
What do you want out of it as a DM? Does the mapping matter to you at all? Can the players dodge the whole thing with a fly spell or a climb check and wooden planks?
I typically do mazes completely off paper. I have a d6 to d20 results sheet with various dead ends, encounters, traps and solutions in the numbers. Successful player skill checks (survival, dungeoneering, intelligence) cross off some numbers. Smart player thinking (marking the walls, etc.) crosses off some numbers. They wander, I roll and eventually they make their way out, sooner if they're smarter or lucky.
If the maze is bypassable, but player progression requires certain amounts of xp, I add bypass encounters (manticore air guardian for example).
The advantage of this is that you can focus totally on flavor and skip the horrors of actual maze map drawing.
|
I think the old-school practice of actually trying to basically reproduce the GM's map is a bit... dated. I had an experience with that in a recent PFS adventure, where one of the lesser success conditions basically required the players to draw what they saw on the scenario map. I thought that was just busywork.
Mazes are a staple of fantasy stories and therefore there ought to be a way to make them work in RPGs. However, it's about perspective. In a movie or a book, the actors are bearing the load of scouting the maze, not the audience. The audience just gets to see the interesting parts. Movies are aimed at pleasing the audience, not the actors. But in an RPG, the actors are the people you need to please.
I would try to do it like this; I'd use graph maps rather than drawing an actual maze. Instead of a hundred passages, I'd create 3-10 nodes and connect some of them with lines. The nodes represent interesting and different places in the maze. Lines show that it's possible to go directly from one place to another without passing through a third place.
So let's start with an entry node, which is the area where the PCs enter the maze. If they're at the entry node, they can leave the maze if they like (provided nobody locked the door behind them). The entry node is connected to at least two other nodes. In fact, every node must connect to at least two other nodes. Also, there's a destination node. It might be an exit to the maze, or a treasure room. The PCs are trying to get there.
To travel from one node to another, the nodes must be connected with a line. Then, the PCs roll some sort of check and if they succeed they move to the node on the other edge of the line. If they fail they move to a randomly determined different node that's connected to the node they just left. Checks might be Engineering, Survival, or something else, depending on the kind of maze you're in. If the PCs achieve a critical success (hugely exceeding the DC) they now sufficiently understand the path between those two nodes and can travel it without checks in the future, and the time required to travel is also decreased.
Nodes are the important locations, which is where you'd put stationary encounters, major clues, treasure, prisoners and so forth. When the PCs are in a new node, take a moment to describe it, and give it a name. These are the places that would get an actual scene in the movie, and the ones you'd put on the map.
The lines between nodes are more abstract. "Many cramped cave passages", "a maze of staircases and short hallways", "lots of hedges that block your view for more than 10ft in any direction". Make sure to add some description. No keyed encounters happen here, although there should be occasional random encounters; not too many though. Keep this part low-CR compared to the nodes.
When the players are in a node, make sure to describe the different choices available to them. Choice is important. The players need to know how many lines they can choose from, and what they're like. There can be more than one set of lines connecting two points, with different terrain. There might be a choice between a wet-slick cave network with lots of obstructions, which is nice if you like mid-range attacks, or a more even terrain, which lets monsters and PCs close in faster.
Players will usually not know which lines lead to which nodes, at first. However, it's possible they have a map, or interrogate locals to get directions. Then you can spoil the destination of some lines.
You might want to look into the Harold of the Ivory Labyrinth. The PCs are spending the whole book going through a literal world size maze, just to get to a continent sized maze inside (that again has to be navigated).
One of the things the book suggests is to actually handwave most of the "maze" elements.
In the large world maze, the players make a survival check. If successful they can move to the next maze area after 1dx days (different values for each maze area). A failed check means they wander this area of the maze for XdY days and then can attempt another check to try to move on.
In the smaller maze, there are about 10 specific locations within the maze. Each location has a list of other locations that it is "attached" to. If the players are wondering aimlessly, there is a percentile table to determine what location they eventually wander into. When the players leave an area, they can attempt a survival check to try to get to a specific location, but a failure means you just roll on the list.
Throw in some random encounters in between locations and their you have a maze that you didn't have to use 5 sheets of graph paper and 8 pencils to make and it helps to eliminate the drudgery of going corridor by corridor. Also having the survival check puts the ball in the player's court. If they are continuing to wonder the maze aimlessly, it is due to their poor rolls and not b/c you as the DM are "being a prick."
I agree with the sentiment of having players navigate the maze through skill checks. It worked really well in Delirium's Tangle as well. Maps always provide a nice point of reference for the players but the power of theater of the mind shouldn't be ignored and it seems to add to the sense of disorientation that a maze is supposed to create.
