Using Boring Meetings to Design Dungeons

The modern world is filled with a certain number of boring, pointless meetings. Maybe the meeting itself is pointless. Or maybe you are simply an ancillary attendee who should have been left off the invitation. For whatever reason, you are stuck here until the meeting ends. With nothing else to keep it busy, your mind wanders back to the dungeon. 

Every time a new participant in the boring meeting talks, try to write down the first remotely interesting word you hear them say that is either a noun, an adjective, or a verb. If they’re talking fast, you might only catch one or two of those words. That’s OK. These are loose guidelines, not strict rules. Keep writing down words until you have at least one verb, one noun, and one adjective. Cluster these in groups until you have a handful of them.

Nouns Are Rooms

Each noun defines the purpose or status of a room in the dungeon that you are creating. Sometimes this will be very literal, like the word “store” suggesting a storehouse. Other times the room you create will only have the vaguest trace of the origin word. Some words will not work at all, but try doing some free association before you give up on a noun that doesn’t have any dungeon energy. You may find it takes you to some unexpected places.

Adjectives Are Contents

Adjectives elaborate on what is in the room. This can take many forms, but the following are good places to start.

Aesthetics: What does the room look like? Smell like? Sound like? 

Purpose: The noun may imply what the room is for, but an adjective that pushes back against that implication may suggest a change in purpose. Integral to many classic dungeons is the idea that a room’s purpose has changed over time.

Occupants: Adjectives can strongly suggest who or what uses a room or has been there recently.

Verbs Are Current Events 

Verbs can suggest monsters, NPCs, and other dungeon activity. They can also suggest hazards, environmental effects, and weather. The random encounters table is a good place to start thinking about verbs. What is happening right now? What happened recently? What will happen soon? If the room is not an empty room, the verb may be the best clue toward what a monster is doing, how a trap threatens interlopers, or how a trick or special feature presents itself to the explorer.

An Example: The Questioning Device

I listened and wrote down the noun "question," the adjective "technological," and the verb "counting."

“Question” could mean many things, from a scrying pool to a riddle. I will make our “question” room an interrogation room. This is a pretty literal interpretation and gives us a grounded place to start.

“Technological” could go in a few directions. The torture machine from The Princess Bride, for example. Perhaps there is a techno-magical machine in this space. Whatever its original purpose, the current dungeon occupants use it to interrogate prisoners.

“Counting” is a great verb because it suggests both subject and object – someone is counting and someone or something is being counted. Perhaps one of a number of prisoners has escaped? A headcount is happening, or has just happened, and the captors have discovered that someone ins missing. They are now using the Questioning Device to try to force the remaining prisoners to tell them where the escapee went.

And so we have a fully formed room. 

Pay Attention, Class

Am I suggesting you be lazy? Rude? Disrespectful to the organizer of your boring meeting? Well, yes and no.

Yes, it is true that I am suggesting you slack off a bit. But I did preface it by saying I was talking about meetings that were unimportant, unnecessary, or overly broad in terms of invitees.

And for what it’s worth, I think this game is a way of genuinely paying attention. The worst sort of inattention is the full-on daydream, where you are thinking about a dungeon that has nothing to do with the meeting. At least in this model, you are paying a minimum level of attention in order to catch those prompt words. Your brain may subconsciously catch more detail than you expect, just because you have given it an ulterior (and more interesting) reason to care.

Dungeon Rooms are Nouns, Dungeon Corridors are Verbs

I sometimes think about dungeon design like this: Rooms are for nouns, corridors are for verbs. 

Explaining what is going on there requires some background.

Why are corridors important at all? Many modern dungeons eschew them. I think part of the challenge is that writers publishing dungeons often want to fit an entire map floor on a single page of a book or PDF. Many otherwise very good dungeons feature cramped, close-together rooms because of product design, not dungeon design. 

But those corridors serve important purposes, many of which are covered well here. So dungeons should have more corridors. How should we design them differently from rooms? 

Dungeon Rooms Are Nouns

The room is defined first and foremost by its contents. The classic dungeon stocking options of “empty/monster/treasure/trap/special” are essentially noun-focused. Something may be happening in the room, but any action is derived from the contents. If you are GMing a minimally keyed dungeon, and the PCs approach a room that simply says “guard post: barricade, 3 goblins” you are going to intuit the action downstream of those nouns. Are the goblins alertly watching out for intruders? Or arguing with each other? Asleep? You may infer the answer, you may roll for it, but it is downstream of the contents. Rooms are noun-forward.

Dungeon Corridors Are Verbs

Corridors are typically not defined by their contents. I’m excluding a “great hall” or “foyer” here. We are talking about corridors that are exclusively transitional spaces between rooms. They are not defined by their contents but instead by action, by what is going on within them.

Wandering monster or random event tables are the classic way of adding verbs to corridors. I strongly agree with Fae Errant's linked post above that there’s typically no need to roll for wandering monsters in rooms; those rolls are doing the most work in corridors. Corridors ensure that there is always a cost to exploring. 

Corridors ask: 

• What are the players doing? Searching, sneaking, pursuing, fleeing? Consider the party's pace.
• What are the monsters doing? Consider using the nested table style of Hot Springs Island, or the supplemental tables in The Monster Overhaul.
• What is the dungeon doing? Depending on the degree of motive agency we assign to the space itself, hazards and obstacles can be thought of as verbs the dungeon itself apples to corridors.
• Has time passed? Have torches guttered out? Spells expired? Corridors ask these questions.
• What has already happened during the delve? Is anything recurring? Is it time for any consequences of prior action to make themselves known?
• What has changed since the PCs last traversed a corridor? What is changing right now?

Touch Grass

The same ideas can apply outside the dungeon. For example, I usually use pointcrawls for outdoor exploration. Each location is very noun-heavy, but the paths between points are the places for verbs. 

Dungeon Thresholds Are Adjectives

This is less essential than the noun/verb distinction, but if you want to take it a step further, consider adjectives as thresholds. Adjectives are relayed to the PCs when they first enter a room, then give way to the nouns as the room is explored in earnest. The adjectives serve to mark the transitional space. How is this room distinct from the corridor you are exiting? The adjectives are often most prominent when the PCs are still deciding whether or not to enter a space.

Adjectives can answer questions like the following: 

• What hidden thing in the room should the PCs be looking for?
• What was the room recently used for? 
• What was the room’s original purpose? 
• What is most immediately noticeable sight, sound, and scent?

Campaign Pitch: Revealing the Secrets of the Mystery in the Most Literal Way

You’re planning a game that revolves around a central mystery. It could be the location of the lost treasure, or the motives of the supernatural creatures hidden from mundane society, or the agenda of the aliens in first contact. Try approaching it like this. 

Create a document that lays out the answer to the whole mystery. It may be similar to your own GM notes, but more elaborate. It can include pictures and diagrams. Think of it like a spoiler-filled after-action report. It can be written in-voice, in diegetic language, but that’s not mandatory. It’s particularly easy to do diegeticly in a modern game like Delta Green, where solving a mystery could be expected to create a paper trail, but it could work in any setting.

Give the players access to the document… but hide literally everything with black redaction bars. This is obviously easiest to do electronically, although you could print out copies for at-the-table reference. When the game starts, the players can tell how long the document is, where the headers are, and how many images and diagrams there are, but that’s about it. They can’t see any of the actual text. Be sure to give them view-and-comment privileges, but not the ability to edit.

As the game proceeds, remove the redaction bars from anything that has been discovered in-game. This serves a few purposes. It formally acknowledges that they have found an undisputed truth of the mystery. It provides a common reference for important information that has already been established. And (for players who care about the distinction) it provides fixed information that was established before the GM knew what the players were going to do; in other words, it “proves” that the answer to the mystery is a fixed target, and not something that the GM is making up in response to player action.

