Category Archives: RPGs

OSR Spellcasting Systems

I’m amazed by the variety of implementations of spellcasting in OSR systems. I asked OSR Reddit what the problems are that these new implementations are trying to solve, what the most popular approaches to spellcasting are, and people’s preferences. I distilled the discussion into the following analysis.

  • Perceived problems with OD&D
    • Boring
    • Vancian magic isn’t magical enough
    • Difference between spell levels and PC levels
    • Set spell lists
    • Doesn’t suit certain settings and types of campaigns
  • Types of spells
    • Set spell lists:
      • By spell level
      • By player level (e.g., level 2 spell for level 2 caster)
      • By schools (e.g., Necromancy)
      • Level-less
    • New spells are discovered in game (as treasure or from other casters)
    • Improvised by player
  • Availability of spells
    • Cantrips can be used any number of times
    • Spell slots:
      • With slots per level
      • Overall number of slots (level-less)
      • Spells have physical form (e.g., scrolls) and take encumbrance slots
    • Any spell they know:
      • Chosen by player
      • Chosen by player if they pass a check (e.g., Intelligence)
      • Determined randomly
    • Casting has a cost:
      • Time — Spellcasting takes 2-3 rounds
      • Reduces hp
      • Mana cost
  • Success of spells
    • Target, if any, gets a saving throw
    • Caster rolls for success
    • Caster failure results in anything from magical side effects to catastrophes
    • Magic Dice (MD) per level, with option to roll as many or as few as you wish (GLOG) but with doubles result in a side effect and triples result in a catastrophe

Based on this research, I decided to do a survey to provide feedback to the OSR (Old School Renaissance) community on what Level 1-3 spells are considered most useful. The results can then be used to create shorter spell lists and player aids.

For each of the six groups of spells, you’ll be asked to rank at least the top three spells for OSR play, from most useful to least useful. You can take the survey here:

https://www.surveygizmo.com/s3/5315422/2f3a7fa043f5

Many thanks!

(Illustration credit: The Sorcerer, by Jennie Wilde.)

Reskinnable Pantheon for a Module

Thought experiment: What should a generic pantheon look like?

For a hexcrawl module I’m writing, I didn’t want to go with specific gods, but I’ve struggled to come up with something generic that the DM could easily adapt to their setting.

I want to make less work for GMs. This is prompted by me encountering things in modules like “A steep hill rises from the forest and at its top there is a shrine where stands an old statue of Yemathic, about 20 feet tall” and not finding enough about Yemathic to know what is intended. (OK, pages and pages on, including passing a few more references, I found out that it was a god or goddess – not sure which – of justice.)

The Greeks and Romans had almost the same gods, just with different names: Ares vs. Mars. Then the Romans tried to translate every local deity into their pantheon: “Oh, your god Taranis is just another name for Jupiter.” So that is the hybrid I’m looking for: “You encounter a wall painting and a shrine to what the runes name as Ausdia, a goddess of the sun, with a solar halo behind her head.” Easy to reskin to Apollo or Belenus or Dol Arrah or Frey or Odur or Pelor or Phlotus or Re-Horakht or the homebrewed solar deity.

One approach would be just to go with the 5e domains: Knowledge, Life, Light, Nature, Tempest, Trickery, and War.

For another approach, I re-read “Appendix B: Gods of the Multiverse” from the 5e PHB and did a word count on the Deity column:

Word Count Word
12 war
11 magic
9 nature
9 sea
7 death
6 knowledge
6 fertility
6 storms
6 good
5 healing
5 giants
5 craft
5 light
5 fire
5 sun
4 music
4 love
4 evil
4 elf

I’m going to go with 8 (so I can use a d8 to look up a god), with a d6 for gender (1-3=god, 4-6=goddess):

  1. god/goddess of death [War domain]
  2. god/goddess of fertility [Life]
  3. god/goddess of knowledge [Knowledge]
  4. god/goddess of magic [Trickery]
  5. god/goddess of nature [Nature]
  6. god/goddess of the sea [Tempest]
  7. god/goddess of the sun [Light]
  8. god/goddess of war [War]

I’m going to leave out alignment references, as I think those are easy for a GM to swap in as needed.

For when I want to use a d12 (ancient ruins):

  1. god/goddess of craft [Knowledge domain]
  2. god/goddess of death [War]
  3. god/goddess of fertility [Life]
  4. god/goddess of fire [Trickery]
  5. god/goddess of healing [Life]
  6. god/goddess of knowledge [Knowledge]
  7. god/goddess of magic [Trickery]
  8. god/goddess of nature [Nature]
  9. god/goddess of the sea [Tempest]
  10. god/goddess of storms [Tempest]
  11. god/goddess of the sun [Light]
  12. god/goddess of war [War]

A common pattern, which I omitted from these, is to have gods of a race: god of giants (5 references in the PHB), god of elves (4), etc. Those can easily be added depending on the history of a location.

Image credit: Metropolitan Museum of Art

One-Page Dungeon Generator

I love the Dungeon Contest’s one-page dungeons, where you can see the entire dungeon and its contents at a glance. Especially those dungeons that get beyond the fixed narrative structure of five-room dungeons. My own attempt for the Dungeon Contest was thematic but probably unoriginal, though it did lead to three great sessions in my last campaign and the rise of a new Big Bad Evil Gal (you can download Catacombs of the Lich Queen here).