I created a labrynth once that consisted of a combination of walls of force, and illusion walls that appeared as transparent walls of force, yet were actually solid depicting a transparent wall with nothing behind it - even though there were often something waiting behind the wall. I also included illusion walls that hid pit traps appearing as no wall with solid floor behind it. From all appearances at start it appeared as a maze of invisible walls.
One creative PC began firing arrow shots from her bow to determine which walls were solid and which weren't.
A rectangle of solid wall surrounded the entire labrynth, though some of the socalled solid walls were illusion walls hiding minotaur in side chambers.
|
| 1 person marked this as a favorite. |
From what I understood, my experience with mazes and labyrinths in general is same what everyone else is gonna say: they are boring, redundant and provide little good game experience. I am sure however that you already heard this multiple times, so I'll jump to advice section.
Mazes & labyrinths can only be done (if the area is large) through the use of skill checks. Players can recognize the failure of a bad die and everyone is familiar with them. Several maze-like "attempts" that I saw in AP or PFS scenario's can confirm it. Even authors themselves realize that drawing the entire maze is loss of time so they turned toward skill checks. When you concluded so, the question is how to build it. I recommend several suggestions below:
a) Several successful skill checks in a row are enough to move PCs through entire maze. Advantage and disadvantage of this approach is that it's simple and fast. It can resolve through simple rolling, little game immersion and several random monster or hazard encounters. It has good and bad sides, but keeps the game moving.
b) Successful use of skill check reveals part of the maze. If you have enough time, you can make this work in both simple and role immersive way. Make the maze on paper (it doesn't have to be extra carefully made) and split it into several sections. Now cut of those sections and each time that party makes a successful skill check, they reveal a new section. It's simple and players don't even have to bother drawing it. Just set it on a table. Perhaps on a larger A3 paper. Whatever you can put in those sections is up to you. Perhaps parts of a big key? Answers to a specific question?
Whichever way you choose to make a maze or labyrinth area, the main idea is to keep the game flowing.
Adam
A lot of good advice here, I think usinge the node-path-system several have suggested with some interesting locations and skill checks will be the most fun. I'll draw up some encounter charts and try it out next week when I'm GM'ing.
Thanks for all of the suggestions, this is why I love these forums.
The main thing is to keep it interesting. If you have corrupt dryads and a hedge maze, throw in some fey creatures. Since fey are weak physically but good with spells and masterminding/charming others, pair these creatures with others in interesting ways. A sprite with its cantrips swapped out for some kind of necromantic ability and then it's riding inside of a bloody skeleton; the silouhette of a satyr playing his pipes while a raven swarm attacks (he's actually invisible somewhere nearby); a cockroach swarm reskinned as common forest beetles but within it's mass are lumpy shapes - these are mites with levels in witch and they're using some kind of homebrewed hex to move in the swarm w/out being injured/distracted.
And I am using all of this!
Great ideas, gave me some others of my own as well!You sir, are awesome.
I don't know how but in 30+ years of running games I never once thought of the node/skill-check scenario. That's genius. So playing off of what I already do (make 5 - 10 room dungeon sections) I'd simply add several feet/miles/whatever of "uninteresting space" connecting them.
The party doesn't know it yet, but they are going to put a stop to the torture, one way or another.
So back to the party. I've created node 1: The Ruined Monastery. There are 6 rooms and the current storyline to this node is that a cult from the town uses some hidden section of the ruins to make vile sacrifices. The party has decided to stop them. I have a handful of goblins serving a cult of 6 humanoids who keep a pair of worgs as guards.
This node then connects to three others: The Worg Den - a nearby series of caves that serve as the breeding ground for the worgs but is also infested with giant burrowing vermin and some random aberrations; The Pit - hidden under a hidden trap door a vertical shaft descends deep into the earth where the cult drops their sacrifices (an outsider served by undead here is currently dealing with an annoyance of kobolds); The Gobbleywood - a hundred acres of dark thickets and bogs housing scattered goblin lairs and other monstrous beasts but also plagued by mischevious fey who want everyone to leave this wilderness so they can end the corruption.
Now the "connectors" then can be described as a paw-path through the rugged hills that circles around the base of the dungeon site for a couple miles where it eventually finds the worg den, a hedgerow that disintegrates into the chaos of the thickets of Gobbleywood and an 80' vertical shaft hidden by a secret door. None of these pathways, in and of itself is particularly interesting so you handwave their navigation by saying "Give me a skill check." In the case of the outdoor paths these might be Survival, Knowledge: Local or Nature or maybe other helpful ones. With The Pit they need to manage a tough Perception check and then, say, four Climb checks before 2 failures. If they make it they're at the bottom; if they fail they're at the bottom with 4d6 falling damage.