But this particular application of redaction does something else. It lets the players see the shape and size of what they don't know. If the entire section under the “Abandoned Mine” heading is revealed, the players can reasonably assume there’s nothing else they need to know about the Abandoned Mine. But if there’s a single sentence still blacked out… well, that should make the players curious. If the players find a map or a photograph, and the GM un-redacts the map in the document, but not the text around it, that raises some interesting questions, right?

They could infer a lot from the organization and the “known unknowns” of what they cannot yet see. Imagine if a section of the revealed text included a hyperlink to an as-yet unrevealed section. Or if one section said “DANGER: Do not engage with [person/entity/object/location] without reviewing [other section that is still redacted]. 

This is probably enough to center the game’s mystery, but there’s room to make it more complex. The GM could tie revealing text to a mechanic, like experience and leveling up. Or give the players an in-game currency to “pay” to reveal certain parts of the text. It could even be a minigame where the players could choose a word and reveal every instance of that word throughout the document, creating a kind of game show word puzzle where they are rewarded for predicting what words might show up a lot, or in particularly important places.

Depending on the desired level of diegesis, it’s up to the group to determine if the characters literally have the document in question or not. At its most literal, it could be classified files or deciphered lore. For a less direct treatment, one could imagine the document as a historical account of the information associated with the adventure. But it’s also OK if it is purely a meta construct. As long as it piques the characters interest in the mystery, it’s doing its job. 

Can AI Turn an Arden Vul Room Key Into Table-Ready Notes?

Last week: Use AI for Routine TTRPG Prep, Not for Creative Work

For this experiment, I’m going to use room Arden Vul’s room 4-47 (“Overseer”) as an example. You can find this entry on page 272 of volume 1. The following description includes minor spoilers about an NPC and their faction, but I wanted to use it because it is such a good example of the issues described in the previous post. I don’t think this gives away too much in the grand scheme of Arden Vul; but if you have plans to play in this campaign in the future, and are sensitive to spoilers, you can skip down to the Conclusions heading for the summary.

Here’s the text as it appears in the Arden Vul book:

4-47: Overseer

Belisarius the Overseer, the second most powerful figure in the cult of Set, lives and works in this large room. The overseer is the ranking ‘secular’ figure of the cult, and is in charge of all guardsmen, military expeditions, security, patrols, and so forth. For all that, Belisarius is a devout, even fanatical, follower of Set, and an adept politician. He is also a misogynist and dislikes High Priestess Stephania intensely. As a result he is constantly plotting her demise, and hopes to replace her with the third priest, Lacedaion. Stephania is aware of these machinations and has taken precautions; she has not acted against Belisarius yet solely because he is so good at what he does. Despite his prejudices, Belisarius would never act to undermine the cult as a whole; only if he were convinced of the motives of outsiders would he consider employing them against Stephania.

The chamber is divided into a public and private area; the latter is to the east, and is shielded by some painted wooden screens. The public area has a large tactical table on which lie (incomplete) maps of the northern and southern caves of Level 4 as well as one ancient map detailing portions of the lower levels (GM’s choice). Also in the public area are several writing desks and chairs containing roster lists, personnel reports, logistical reports, and so forth. Two aides-de-camp (from the Alpha Sept) are always present here. At the north wall, next to the door to the caves, stands a large cabinet in which may be found a sizable number of curios looted from the halls as well as 12 bound codices (on theological, military, and motivational subjects). The private area contains a simple bed, an armoire, another writing desk, a weapon and armor rack, a practice dummy, two cots for the aides, a set of four easy chairs around a coffee table, and a cabinet containing 22 bottles of excellent brandy. In addition to his administrative skills, Belisarius is a formidable warrior.

The north door leads to a narrow tunnel that provides Belisarius with private access to the cavern at 4-66 via a one-way door. He doesn’t use this door very often, and the corridor is thus thick with dust.

There is a 50% chance that Belisarius is present in his quarters. The rest of the time he is elsewhere in the complex, typically in 4-2, 4-3, 4-7, 4-43, 4-51, 4-53, or 4-61. He has a 25% chance to be in 4-48 whenever the second mystery is performed [sic]

Secret Door (east): This door is normal to find and locked, and leads to a set of secret  interrogation chambers. Only the overseer and his aides-de-camp know of the existence of the door and the interrogation chambers. The key to the door is hidden in Belisarius’s liquor cabinet.

Inhabitants:

 1. Belisarius the Overseer, 7th-level male Archontean fighter: SZ M; AL LE; AC 2; MV 120’; HD 7; HP 70; #AT 3/2; Dmg 3-8 (javelin +2) or 4-11 (long sword +2). Abilities: STR 17, INT 13, WIS 16, DEX 11, CON 16, CHA 15. Possessions: chain mail +3; returning javelin +2 (see new magic items), long sword +2; potion of extra healing (x2); potion of fly; and a key-ring with keys to SL8-1 and 4-106. 

2. 2 aides-de-camp, 3rd-level female Archontean fighters: SZ M; AL NE; AC 5; MV 120’; HD 3; HP 25, 22; #AT 1; Dmg 1-8 (mace), 1-6 (spear) or 1-4 (light crossbow). Possessions: Set armor (see Arden Vul items); spear; mace, and light crossbow.

Stop and think for a moment about how this information is organized. It’s not random. The most important information is up top (who Belisarius is and what his motivations are). It is followed by the physical contents of the room, then more conditionally relevant information, like how the room connects to other spaces. It ends with the stats for the potential occupants. For purposes of the DM reading the text in advance of running the dungeon, this is basically fine.

But for use at the table? This could be much better.

The first paragraph is all about Belisarius’s motives and (implicitly) how he will act and react if the PCs become involved in the politics of the cult of Set. This is the most important information in a narrative sense, but the least important information in the context of dungeon exploration. This paragraph is actively hampering the product’s usefulness at the table, where speed of adjudication would be the DM’s priority. If the PCs are engaging diplomatically with members of the cult of Set, the DM will probably need this information long before the characters wander into Belisarius’s bedroom.

The second paragraph begins to lay out the practical details of the room. Halfway through this paragraph, we learn that “two aides-de-camp are always present here.” It’s easy to imagine a DM describing this room, with its tables and maps, and then suddenly, in a surprised voice, saying “Oh! Uh, and… there are two fighters here?” At minimum, some kind of bolding should indicate the presence of monsters and NPCs. The paragraph goes on to list more contents of the room, and the final sentence states that “In addition to his administrative skills, Belisarius is a formidable warrior.” This would be more at home in the first paragraph explaining who he is, or omitted entirely (his formidability in combat is clearly telegraphed by his stat block).

The third paragraph addresses the northern exit and is almost entirely redundant with information that can be intuited from looking at the map (another common point of complaint in Ten Foot Pole reviews). The fourth paragraph provides context for the secret door in the east wall. The key that unlocks this secret door is first mentioned here, not earlier, when its actual hiding place is described. So a DM could easily adjudicate the PCs searching the liquor cabinet without realizing they should find the key that is hidden there.

So, how good of a job can AI do unpacking this information? Let’s find out.

Here’s the prompt I gave ChatGPT: “Can you improve the organization and formatting of the following text? This is a room description for a roleplaying game, but the text needs to be reformatted so the person running the game can more quickly find the most important information. Details about people, creatures, or other things adventurers might interact with when exploring this space are most important and should be emphasized. Use bolding, bullets, and any other formatting tools you want to make the text easier to reference on the fly.”

The output was as follows: 

Key NPCs and Creatures

1. Belisarius the Overseer

• Role: Second most powerful figure in the cult of Set; Overseer of guardsmen, military expeditions, and security.
• Personality: Devout, fanatical follower of Set; adept politician; misogynist; intensely dislikes High Priestess Stephania and plots her demise but remains loyal to the cult as a whole.
• Combat Stats:
• Class/Level: 7th-level male Archontean fighter
• Alignment: Lawful Evil (LE)
• Armor Class (AC): 2
• Movement (MV): 120’
• Hit Dice (HD): 7
• Hit Points (HP): 70
• Attacks (#AT): 3/2
• Damage (Dmg): 3-8 (javelin +2) or 4-11 (long sword +2)
• Abilities: STR 17, INT 13, WIS 16, DEX 11, CON 16, CHA 15
• Possessions: Chain mail +3, returning javelin +2, long sword +2, two potions of extra healing, potion of fly, key-ring (keys to SL8-1 and 4-106).

