The sphinx blocks your path. She speaks in a calm, booming voice that echoes through the dungeon.*
“Altar of the Lupine Lords.Answer: Let’s play something else.
Round she is, yet flat as a board
Altar of the Lupine Lords.
Jewel on black velvet, pearl in the sea
Unchanged but ere changing, eternally.”
*This riddle is the first thing most critics of puzzles in TTRPGs think of. It's from White Plume Mountain.
Not all OSR-style challenges are puzzles and not all puzzles are OSR-style challenges. But the puzzles I like are OSR-style puzzles.
I've talked about running puzzles well, but I've yet to talk about how to make good puzzles. Puzzles in TTRPG games have historically been so bad that I've probably already lost readers with the title of this post. Some people dismiss puzzles as a waste of time. That’s a shame, because in my own humble opinion, puzzles can be just as engaging and even more rewarding than the “OSR-style challenges” that referees already present to players. Puzzles are a lot harder to do well, but they are worth it.
What are OSR style challenges?
- No obvious solution. (Straight combat is always obvious.)
- Many possible solutions.
- Solvable via common sense (as opposed to system mastery).
- No special tools required (no unique spells, no plot McGuffins at the bottom of a dungeon).
- Not solvable by a specific class or ability.
One example from the blog post that sticks in my mind is:
58. Cross a moat filled with crocodiles.
What’s a puzzle?
Let’s tweak the Oxford Languages definition a bit and define a “puzzle” as “a problem designed to test ingenuity or knowledge.”
Whose ingenuity or knowledge?
When talking about TTRPGs we usually mean the player. Some would argue that the player’s character should be the one who is being tested and they should get to roll dice to determine success. To those people, I’d say they were missing the point of puzzles in adventure games, or the point of this blog post at least.
So for the purposes of this discussion, we’ll say that a “puzzle” is "a problem designed to test ingenuity or knowledge of a player who is roleplaying one or more characters in a fictional space". Or… just “player” because I don’t want to write all that.
This is still a very broad definition, but it helps eliminate some quick cases right away.
A locked door and a 1st level adventuring party.
This is NOT an OSR-style challenge and it is NOT a puzzle.
The party could pick the lock, they could force or break the door, or they could open the lock with a key. These are standard solutions that don’t require a lot of thought. Would you laud a player’s decision to force down the door as ingenious? Probably not.
Let’s do better.
A locked IRON door and a 1st level adventuring party. AND it’s common knowledge there’s a Troll around here with a rusty key dangling around their neck.
This IS an OSR-style challenge, but is it a puzzle?
There are many ways to obtain the key (or not) and open the door, and none of them are particularly straightforward. Are there ingenious ways to obtain this key? I think so. The PCs can trick the Troll or do favors for the Troll or wage a faction war against the Troll, etc. Opening the door with the key is the straightforward part. Or the PCs can just ignore all of that and make their own door with time, sweat, and tears.
While this challenge seems like it fits our definition of “puzzle” in that it is designed to test ingenuity, it doesn’t feel particularly “puzzle-like”. There’s planning and cleverness, but there’s no “eureka effect” or “aha! moment”.
The "eureka effect", or “aha! moment”, according to Wikipedia, is “The common human experience of suddenly understanding a previously incomprehensible problem or concept.”
Let’s tweak our definition again. A puzzle is "a test of a player's ingenuity and designed to invoke the experience of suddenly understanding a previously incomprehensible problem or concept."
Let’s make a PUZZLE now.
A locked iron door and a 1st level adventuring party. AND a massive iron bell hangs from the ceiling. FLECKS OF PURPLE SLUDGE blemish the rock walls.
PCs ring the bell. It’s deafeningly, ear-bleedingly loud.
“No, thank you!”, exclaim the PCs, afraid the noise will draw some horrible thing to them. “Tra la la.”
Two rooms later, purple oozes block the path forward. There’s shiny key hanging on a hook way out of reach. It’d be easy to get if it weren't for these darned purple people eaters. The "Molotov lamp oil" strategy is no good. The PCs have only made the oozes angry. These oozes eat fire for breakfast, with a side of blood sausage. The fighter’s armor is disintegrated. The elf has lost their arm. The PCs retreat.
In another room, the purple oozes are absent. Their gooey trails stop and veer off from here. There's a house-sized nest here with some big honkin’ eggs and some shiny treasures. Also, there's a horse-sized, screeching goat-bird with razor talons. The screech upends the PCs and makes their ears bleed.
Damn. Did we just design a puzzle dungeon? Almost.
At this point, there's a bunch of information about the mechanics of the dungeon available to PCs who experiment and interact with their environment. Score.
The PCs might get that eureka effect when they see flecks of purple sludge around the lair of the goat screecher.
Eventually we want them to realize that sound is the purple ooze’s weakness. Ringing the bell or the screech of certain creatures will cause the ooze to become paralyzed and start flipping out while pulsing flecks of purple sludge.
The flecks of purple sludge provide re-contextualization for the players. If we’ve been doing our job right, we’ve been painstakingly pointing it out for the players reminding them it’s there. And when players ask, we tell them when the flecks are NOT there. We want these areas to stick in the player’s minds and seed that aha! moment.
This is pretty good. Lower case “good”.
- Clear end-goal. Get the key to open the door.
- Clear challenge with no obvious solution. Get past the purple oozes.
- Multiple solutions. Kite the oozes away, lug that big-ass bell around and ring it in front of the oozes, kite the goat screecher to the purple oozes.
- Solvable via common sense and not special tools. While the massive bell could be considered a special tool, it’s not required to solve the puzzle. In fact, it’s still a pain in the butt to lug around. I wouldn’t call it a “plot McGuffin”.