Do I have the method correct? If so does anyone ever add interesting in-between adventures? I'm running a homebrewed megadungeon and the trouble I'm having is when my PCs make a lucky choice and end up moving way deep into the place, suddenly they're in an area their current APL can't handle so I either TPK or scale back the adventure. I suppose with the node/connector thing you could put in warnings like "Yes, you've opened the trap door to the pit but an overwhelming stench of death meets you. There is a palpable sense of evil, such that you've never experienced before. You have a sudden recollection of an old legend of this place that there was a well in the heart of the monastery that a powerful paladin descended and never returned from; the symbol of her god, Iomedae is etched just inside the lip of the descent"
Anyway thanks everyone in this thread for helping an old dog learn a new trick/trap.
@Mark Hoover : what I would do (and I am next week) is to add "doors", they don't neccisarily have to be actual doors, but they perform the same function - you can't pass.
maybe the key is just the right directions or a map from an npc, maybe it's some sort of magical or alchemical protection agains a hazard that hindered the partys progress.
It could be a magic lantern that keeps the draining fog away, a set of skiis, climbing gear, a potion of flight to get past a big wall or anything similiar.
but just make sure it's not something that party already know how to do - if you tell the fighter with a full climbing gear that he needs to go get "the special climbing gear" from somewhere else, he will call b+!$%#@#, and/or possible flip the table
1 starting location: the entrance to the hedge, there are 2 paths through the tangled briars. Each of the 2 next nodes also have 2 paths each, but they share one destination. the last 3 nodes have a single path leading to their end.
000090000
7000+0008
|000+000|
2-3-4-5-6
000\0/000
000010000
(it's hard to get the symbols to align, hope it's understandable)
taking a page from all bioware games EVER, I am going to have 1 or 2 "keys" hidden in the 2 end nodes (7 and 8) that will "unlock" the path to the boss (9).
between each node I'll place a skill check, failing it will send the party to random node.
for s~&~s'n'giggles I might even make all the location of the nodes (except 1,7,8 &9) random -I'll just roll a dice when they enter and write in the number on my map as they explore.
So far I have:
some deranged sprites animating some corpses in a clearing with a ruined wall of some overgrown fort that has fallen apart.
a lone uncorrupted satyr fighting off several attackers, the pc's can save him and get some explanation of the what and why
(he will probably have an eyepatch and be real badasss and bleeding and such, because that's the kind of npc he is)
a farmhouse filled with witch mites and bugs, there is probably something cool to find there - like a weapon that bypasses fey DR, because none of the players have one.
Crazy fungus dryads - maybe doing something effed up, to clarify that they are infected and EVOL. maybe some kind of "bathing nymps"-like scene, only gruesome.
Somebody needing to be rescued from something big, treant/shambling mound/hangmans tree. lots of "I can't take any more of the effing trees in this effing forest!"
7 and 8 will have some plot stuff that will allow access to 9.
9 will be an all out battle royale in a defiled druid grove with a fungus queen (that I might need to scale down because the party is barly lvl 9) and a pile of everything they have fought so far.
She will probably say something heartwrenching as she dies "mother? is that you?" etc.
yay, the heros saved the land from the evil of whatever, have some loot, time to go see the mayor and get rewards, btw have you heard about our bandit problem to the south?
still need some detail but I think I have a plan, steal it if you want!
|
I rarely do labyrinths.. and if I do I need to answer a few questions first...
1) why is there a labyrinth?
2) who built the labyrinth?
3) why do I want my players in the labyrinth?
I'll elaborate later...
Edit:
#3 is the most important. There are a few answers to this...
- want a dungeon crawl without the dungeon...
- mapmaking
- minotaur encounter#1 comes next...
- to harden access to any1 who can't just teleport in...
- to keep prisoners
- to entertain your kids
- to hold a minotaur..#2 answers labyrinth material
- a druid probably use hedges
- wall of stone/iron/force...
- dwarfs might have excavated one...---
Once these questions are answered I figure out what to do...
If you know that your players don't find it fun to map a labyrinth, don't do the mapping... Keep it small and simple.
Make sure the inhabitants in/behind the labyrinth have easy ways to get through/around... (yes, this gives the players a chance to dodge the labyrinth...)
I'm always surprised minotaurs don't build the damn things more themselves. I mean they have a natural innate ability to navigate the things. That's like humans knowing they like to live in these things called 'houses' that offer them protection and warmth but having to constantly live in hollowed out houses made by eccentric weirdoes.