2. Aides-de-Camp (2)

• Role: Assist Belisarius with administrative tasks; always present in the public area.
• Combat Stats:
• Class/Level: 3rd-level female Archontean fighters
• Alignment: Neutral Evil (NE)
• Armor Class (AC): 5
• Movement (MV): 120’
• Hit Dice (HD): 3
• Hit Points (HP): 25, 22
• Attacks (#AT): 1
• Damage (Dmg): 1-8 (mace), 1-6 (spear), or 1-4 (light crossbow)
• Possessions: Set armor (unique to Arden Vul), spear, mace, light crossbow.

 

Room Layout and Features

Public Area

• Tactical Table:
• Contains incomplete maps of the northern and southern caves of Level 4.
• Includes an ancient map detailing portions of the lower levels (GM’s choice).
• Writing Desks:
• Hold roster lists, personnel reports, logistical data, and similar documents.
• Cabinet (North Wall):
• Contains curios looted from the halls and 12 bound codices (topics: theological, military, motivational).

Private Area

• Sleeping Quarters:
• Simple bed, armoire, writing desk, and cots for the aides.
• Recreational Area:
• Four easy chairs, coffee table, and a cabinet with 22 bottles of excellent brandy.
• Training Equipment:
• Weapon and armor rack, practice dummy.

Secret Features:

• Secret Door (East):
• Leads to interrogation chambers.
• Lock Details: Normal to find and locked.
• Key Location: Hidden in Belisarius’s liquor cabinet.
• Knowledge of this door’s existence is limited to Belisarius and his aides.

Exits

• North Door:
• Leads to a dusty, narrow tunnel providing private access to cavern 4-66 (via one-way door).
• Rarely used; thick with dust.

NPC Presence and Movement

Belisarius:

• Chance of Presence: 50% in his quarters.
• Typical Locations: 4-2, 4-3, 4-7, 4-43, 4-51, 4-53, or 4-61.
• Special Event: 25% chance to be in 4-48 during the second mystery.

GM Notes

• Belisarius's Plotting: Actively plots against High Priestess Stephania, intending to replace her with Lacedaion.
• Stephania’s Awareness: Aware of his schemes but tolerates him due to his efficiency.
• Interrogation Chambers: These are secret and only accessible through the locked door in the private area.
• Maps and Codices: Could provide adventurers with valuable intelligence if accessed.

This is already significantly improved. It’s overkill on the bullets, particularly in giving many bullets only a single sub-bullet (from an information design perspective, a single sub-bullet should never  rarely appear alone, as it defeats the whole purpose of bullets, i.e., parsing out a list in a scannable format). But everything is much more clear than the original text, and the machine didn’t even need much coaching to reorganize the information in order of occupants → features → exits → miscellaneous.

I asked it to make a few changes, moving the NPCs presence to the top; merging the secret door info with the exit info; and condensing some of the bullets to shorten the overall vertical “height” of the entry. It produced this:

Key NPCs and Creatures

1. Belisarius the Overseer

• Role: Second most powerful figure in the cult of Set; Overseer of guardsmen, military expeditions, and security.
• Personality: Devout, fanatical follower of Set; adept politician; misogynist; intensely dislikes High Priestess Stephania and plots her demise but remains loyal to the cult as a whole.
• Chance of Presence: 50% in his quarters. Otherwise, typically in 4-2, 4-3, 4-7, 4-43, 4-51, 4-53, or 4-61. Special Event: 25% chance to be in 4-48 during the second mystery.
• Combat Stats:
• Class/Level: 7th-level male Archontean fighter; Alignment: Lawful Evil (LE); Armor Class (AC): 2
• Movement (MV): 120’; Hit Dice (HD): 7; Hit Points (HP): 70
• Attacks (#AT): 3/2; Damage (Dmg): 3-8 (javelin +2) or 4-11 (long sword +2)
• Abilities: STR 17, INT 13, WIS 16, DEX 11, CON 16, CHA 15
• Possessions: Chain mail +3, returning javelin +2, long sword +2, two potions of extra healing, potion of fly, key-ring (keys to SL8-1 and 4-106).

2. Aides-de-Camp (2)

• Role: Assist Belisarius with administrative tasks; always present in the public area.
• Combat Stats:
• Class/Level: 3rd-level female Archontean fighters; Alignment: Neutral Evil (NE)
• Armor Class (AC): 5; Movement (MV): 120’; Hit Dice (HD): 3; Hit Points (HP): 25, 22
• Attacks (#AT): 1; Damage (Dmg): 1-8 (mace), 1-6 (spear), or 1-4 (light crossbow)
• Possessions: Set armor (unique to Arden Vul), spear, mace, light crossbow.

Room Layout and Features

• Public Area:
• Tactical Table: Incomplete maps of the northern/southern caves of Level 4; ancient map detailing portions of lower levels (GM’s choice).
• Writing Desks: Roster lists, personnel reports, logistical data.
• Cabinet (North Wall): Contains curios looted from the halls and 12 bound codices (topics: theological, military, motivational).
• Private Area:
• Sleeping Quarters: Simple bed, armoire, writing desk, and cots for the aides.
• Recreational Area: Four easy chairs, coffee table, and a cabinet with 22 bottles of excellent brandy.
• Training Equipment: Weapon and armor rack, practice dummy.

Exits

• North Door:
• Leads to a dusty, narrow tunnel providing private access to cavern 4-66 (via one-way door). Rarely used; thick with dust.
• Secret Door (East):
• Leads to interrogation chambers. Lock is normal to find and locked.
• Key Location: Hidden in Belisarius’s liquor cabinet.
• Only Belisarius and his aides know of its existence

GM Notes

• Belisarius's Plotting: Actively plots against High Priestess Stephania, intending to replace her with Lacedaion.
• Stephania’s Awareness: Aware of his schemes but tolerates him due to his efficiency.
• Interrogation Chambers: These are secret and only accessible through the locked door in the private area.
• Maps and Codices: Could provide adventurers with valuable intelligence if accessed.

This is pretty close to how I would organize this information if I had converted this room key into table notes by hand.

Finally, I asked it to add a table of contents at the top with a hyperlink to each section, and it provided blue-colored text that looked right… but didn’t appear to form an actual hyperlink, and didn’t work when I copied it into other programs. So there are limits to what it can do, but I think some combination of better prompting and using a more advanced AI could get there.

Conclusions

This is really only scratching the surface. If you gave an AI the entire text of an adventure you were going to run, you could ask it to create cross-references and summaries. It could generate random encounter tables based on dungeon denizens. It could provide a treasure index. 

Yes, all of these tasks are things the GM can do themselves. But they are time-consuming. AI helps in the same way that conventional word processing software saves time, which can be reallocated to the more creative parts of session prep.

There are some issues to navigate before using this more broadly. Some AIs might introduce external information, hallucinate, or garble details (although this is less of an issue when asking it to rearrange or summarize content, versus creating new content). And we should be mindful of any stipulations that creators put on the use of their work vis-a-vis AI, particularly concerns over training data. ChatGPT does not use submitted content to train its model, but other AIs might. If a creator includes stipulations about using their content in this way, err on the side of caution and keep it away from the AI.

But GMs should absolutely continue to explore AI as a prep assistant. The world has a finite number of highlighters, and your wrist is probably going to give out before you get halfway through marking up a beast like Arden Vul. Don't let something as simple as organization and formatting stand between you and running an epic megadungeon like this.