- No fancy spell or class ability that’s required to solve the puzzle.
- Seeding in those Aha! moments. We've got a bit of re-contextualization with the purple flecks, but it's easily missed. We probably need some more.
I say this is pretty good because it depends on the PCs making a pretty big leap in logic. With these kind of puzzles, while there’s no requirement of system mastery, there is a requirement of mechanics mastery.
It’s much less of a jump in logic, if we incrementally teach the players the mechanics of the world.
This dungeon reacts to sound. That’s its theme. Let’s dig into a bit more.
I want a cool item to find, ala The Legend of Zelda. The bell is fine, but it’s heavy and bulky. The PCs should be given the option to upgrade given a challenge. How about we throw “Moonfen’s magic tuning fork” in the goat screecher nest? Get it stuck in the PCs' craw and don’t hide the fact that it’s there. It’s another possible aha! moment later when they realize that the theme is sound. The tuning fork is not required, it’s just nice to have and a good reward for ingenuity.
We need more things that react to sound. Let’s sprinkle in some thick crystal that’s painstakingly slow to break with tools, but shatters like the T-1000 when the high pitched sounds start. Actually, let’s make it purple crystal. It’s the crusty hardened shell of purple ooze mucus. And let’s get this crap aaaaaaaaall over some shiny treasure. We’re going to block archways with it and cover acid melted, mummified corpses with it. This stuff is gonna be all over the place and block some cool things that PCs want. We’re also going to put some bits of broken purple crystal interwoven in the goat screecher nest. It’s gotta make due with whatever supplies it has, and there will be a lot of this broken stuff wherever it screeches.
Not everything has to be related to the theme. In fact, it’s better to have some stuff that’s not. PCs relying on the bell too much? Let’s get some rust monsters in here. The bell was previously inaccessible to them because it was tied high to the ceiling, but now that the PCs have been carrying it around, the scent of that sweet sweet iron is permeating through the dungeon and whetting the ole’ rust monster whistle.
Oh, hang on. Now I want a whistle. The sound can't be heard by human ears. Maybe elf ears… but definitely heard by the goat screecher. Heard from anywhere in the dungeon. And the goat screecher doesn’t like it one bit.
If you’re feeling especially bold, throw in something that subverts the PCs expectations. Lets put in a room with doors that react to sound, but not in a way the PCs have seen before. They need to be QUIET to open these doors. I’m talking, putting-out-noisy-burning-torches quiet. Drop ample hints and keep it optional. You’re players might get another aha! moment out of it.
Now this is an uppercase “GOOD” puzzle. It’s a puzzle dungeon!
- Teaches the mechanics of the world. The PC’s have ample opportunity AND motivation to learn the mechanics of the dungeon. The referee has lots of options for further puzzles that build on each other. Smaller challenges that feed into bigger and more complex ones that rely on planning.
- Seeding those Aha! moments with THEME. This puppy is screaming “SOUND!” at the PCs. We've got plenty for them to experiment and interact with and some bait for them to think about how to get. Perfect environment for the eureka effect.
Can we make it a GREAT puzzle?
Yes, but… this is the toughest pill to swallow and I’ve saved it until the very end when I’ve (hopefully) got you all excited about making OSR-Style puzzles. The hardest part about making puzzles fun and exciting has to do with the fact that you are designing them purely in the theoretical. You won’t know if a puzzle is going to land until you run it. You need to test it out and be willing to change things that don’t work and add more to make the experience smoother.
An obvious blocker here is if you’re only playing with one group. You can’t get the eureka effect yourself and you can’t get the same group of players to experience the same eureka effect twice.
So start small, play with lots of people, build, tweak, change, obsess. Or not. You’ll still have a GOOD puzzle. Which is good! I can’t stress enough how much better your puzzles will be than 90% of the published material out there. Don’t be afraid to unleash your hard work for your players. Be upfront about what it is and why you’re excited about it, but make it optional. You’re players don’t need to engage with or solve your puzzles. Allow them to break your puzzles as this is the BEST way to improve them. You can do this. Really!
Further reading... or viewing, rather.
Now we are getting into advanced territory and puzzle theory. There are systematic ways to structure puzzle dungeons through teachable loops and category buckets for puzzles.
If you want me to write more on these techniques, I can. Leave a comment. Or you could just listen to people much wiser than me about the topic.
Game Master's Tool Kit on youtube has a series of videos breaking down why The Legend of Zelda's puzzle dungeons are so good. They talk about how they are constructed and why they work.
There's also a great video about what makes puzzles good in general.
I was also made aware while writing this blog post that this guide exists, written by Sersa Victory. It does a great job visually explaining classic puzzle loops while focusing on TTRPG play.
Are there adventure modules already like this out there?
And I'd be remiss if I didn't give recognition for one of the best puzzle dungeons in the scene. Fabien's Atelier. It's one of the adventures in Wyvern Songs, by Brad Kerr.
Now for my hopes and dreams.
I hope I’ve changed the minds of some of the more stalwart players and refs that are opposed to puzzles in their games entirely. And I hope that this post is able to act as a bit of a meandering guide for referees creating their own puzzles. And at the very least, I hope this post encourages at least one referee to think twice before including an impassable block at the beginning of a dungeon that disappears after the PCs answer a tonally dissonant riddle with the words: “the moon”.
If you prescribe to these notions or if I've made a believer out of you, leave a comment or something. Tell me your favorite puzzle. If you still think puzzles suck, leave a comment or something. Tell me what you hate and I'll try to change your mind again.
P.S. I don't hate white plume mountain. That module is awesome. There are some amazing challenges right next to some pretty bad ones.