As a DM, you can literally disguise a maze as an entire dungeon complex, with different people living in different parts of the winding madness.
Mazes can be fun, just not 'five foot corridor' mazes of blank meaningless walls.
Those things are atrocious nightmares.
One thought is to run it as a skill challenge. This way, you only need to map out certain sections where combat happens.
|
|
One thought is to run it as a skill challenge. This way, you only need to map out certain sections where combat happens.
In essence, yes. But it's important to make it into a good skill challenge; one that requires the players to weigh choices that matter, rather than merely brute-forcing their way through with high rolls.
My graph idea attempts to do that; there's an element of exploration and making choices in there, but also of skill. If the PCs suck at navigating, they'll be running in circles for a while, meeting lots of annoying small encounters that drain resources.
By the way: I played Delirium's Tangle, and I really didn't like the maze exploration aspect. We lacked the right skillset and had to just force our way through with copious use of a Wand of CLW. It was just a "naked" skill challenge, there wasn't really any player input involved.
We got to choose the skill to use, but I didn't feel that the choice of skill had any impact other than to determine our bonus/DC. Nothing changed in the encounters based on the choice of skill.
In the end we hit the center because the scenario says "after X failed checks, you succeed anyway". It put an end to our boredom, but it wasn't really satisfying.
---
A super-easy way to do it is to turn that 90 degrees. Say that navigating the maze will take X checks, regardless of success. But every success avoids a dangerous hazard, while every failure means a nasty encounter or setback of some kind.
A bit more sophisticated is to phrase it as a series of choices: between easy with no rewards, or harder with some rewards. Example:
"You can take the long way around, which will cost you some time, or you can try the unsafe-looking rope bridge."
"You can just continue here, or you might climb up to the statue on the ledge to try to grab its gemstone eyes."
As a variant there's the choice of perils. "Do you want to brave the passage filled with spider webs, or the one that smells like sulphur?"
These are all still not truly mazes, because there's no getting lost involved. You're basically on a railroad to the destination. But you can pick some of the routing.
---
You can simulate getting lost fairly easily, by just moving people back some spaces on the track. But that doesn't really mean anything unless being moved backwards is bad in some way. That can be easily implemented though;
- It's not safe to rest in the maze. The longer you're there, the more wandering monsters you run into. Getting some good rest takes work, and it's not guaranteed to succeed every time.
- Food supplies are limited. You can't just grind the maze for a few months. Note that this doesn't work after level 5 or so, when Create Food & Drink comes online.
- Time pressure. While you're wasting time in the maze, things are happening. Enemies are gaining on you (or getting away), princesses are getting eaten by minotaurs, curses are reaching the moon phase where they become incurable, etcetera. This is probably the most reusable of all the reasons.
- Enemies are adapting. The longer you stay in the maze, the better enemies get at attacking you. There might be an overseer watching you, sending in more specialized troops. Or you might be the target of an alien reality TV show, and the audience wants new, fresh, nasty things to happen.
- The GM won't let you level up while in the maze. A bit meta, but it could work.
---
The graph system I outlined before goes another step further, because it actually introduced multiple ways to get somewhere. And that's when foreknowledge becomes interesting. If the PCs know that their ancestor reached the heart of the maze through the Black Marble Halls and then the White Marble halls, they'll rejoice if they run into the Black Marble Halls. But they might have a note from an enemy who mentions a different route through the Magma Caverns. Now the PCs have a choice; do they try to take the hot or the cold route? Or do they try to discover an entirely new way?
Phussy Owl: how do you describe the connector lines? Is each node meant to represent a single encounter or an area containing multiple encounters?
The classic example is the underground dungeon. In that milieu would a node be a single room? That would make the connectors hallways, secret passages, bridges across chasms, etc.
If however a node is a group of rooms inside a larger facility then I'd think the challenge would be managing the right connector going to the wrong location. If one node of a labyrinth is a collection of 6 rooms with a sewer pipe leading out one end and a flight of stairs heading down from the other but the PCs' skill checks result in a failure, how do you justify them taking the stairs and ending up in the sewer they were trying to avoid?
I usually tell about the place and just like i dont usually describe every door in a City i dont bother with every hallway in a labyrint. The PC sees them the players dosent need the long an tedious tour. To avoid getting lost i use skill checks modified by clever tricks, like marking the Way.
Edit: and i dont have a detailed map. I make a story map, sort of, and leave room for the players to suprise me or hijack the story.
To the original question: I don't. It is typically a waste of time, possibly with the exception of for the mapmaker. I have yet to see a situation where it was fun. That said, the boundary between labyrinths and normal dungeons CAN be a bit difficult to pin down. Jzadirune springs to mind.