Use AI for Routine TTRPG Prep, Not for Creative Work

The Halls of Arden Vul by Richard Barton is a truly massive megadungeon. It’s a love letter to AD&D and the classic megadungeons of the 1980s.

The good part of the classic feel is obvious. Listening to the 3d6 DTL podcast’s long-running Arden Vul campaign, it’s impressive to see the density of factions, the depth of the dungeon’s history, and the opportunities for interactivity. I would love to run it someday. 

The bad part comes down to formatting and at-the-table ease of use. I wouldn’t say it doesn't do anything to incorporate the advances TTRPG usability over the last 40 years; it includes expansive overviews and aggressive cross-referencing, an improvement from the “figure it out yourself” opacity of a lot of ‘80s D&D products. 

But it doesn’t go nearly far enough. The text is dense, written for casual reading rather than for quick reference. Ancient history is intermingled with recent events. There’s little use of bullets, bolding, or sidebars. Even little things reflect a retrograde formatting mentality; listing damage as “2-7” rather than “1d6+1” is a small detail, but it is a pointlessly archaic Gygaxism that can trip people up; particularly new DMs, or those who don’t easily intuit die increments from numerical ranges. Compare this to a modern product, like Necrotic Gnome’s OSE releases, with their signature house style, carefully calibrated for use at the table. Arden Vul is old school not just in play culture, but also in formatting and presentation.

Don’t take my word for it. Read Bryce’s review on Tenfootpole. He obviously has high praise for the amount of evocative detail and interactivity on display in Arden Vul. But his reviews also place a premium on ease of use at the table, and many of his criticisms of Arden Vul land here. See the “Jhentis the Ghoul” section of the linked review for an apt summary of the problem.

Browsing my own copy of Arden Vul, I have to agree with him. It is frustrating that such a massive, fascinating work should be held back by mere formatting and organization issues. Implicitly, the DM running this is expected to make up the difference. As Bryce puts it in the review: “Highlighters out! Actually, better buy a gross of them, you’ve got 1122 pages to read, absorb, and highlight.”

But it’s 2025. We aren’t limited to highlighters, right? Can’t we use… technology? Can we get the AI to do it for us?

On the Use and Misuse of Generative AI in TTRPGs

Generative AI is a hot topic in many spaces, including indie creative scenes like TTRPGs. While some indie and OSR creators have dabbled in using generative AI, there's a lot of ire directed at these tools as well, particularly from artists who object to Midjourney and similar image-generation software.

I’m sympathetic to concerns about AI content replacing human-made content. In the past, I used AI images on this blog, just to add some color to my posts. But recently, I’ve been losing interest in those image-generation tools. While my use of them did not displace any work that would have been commissioned to a human artist, I’ve nonetheless found that as AI-generated images have gotten “better,” they’ve lost some of the weirdness they had just a few years ago, and are turning into professional-but-bland tools, better suited for workplace PowerPoint presentations than TTRPG content. There’s much more to say on this topic, but that’s a whole post of its own, for another time.

AI image generation in action. Technically, yes, this delivers on the prompt request of "robot rearranging things in a dungeon room." But everything in it is either sharp-but-derivative (the very generic robot); blurred to the point of abstraction (the figures, the chest (?) on the left); or failing on a basic level to model a coherent 3D space (the archway, the candles, the stairs, the robot's lower body). Image generation like this is not a good replacement for human-made creative work. 

But this skepticism of generative AI as a replacement for creative work shouldn’t blind us to the usefulness of AI as an assistant for routine tasks. AI tools in their current form are best at handling the least-interesting and dullest parts of the process, freeing up creators to focus on the really creative stuff. And one of the places AI can help the most is in session prep.

Next Week: Can AI Turn an Arden Vul Room Key Into Table-Ready Notes?

In Praise of Location-Based Scenario Design

Two of the recent games I have run are “travel” games. In one, the premise is that the characters wake up on a train without their memories. The train moves from place to place, and as they explore each stop on the train’s journey, they uncover clues related to their predicament. In the other game, the players are afflicted with a mysterious magical condition that causes them to Plane Shift randomly through the multiverse, encountering dangerous and unexpected situations.

I’ve enjoyed running both of these games, and I don’t regret pitching them to the players… but they had the unexpected side-effect of making me miss location-based play.

You won’t typically see it printed on the label, but “location-based play” is important to old-school D&D and similar games. The dungeon is not just an abstraction, or the place where the monsters are, but also a literal constraint around the possible actions the characters can take. The location where the adventure takes place creates time pressure, facilitates antagonist action, and intrinsically offers a series of risks and rewards.

The traveling games I ran were more scene-based. The scope of action is less constrained, if at all. Pressure and conflict have to be induced more deliberately.

Running these scene-based games made me appreciate how much location-based play does some of the work on the GM’s behalf. Listening to 3d6 Down the Line’s excellent Adren Vul podcast also drove the point home. A fixed location like a megadungeon organically induces factions, monsters, NPCs, traps, hazards, treasure, and other “game objects” to bounce off each other. 

One of the best aspects of location-based play is that this occurs organically. The GM barely needs to think about it; these interactions intuitively descend from the character of the location itself.  

A scene-based game can and often should re-introduce known concepts periodically, yes. But it takes more work. How did they get there? Why have they shown up again in such a different context? Embracing serendipity as a law of the universe can help, but it will also tax verisimilitude more heavily than simple physical proximity.

A rich location can create its own intrigue and adventure -- illustration by Artur Skizhali-Veys

The difference between location-based and scene-based scenarios may seem obvious, but it’s quite common to see people mix them up. Take, for example, the five-room dungeon concept, created by Johnn Four. The five-room dungeon idea is a preparation shortcut for a TTRPG session. At a glance, “five-room dungeon” sounds like a location-based style of play; it’s a way to make a dungeon, right?

Not really; reading in more detail, it’s clear this is actually a scene-based approach. Four cites Joseph Campbell of “Hero With a Thousand Faces” fame and says “It's the story framework that matters most for great gaming.” That statement likely sounds fine to trad or modern gamers, but is antithetical to OSR gamers, i.e., those who most promote location-based situations. 

Four seems to recognize that the name is a misnomer, clarifying that it is not limited to fantasy games. To the degree five-room dungeons involve locations, they are essentially a way of building what old-school games would call a “lair”; a small dangerous place, perhaps waiting to be discovered while exploring a hex. The lair may have one or more monsters, and the possibility of treasure, hazards, and challenges. It’s probably enough to fill a single session of play. But it fundamentally lacks the promise of exploration that characterizes dungeons, and it is expressly linear. 

There’s enough confusion out there that I saw someone in one instance trying to create a megadungeon by clustering five-room dungeons together! Could that work? Well, maybe. But I think it would have to be done very carefully and deliberately, as it’s a matter of combining two scenario design techniques from diametrically different schools of play.

Feels Within Wheels: An NPC Emotion Generator

I am occasionally guilty of watching helplessly as the themes of my blog posts and the focus of my game prep diverge and wander off in different directions, never again to meet. But not this time!

Here’s an old post that I actually use: Feels Within Wheels. I took an “emotion wheel” as a prompt for NPC behavior.

Oh Lemur of Serenity, who watches over us all, please grant me the peace I seek. Gif source here.

I’ve been using this in my games to good effect, but the table is a little awkward to use directly as a die-rolling prompt. The middle and outer rings of the wheel feature irregular numbers of entries, and some of those numbers don’t correspond to die sizes. It’s always possible to scale them up to a die size (e.g., a choice between five options can be determined by a d10 roll halved), but it's extra brainwork, and in the middle of a session when characters are running around with their hair on fire (figuratively or literally), anything that streamlines decision points is helpful. So let’s make another random Perchance generator.

https://perchance.org/feelswithinwheels

While populating this generator, I noticed some odd details in the wheel I used in my prior post. Why does “dismayed” show up twice on the outer wheel? Is there a meaningful difference between being astounded and astonished? Why is “illustrious” on the table? If I asked someone how they were doing on a given day, and they told me they were feeling illustrious, I would suspect they had been replaced by an alien or an artificial intelligence.

We could fix those things… but ironing out small details isn’t actually that important. A tool like this can be a fuzzy tool. It provides some basic inspiration, but it doesn’t need to be tightly wound or perfectly edited. Some rough edges are OK. Enjoy the generator, and have an illustrious day.

A Random Generator Is Worth a Thousand Hours of Prep

One of the games I ran this year involved adventurers hopping unexpectedly from plane to plane. While prepping for the game, it quickly became obvious that this was not something that I could prep in the same way I would prep a megadungeon or a scene-based mystery.

So how do we produce interesting planar content without prepping dozens of worlds in exhaustive detail? How do we make sure characters “get” each plane they visit? Especially when they’re not spending much time on each one?

Planes of existence in a fantasy world are exaggerations of the “natural” world (the prime material). They are defined by their defiance of rules or expectations we have about the regular world. To boil it down to a very simplified binary, it helps to think about the two sides of what makes a plane different from the normal world.

Concepts that epitomize the plane. Something inherent and fundamental to the place that defines it. It should usually be immediately obvious and prevalent throughout the plane.

Concepts that are antithetical to the plane. Something that is against the plane’s nature. It should either be prominent for this reason, or conspicuous in its absence.

It’s important to think about both aspects. It’s all well and good to say the elemental plane of fire is epitomized by fire. It’s hot. Fires raging, smoke, lava, and so forth. But that’s not enough. Think about near-absolute absence of water. What does the air feel like as a consequence of that? What is the weather like? Are native creatures violently allergic to water? Or is it a precious treasure to them? Possibly some of both, living side by side?

Let’s put this in motion with some prompt tables. 

What are the local landmarks? What captures the eyes of the visitor?

Epitome

• A castle, city, or other built structure. It’s composed of the essence of the plane, or its structure and function is defined by the presence and abundance of that essence.
• A mountain, body of water, or other superficially “natural” feature. It behaves in a strange or exotic way that reflects the essence of the plane.
• A vortex, portal, or other magical, extraplanar juncture or aperture. Its presence indicates the permeability of the plane, as well as its adjacency to affiliated planes. 
• The site of disaster or change characteristic of the creative or destructive forces of the plane.
• Valuables that are rare on the prime material plane are abundant here (e.g., valuable gems on the elemental plane of earth). 

Antithesis

• The absence of something we take for granted in the normal world (e.g., no fire on a frozen or entropic world).
• A concentration of a rare resource, whether precious or merely exotic (e.g., an oasis on a desert plane).
• A structure, object, or feature that was native to another plane, but was moved to this plane, intentionally or accidentally. Its original nature has been warped or changed by this plane.
• Something removed or exploited and taken elsewhere, leaving tangible absence in its place.
• Something antithetical to this plane, but reshaped by the epitomizing forces here.

What is going on there? How is the situation ripe for adventure?

Epitome 

• The plane’s essence is difficult to understand or interact with, or otherwise defies material plane logic. 
• An event (natural or social) is occurring that restricts or slows visitors’ ability to travel and explore.
• A power from a sympathetic or aligned plane is trying to influence, ally with, or absorb the plane.
• Political, commercial, or social activity focuses on a commodity or treasure that can only be grown, made, or refined in the unique environment of the plane. 
• A gift or creation of the ruler or power on the plane, unique to this place and never taken off-plane, has been damaged, compromised, stolen, or otherwise altered. 

Antithesis

• The antithesis of the plane, something that would be expected in the prime material, is totally absent. Natural laws may be distorted to account for its absence.
• The antithesis of the plane is imprisoned, contained, rationed, or besieged.
• The antithesis of the plane has been memorialized, shunned, sanctioned, or put on display.
• The plane’s enemies or natural opposites are invading.
• Magic is altered in some fundamental way by the absence of something that would normally power, channel, or enable it. 

Who is nearby? Factions? NPCs? Monsters?

Epitome

• Natives of the plane, whose nature is linked to the essence of the plane. They are not merely planar loyalists; their very understanding of the cosmos is defined by the epitome of their plane.
• True believers, either in the ruler of the plane, or the nature of the plane itself. They either transmigrated here after death, or traveled here by magical means.
• The ruler of the plane. Whether a demigod, demon-king, or something stranger.
• Created creatures made of the essence of the plane by archwizards, gods, or others who use the plane’s essence as raw materials.
• Creatures from allied planes who have come to visit, trade, or evangelize. 
• Guardians, persecutors, or judges of the epitome, who seek to destroy, expel, or dominate the antithesis. 

Antithesis

• Creatures trapped here, either intentionally by denizens of the plane, or those stranded by accident.
• Creatures that are valued or respected because – due to their antithetical nature – they can do things or provide value that native denizens cannot. 
• Invaders from an opposed plane who have come here to destroy or conquer part or all of this plane.
• Explorers seeking to secure the epitome of the plane for use as an antithesis on their own plane.

What hazards, traps, or dangers are here?

Epitome

• The essence of the plane is hostile or otherwise dangerous to travelers. Simple actions like movement, breathing, or eating and drinking may be difficult.
• An out-of-control or escalating expression of the epitome is becoming more extreme over time.
• The landscape or physical properties of the plane are changing in a way that defies material plane laws.
• Traps, barriers, or other intentional dangers have been established to keep planar visitors either from accessing sensitive areas, or out of the plane entirely.

Antithesis

• A forced merger or overlap with an opposed plane creates violent or unpredictable interactions.
• Open conflict between factions or individuals over antithetical elements. Multiple factions may seek to recruit outsiders. Innocent bystanders may be caught in the crossfire.
• Remainders of a long-ago planar conflict between epitome and antithesis persist to the present (e.g., metaphysical minefields).
• Weapons deployed in the plane broadly attack a weakness or vulnerability inherent to the epitome of the plane, endangering anyone unfortunate enough to be nearby. 
• Something hazardous was placed here for containment or safekeeping, because the nature of the plane itself or the behavior of its denizens can suppress, control, or monitor the antithetical thing.  
• Something stolen from an antithetical plane, which is disruptive or wrong in this place.

Here is our random generator that incorporates all these bits into just four prompts. We can adjust it some more going forward; maybe we want the epitomes to be several times more likely than the antithesis.

https://perchance.org/whatishappeningonthisplane

Bonus Table: How Did You Get Into This Mess? (d12)

1. A wizard did it. A spell was cast upon you. It went wrong.
2. A wizard did it (the wizard was you). You tried to cast a spell. It went wrong.
3. Portal passenger. You wandered into a portal and now you’re here, wherever here is.
4. For science. An experiment went awry. Whether you were an experimenter or a test subject doesn’t matter now.
5. Trapped! A nefarious entity created an inter-planar trap. Congratulations, you have sprung that trap.
6. Hot pursuit. Something is chasing you across the planes. You don’t know what it is, but it seems to be accelerating and salivating.
7. Left behind. You were hired for some specific job or expertise, but the person who hired you left you behind. 
8. Cursed! Whether it happened to a distant ancestor long ago, or to you personally last week, the terrible curse has sent you hurtling between worlds.
9. Debt. You owe such a vast amount that when a particularly dodgy character offered you an especially suspicious way to get out of town -- really, really far out of town -- you took it.
10. Transmigration. You died. While your soul was on its way to the afterlife, it took a wrong turn, and you are now lost. Depending on where you were destined to go in the afterlife, this may be either relatively good or relatively bad news for you.
11. Prison break. You were trapped in Tartarus, a demiplane, or some similar extraplanar confinement. You're free now but you didn't plan too far beyond your escape.
12. Sole survivor. You were part of a highly larger team that was intentionally traveling the planes. You're not in a good situation now, but you wouldn't trade places with your recently deceased companions.

Going Up, Going Down: Incrementing Die Sizes for Information Density

I've occasionally seen people online present six related random tables, with each table tied to one of the classic polyhedral die sizes. For example, rolling a d4 to choose a location, then a d6 to determine what’s currently happening there, then a d8 to decide what monster is present, and so on.

I don’t know where this idea originated, or if there’s a proper name for it, or how far back it goes. It’s just a nice execution, because it’s satisfying to roll a fistful of dice all at once and then discern the results. It’s a fun prep technique, but a little too time-consuming for use at the table. Still, there’s something compelling about using all six of the main dice in one throw. What kind of similar ideas would be more useful at the table?

Picture a single random table, numbered 1-20. The entries near the bottom of the table are positive (or at least neutral) for the PCs, while the ones at the top are increasingly negative. For example, in a dungeon, a result of 1 could be something simple like a torch burning out, while 20 would indicate a random encounter with the most dangerous monster in the joint.

When you first roll on this table, use a d4. Interpret the outcome of the result, then cross it out. If you roll that same result a second time, nothing happens, but you permanently increase the die size to d6. Continue in the same fashion, increasing the die size to d8 and so on whenever you hit a previously rolled result that has already been crossed out. If you reach the d20 and can no longer increment upwards, treat re-rolls of previously rolled monsters as the nearest as-yet-unrolled result (and cross it out afterward). 

Rolling Downhill

The above method builds in a time delay for the really tough stuff. The increasing die means that the worst danger from the deepest part of the dungeon won’t show up while exploring the first room; but the incrementing dice and crossed-out possibilities mean that the more time the party spends in the dungeon, the more certain it is that those events will eventually happen.

What if we want at least a chance that the biggest dangers show up early? We can turn this around and put the bad stuff at the low end of the number range, and the good (or at least relatively "less-bad") results at the top.

Start with the d20, and roll as usual on the 1-20 table, crossing off results as you go. When you reroll a previously rolled result, switch to the next-smallest die. So the DM could roll 6, 11, 3, 20, 11, crossing out each of those results. When the second 11 is rolled, the die increments down to the d12; results 13 and above are now out of range.

This one will run on a much more aggressive clock, and after a handful of encounter rolls, the PCs will quickly get into hot water. For some counterplay (and a chance to still use numbers that have fallen out of the shrinking die range) give the players a way to release steam from the dungeon. Say that every favorable negotiation or parley with an NPC or a faction can increment the die back up by a level. Or exploring a certain number of rooms. Whatever connects to the intended gameplay loop and provides some motivation to the PCs.

There’s nothing special about either of these systems that makes them “better” than conventional random event or dungeon exploration tables. They merely take advantage of the natural tactile appeal of the varying sizes of polyhedral dice. And for in-person games, the d20 (in the first system) or the d4 (in the second system) is a powerful sign to the players that they are in as much danger as the dungeon has to offer. 

Parceling out the GM's Hats

There are TTRPG games with a traditional GM, and there are GM-less games. Is there any useful space in between? 

I previously wrote about games in which the GM role switches mid-session. As I said at the time, I think that mid-session-switching games will only appeal to a narrow band of experienced players. If your entire group is made up of GMs who have past experience running games, sure, no problem. But that’s not most groups.

I started thinking about this some more after some conversations with several players who had strong creative instincts, but were reluctant to run their own games. I generally believe that most people are too hesitant to try running a game. They believe it’s a unique skill that they don’t possess. I usually advise them that it’s much easier than they realize, and they should at least run a one-shot to try it out. And I still believe that.

But after those conversations, I also realized that the reasons players don’t want to run games are more nuanced than just hesitation. Some potential GMs love worldbuilding, but hate rules. Some are interested in tactical and crunchy mechanics, but don’t like portraying NPCs. Some are happy to act as a master of ceremonies, but don’t want to serve as a judge. 

Some of these preferences are things these people should interrogate to see just how deeply they are rooted. But some of them are just hard stops, and keep people from running games, if they aren’t comfortable wearing each of the hats on the GM’s many-armed hat rack.

So could we divide GM duties into different roles?

A couple caveats before I proceed. I have not (yet) tested this idea in a game, so it could shatter on impact with actual players. I’m also sure that some game has done this before; I just don’t know which game it is. I’m not suggesting that this is a completely new idea.

The Architect. This is the part of the GM job concerned with creating stuff during prep. It might be worldbuilding or dungeon maps or even creating monsters. Players who love laboring over intricate backstories and lore are well-suited to this role.

The Actor. This role characterizes NPCs. Players who enjoy roleplay itself will gravitate toward this role.

The Adjudicator. This is the judge, the GM who interprets the dice and the action of the world. Players who know the rules front to back often enjoy this role.

The Antagonist. The antagonist is simply a player who controls monsters and other hostile creatures. Their resources are limited to what the Architect has placed in the current milieu; but they can otherwise devote their tactical energy to providing a challenge for the players. A player who loves combat – or just coming up with complex plans and ambushes – can handle this role.

This is a simplified model; we could make a longer list if we needed to. And most games won’t need or want four different GMs. But two people could wear two hats each; for example, a creative type who enjoys serving as architect and actor, while leaving the crunchy stuff to a second GM who acts as adjudicator and antagonist. And these roles can’t cover everything the GM is required to do; the GMs would need to navigate places where their roles leave gaps, or overlap. But I believe it would be an interesting experiment.

Puzzle Escalation (Not Obstruction)

Last week: Solving the Problem of Puzzles in RPGs

In far too many games, when players get a puzzle wrong, the GM’s answer is “you make no progress.” This creates a frustrating dead-end, or invites uninteresting brute-force solutions. Taking damage is slightly better, but isn’t much of a deterrent in new-school games where PCs have piles of HP and ample healing, and also may not work well for games where physical violence is rare or frequently fatal.

Puzzles should instead escalate the current situation. Downsides of failing a move in a PBTA game are good for ideas; phrases like “you leave something behind,” “you owe someone a debt,” or “you lose face with a faction.” In a dungeon or other dangerous or hostile space, it can be as simple as asking what other forces are at work here, and how the overall situation worsens for the PCs when they beef a puzzle.

Puzzles that were created intentionally and follow some kind of functional design should have safeguards. Puzzles that are natural, or unintentional, or represent the degraded or decayed version of a previously functional system, should present hazards.

You Cannot Have a Meaningful Puzzle if Strict Time Records are Not Kept

As with exploration and trapfinding, most puzzles depend on external time pressure. A tight clock applied to a puzzle can solve most of the problems I identified at the outset of this post. In a dungeon scenario, this can simply be part of the overall tracking of time within the dungeon, with the built-in penalty for dwelling too long in one space.

Cooldowns can also help. The computer terminal locks for an hour after three incorrect password entries. The temple inscription can only be read during a 15 minute period when the sun is directly overhead and shines through an aperture. If the PCs are going down the wrong path with their questions, the NPC who is the “voice” for the puzzle simply refuses to talk to them further until the next day.

Mister Police, I Gave You All the Clues

Any puzzle can be solved if the PCs find enough clues. The GM can give them more and more over time, up to and including giving them the whole answer… with two caveats.

The first caveat is that the answer should come piecemeal, broken up into many small bits. If the party solves it early, with fewer clues, they are essentially rewarded with time saved (again reinforcing the importance of time). If they don't solve it early, but stick with it, they will eventually get enough clues that the solution will become obvious.

The second caveat is that clues should be diegetic objects. They should not be the results of ability checks or similar mechanical output from the character sheets. Making them diegetic objects is a great way to take otherwise-irrelevant background information about the game and work it into the  actual execution of the session. 

Solving the Problem of Puzzles in RPGs

What exactly is inside the dungeon? Monsters and treasure, definitely. Some NPCs and treasure, sure. Traps and hazards, of course. And… puzzles. 

But puzzles can cause problems. The game grinds to a halt as the DM explains and re-explains an incongruous logic pretzel or goofy riddle. The natural born puzzlers in the group go to work, while the rest of the table checks out.

Even with the best of intentions, people are left behind. In one game where I was a player, our party encountered a puzzle that was an elaborate gag… referencing a TV show that only some of the players in the game had seen. (The GM in question did a great job generally, so I was happy to sit on the sideline in that instance; but the example was still instructive).

By player request, puzzles are part of the mystery of my Knave game, Strangers on a Train, and I have found it’s possible to gradually, slowly, get better at making puzzles. And puzzles are worth keeping in our games.

Tricks, Not Puzzles

Most people know that old-school D&D included instructions for populating dungeons with monsters, treasure, and traps. But new-school D&D players overlook some of the other guidelines for keying the dungeon. Old-school D&D did not suggest (at least not principally) featuring “puzzles” in the sense they are commonly understood today, but rather included “tricks.” The difference is informative.

A puzzle, as typically deployed in RPGs, has one correct answer. There is no partial success. There is no amusing mistake that still moves the scenario forward. There often is not even room to improvise workarounds that bypass the puzzle entirely. Poorly written 5E adventures merely offer a greasy trail of ability checks to give hints to bored PCs until they bump into the right answer.

A trick has no answer because it is not asking questions. It is probably more harmful than helpful to the average PC, but with the tantalizing possibility of lucky benefits or creative leveraging. Because a trick doesn’t have a “right” answer, it’s much easier to plan one. A puzzle, like a joke, needs to have a satisfying conclusion. A trick is more like an amusing non-sequitur that sets the stage for improvisation.

The remainder of this post will talk about puzzles as they’re more commonly understood, but a good time-saving first step is always to ask “could this just be a trick instead?”

Built into the Background 

When used poorly, puzzles have the proverbial glowing neon sign pointing to them, announcing their presence. They are “mini-games,” which is to say, a break in the flow of the actual game everyone showed up to play.

A really good puzzle is one the players solve without realizing they’re doing a puzzle. Piecing together the clues to the location of the lich’s phylactery or identifying the murderer may in a sense be a “puzzle,” but if the players solve it purely through asking questions and interacting with the fiction, all the better! The best puzzle is the kind that never announces itself as such.

The Puzzle Pieces 

Begin with a collection of related pieces of information. A list can be enough. A dozen animal names. Digits with a non-obvious connection, like ascending prime numbers. A collection of paintings by one particular artist.

Now imagine that someone has access to only a quarter of the list. Could they intuit the pattern, and fill in the rest?

Take this process and phrase it as a question. In its simplest form, it could be “if X, then Y, what comes next?” Or “if you have seven objects with such-and-such characteristics, and seven places to put them, how much information would you need to place all of them correctly (without brute forcing it)?”

The puzzle will feel more natural if the puzzle pieces are not all grouped conveniently around the puzzle itself. The pieces can be bits of information that dungeon residents know. They can be evidence near the crime scene. They can be treasure. They can be pieces of miscellaneous lore or background information that don't fit elsewhere in the game.

After we’ve finished our list of puzzle pieces, we can work backwards to “...why?” Depending on the type of game, this may be easy or difficult. A funhouse dungeon doesn’t really need to explain its puzzles; they’re present because that’s the style of play the players signed up for. But most other games will want more context. Remember, this doesn't have to suggest realism; it simply has to convey verisimilitude, which is to say, it must be true to the shared reality of the game itself.

So why is there a "puzzle" in the game? 

• The obstacle has deteriorated from something that used to unambiguously hide or limit access, and can now be "solved" in its weakened state.
• The obstacle includes "clues" that were a reminder for its creator, or an explanation intended for someone else.
• The obstacle is a test to determine worthiness (but be careful with this one, as it's typically overused).
• The obstacle was designed as a game first and foremost, and only later adapted to gate something valuable or useful, probably by an entirely different party from the creator.
• Someone has previously partially solved the puzzle, and showed their work in the process, providing enough information to get started.
• The "puzzle" is an obstacle that is malfunctioning in some way that has subverted its original purpose and made it cryptic and opaque.
• The obstacle's clear meaning and purpose is in a lost language, and the “puzzle” lies in ambiguous translation.
• The obstacle is possessed or guarded by a trickster spirit (such as a mischievous imp or rogue AI) that amuses itself by confounding those who wish to move past it.

Next week: Puzzle Escalation (Not Obstruction)

Life on the Grid

I don’t like to bring a lot of gridded maps into my in-person* RPG sessions. I prefer theater of the mind for combat and other tense, moment-by-moment scenarios. I typically only want maps for the most spatially complex situations, where they’re absolutely necessary to ensure player comprehension.

But I like to interrogate my preferences, and it takes me to some interesting places…

So Transparent

Getting off the grid was a gradual process for me. I went from the big Chessex map (nice, but hard to take on the go), to the interlocking dry erase tiles (more mobile, but still restrictive), to these transparent gridded overlays. 

The obvious use for the transparent overlay is to conveniently turn a gridless map or landscape into a battlemap. The less-obvious use is to lay them over any image and use the resulting grid to facilitate play, whether we interpret the spatial relationships it creates literally or figuratively.

For our Strangers on a Train game, I can print any sort of image of outdoor terrain, lay the grid on top of it, and immediately create a pointcrawl with measurable distances. This technique isn’t limited to top-down views; for images conveyed from more conventional perspectives, the grid can also suggest height. Encounter checks trigger (and time advances) whenever the group moves across one of the gridded lines.

Spatial Relations for Every Situation 

But the image doesn’t even have to represent a place. Any handout can benefit from an overlaid grid. 

Progress clock. It’s one thing to say the bomb explodes after five setbacks. It’s another thing to watch a token move across a grid laid over an image of a bomb on the center of the table.

The kaiju. Print an image of a giant creature. Lay the grid on top of it. Player tokens or miniatures reinforce the relative size of the creature, and the grid forms a map for climbing around its body and attacking its weak points.

The puzzle. Lay the grid out on top of a puzzling image, and invite PCs to indicate which bits of it they will investigate, in turn.

The factions. Print an image displaying rival factions. Use the grid to map the ebb and flow of their power and influence; or alternately, plot the PCs’ relative prestige with either side.

The battle. Take a painting with a lot going on, like Thomas Cole’s The Course of Empire; Destruction and lay the grid over it. The grid subdivides the space into grokkable chunks, and gives players a way to choose what to interact with in a scenario that’s bigger than they are.

Advanced techniques. Is a square grid too boring for you? Grab any image you like and run it through a Voroni Generator. Great for a pointcrawl through a twisted, physically abstracted space like the plane of Limbo, or corrupted cyberspace.

*I use the grid more liberally for online play. Without people in the same room, face-to-face with each other, there’s a greater need for imagery to visually center the players and keep them on the same page. 

The Steelman Argument for Fudge

Last week: The Argument for Fudge-Free D&D

Obviously, I am anti-fudge. I am against fudging rolls of the dice, and equally against fudging other aspects of the game via Mike’s “dials.” But for any strongly held belief, it’s worthwhile to take some time and try to find the strongest arguments against one’s own position. A “steelman argument” is the opposite of a strawman argument – attempting to look at an opponent’s argument in the strongest possible light. Below are my attempts to steelman the fudge.

It's worth acknowledging that many home games are built around untested alpha content. Many DMs don’t have the luxury of running scenarios multiple times. If they are running published content, it has (hopefully…) been tested, but who knows how extensively. If the DM is homebrewing their game content, that content is reaching the table in a purely untested form. Most of the content I have run in long-running games was like this. And as such, it was untested, and I (as the adjudicating DM) had to address instances where I (as the planning DM) had made mistakes or failed to anticipate how the content would actually work in practice.

My solution to this was usually to use diegetic solutions and flexible adjudication tools rather than fudging the dice or the dials. But I understand why many DMs find that so challenging, and it can sometimes be really difficult to find a diegetic solution to a particular game situation (whereas fudging avoids the need for a diegetic explanation).   

It’s possible I’m underestimating the DM's capacity to make the game “more fun” in the moment. Perhaps some DMs can make these adjustments in a way that reliably does push the game in more interesting or fun directions. Perhaps I have a below-average sense of these adjustments when adjudicating the game, and it's easier to do without risking player engagement than I think. I’m skeptical, but I could be wrong.

It could be that a game has more tolerance for fudge than I believe it does, as long as the players trust the DM. Mike includes some caveats in the article suggesting that frequent or arbitrary fudging is bad. I would obviously go much further than he does. But a steelman argument for fudge would suggest that fudging has to be very frequent or severe to really harm a game. In this view, the degree of fudging Mike suggests won’t be nearly as apparent as I suspect it will be.

A Player Transparency Compromise

Acknowledging that one, two, or all three of the above steelman positions could be true, I would offer one final position of compromise between my position and the pro-fudge DMs out there. I believe the DM who wants to fudge in this manner should clearly state their intent to do so during session zero, or an equivalent communication before the game begins. (It’s possible Mike does in fact do this in his games; the article in question doesn’t specify either way.)

Fudge as a session zero topic can sit right alongside the discussion of the desired level of lethality and challenge in the game. The DM who intends to do this should simply say “I’ll sometimes fudge monster HP in the interest of making the game more fun.” If I were a prospective player, this would at least alert me that this wasn’t the game for me. And other players could build their expectations for the game around this expectation.

I believe this compromise position doesn’t contradict either Mike’s points in his article, or the steelman positions I outlined above. I would challenge fudge DMs – are you transparent with your players about what you’re doing? If you are not, I’d be interested to know why.

The Argument for Fudge-Free D&D

“Fudging dice rolls is bad” is now a truism in most online RPG discussion spaces. Along with familiar phrases “don’t railroad” and “talk to your players,” “don’t fudge” has been repeated exhaustively wherever RPG advice is shared. I wrote about this subject previously because I wanted to get my thoughts down on paper, but I don’t harbor any illusion that those posts were particularly novel. This is well-trodden ground.

But pro-fudge people don't give up easily. They look for other ways to bring the fudge back into the game. For example, moving the fudging action from the results on the dice to other parts of the game. I don’t believe this is any better than conventional dice fudging.

Before I proceed, I want to contextualize my criticism. I don’t write a lot of posts criticizing other writers’ views. The reason I’m doing so here is because Mike Shea’s (AKA Sly Flourish) GMing advice is generally sound, particularly for his putative audience of new and mid-level GMs who are still figuring out how to run games. I also think his signature “lazy DM” style has much to recommend it in a hobby larded with excessive, useless session prepping. I have his Return of the Lazy Dungeon Master book and would recommend it for a new 5E DM looking for prep tools. If Mike routinely gave out bad advice, I wouldn’t bother to comment on a post I disagreed with.

I further recommend you read Mike’s article in full before proceeding, so you can make your own evaluation of his argument before you read mine. While I have endeavored to convey his statements accurately, I welcome counterpoints in the comments about anything I missed, misread, or interpreted unfairly.

“Fudge monster hit points whenever it leads to a more fun game”

When and to what degree fudging would be “more fun” than the alternative is not as obvious as it sounds.

Gauging what is more or less “fun” while an RPG session is in motion sounds like it would be easy. But it’s not. It’s like trying to figure out what is wrong with your car while you are speeding down the highway.

Why? I contend that one of the most appealing characteristics of TTRPGs as a form of entertainment – contrasted with what board games and video games and sports and many other pastimes have to offer – is just how hard it is to predict what is going to happen. The incredibly wide range of unexpected (but organic and logical) outcomes from the fiction, which the players never would have seen coming, is a big part of the appeal to the hobby. 

When the DM starts tinkering with the fiction to make it “more fun,” they are constricting that range of outcomes. I remember very clearly the first time I realized, as a player, that the DM was doing this. The DM in question (who did a great job overall) confronted our mid-level party with a beholder. The other characters fled, but my character fell behind. I prepared to make a last stand and hold back the beholder for as long as possible, to buy more time for the rest to escape. I obviously hadn’t planned for this outcome, and the idea that this character would suddenly die in this manner was interesting, and interesting in a way that I’ve really only encountered in TTRPGs.

But the DM fudged. To their credit, most or all of the fudging was diegetic – deus ex machina by NPC. I’m sure the DM was trying to make the game “more fun.” But for me personally, the cost of the intervention outweighed its supposed benefits. 

“Fudge When it Ends a Battle That's Overstayed It's Welcome”

Most fights should end well before either side hits 0 HP.

It’s wild to watch people twist themselves into knots trying to “solve” a “problem” that was sufficiently addressed 40 years ago in Moldvay Basic. If the DM uses basic D&D morale, or an equivalent system, monsters will flee once the battle has overstayed its welcome.

Mike writes that “It doesn't make sense for [enemies] to just quit and walk away” after the enemy boss is dead or the main goal of the battle has been achieved… when this is exactly what would make the most sense. Excepting constructs or other “unthinking” monsters, most minions would certainly surrender or flee in that circumstance, and the battle would end in a logical and interesting way.

This is not arbitrary or immersion-breaking; quite the contrary. The idea that enemies break and run is a much more realistic depiction of fantastic (medieval or, more generally, pre-modern) combat than the video game slog to zero Mike appears to be envisioning. Most of the casualties in historical battles occurred after one side was routed and fled the battlefield. I’ve often run D&D 5E combat this way, and it works perfectly fine.

Based on other things he’s written, I don’t think Mike is opposed to non-lethal conclusions to combat in general. He may just consider that outside the scope of this particular article. But a diegetic solution like morale solves this problem so much more elegantly than rigging the fight behind the screen, and I'm left wondering why fudging “dials” is even a viable option by comparison. Combats that end when enemy morale breaks should be one of the most common resolutions to combat. 

“When You're Beefing Up Bosses” and “When It Makes Sense for the Story”

Fixing monster HP totals should happen during prep, not while running the game.

These two are addressing the same idea – changing HP mid-combat to hotfix D&D 5E’s creaky design.

Hacking the game is hard enough during prep. Trying to do it while running a session is even more difficult. To return to my original analogy, this is like saying to a passenger in your car, “Sorry, I didn’t have time to go to the mechanic this week. I’m just going to lean out the window with this lug wrench and do some tightening!”

For emphasis: It is perfectly fine for the planning DM, during prep, to adjust the HP of monsters they’re considering using. The planning DM can alter the fiction in many ways during prep, and that’s fine until the PCs encounter the fiction in-game. The adjudicating DM has a different responsibility, and should be skeptical of their own instinct to patch the game while it is underway (on top of their many other responsibilities as master of ceremonies and referee).

This use of fudging also raises a further question: “why have hit points at all?” It may sound like I’m being glib, but it’s a serious question. In a Free Kriegsspiel Revolution game, the fight with the lich would simply continue until the game participants agreed it made sense for it to end. Everyone understands that there is no number for the PCs to push to zero, so there’s nothing to fudge. 

There are other RPG systems that are otherwise designed to better facilitate the cinematic thrill of fighting the boss – with the dramatic moments arriving at just the right time, and the boss dying right at the climax of the fight. D&D 5E is not that game, and trying to force it into that box just chips away at the game’s verisimilitude and shared investment.

And yes, there is a cost. Players are not dumb. They will figure out what’s happening, and learn that the resolution of combat isn’t based on their skill, or an impartial and genuine engagement with the fiction, but rather the DM uncommunicated and ambiguous sense of fun, and their desire to perform emergency maintenance while going 70 miles per hour.

Next week: The Steelman Argument for Fudge