So I was disappointed when I ran Hex Describe for the first time and realized it lacked any single-page dungeons in its hexes. I suggested something quite simple, providing five bullets built from three types of rooms (entrance, interior, final):

  • Natural cavern
  • Mushroom farm
  • Kitchen
  • Guardroom
  • Throne room

Well Ktrey Parker ran with it, adding themes and adapting his tables, and Alex Schroeder built a dungeon-map generator, and together we iterated and iterated, ending up here:

Check it out! You can generate your own random dungeons, varying in size from 5 to 14 rooms. And, if you’d like to improve the generator, we’re always looking for additional authors to add to our tables.

Use Text Mapper to Create Random Maps and Hex Describe to Create Random Campaigns

For my current hexcrawl campaign, the Hexedland, I created a 10-by-10 hexmap using Text Mapper. I used the random generator based on Erin D. Smale’s algorithm, then keep tweaking the results by hand until I got something I liked. For instance, I wanted the starting hex to border each type of other terrain (forest, mountains, hills, swamp, lake, grassland) so that players could choose the type of environment (and therefore monsters) they wanted to encounter. I lengthened a mountain range that divided the middle of the map, and had the west side be arid and the east side be fertile, indicating that rains blow westward.

Editing the map doesn’t involve drawing but involves changing descriptions of hexes:

0101 green firs thorp
0102 light-grey mountain cliff1
0103 light-grey mountain
0104 light-green firs thorp
0105 green fir-forest
0106 light-grey forest-hill
0107 light-grey mountain
0108 light-green firs thorp
0109 grey swamp
0110 grey swamp
0201 light-grey mountain cliff0
0202 white mountain
0203 light-grey mountain
0204 light-green fir-forest
0205 light-grey mountain cliff0

I also ended up adding names into the map for some of the major landmarks. The following illustration is just of a random map, as I don’t want my players to see the actual map I’m using. Half the fun is in the discovery.

hexmap

Now why create maps like this?

The key reason is that you can then copy that map into Hex Describe and create an entire campaign!

All of a sudden you have 100 regions that your players can go explore! And of course you can hack and edit any description to better suit your players’ interest. But it is much easier to start with 100 descriptions already generated then to start with a blank page.

So far I’ve extensively rewritten each of the hexes my players have visited, keeping some elements from Hex Describe’s output. For instance, I’m replacing the Halberds & Helmets cosmology with the gods from my one-page pantheon. But the random hexes generate ideas that push me to be more creative and inspire confidence that, should my players set off in an unexpected direction (as they often do) or get lost (less often), I can describe where they end up.

I encourage you to give both Text Mapper and Hex Describe a spin!

One-Page Pantheons

The general advice for DMs building their own campaign worlds is to recognize that most of world creation won’t end up being experienced by the players. While you can go full Tolkien if you wish (to obey your own muse), you’re typically better off creating simple systems and then using fractal design to zoom in on those parts that players show an interest in.

For my Hexedland campaign, rather than write The Silmarillion, I developed a one-page pantheon, where the description of each god shared the major myth associated with that god. Read together, this provides a summary of the mythos. What follows is what I provided players at Session Zero:

Humans believe in all 12 gods, but most have two gods they consider patrons. They wear an upper arm ring with a symbol of their favored god and a finger ring with the symbol of their second god. They do not need to pick a god that exactly matches their alignment.

Common beliefs no matter which gods a player favors:

  • Gods are not omniscient. They rely on prayers to learn what is happening in the world.
  • Gods are not omnipotent either. They channel their actions through clerics and believers.
  • Some people became gods, either through Diahaj granting them apotheosis or through them recruiting enough worshipers to achieve it on their own (Tolcu-Tolcu).
  • Where a spell on a scroll can only be read by magic users, a spell on a prayer leaf can be read by anyone. A prayer leaf is a piece of parchment folded in half and then folded in thirds and placed in a small leather pouch.
  • Each god has an associated taboo, an activity a character must avoid. For instance, followers of Wenmaju avoid dry food (rations), while followers of Pentwer won’t set traps or associate with someone who sets traps. Violating a taboo will suspend a player’s ability to use any prayer leaves until a tithe has been paid to the right temple in Auspele.

The twelve gods often contest one another in mortal affairs. Other, more idiosyncratic gods are worshipped elsewhere.

d12Deity / Algnmnt.Archetype / TabooSymbolBackground
1Ausdia
LG
Creator / public eatingSunburst with 12 raysShe created the sun, the trinity, and peopled the world with immortal plants and walking trees. No one ate anyone else.
2Neccelio
CE
Lawbreaker / jokesCave entranceThe god of the trinity, he rebelled and created animals and night and death. Sometimes called “The Devourer.”
3Sterdia
LG
Caregiver / selfishnessCrescent moon & starThe goddess of the trinity, after the first nightfall, she grew the moon and the stars.
4Diahaj
CG
Magician / routinePainted skullThe other-gendered god of the trinity, they transformed substances, resurrected the dead, and deified mortals.
5Wenmaju
N
Lover /
dry foods
Two eyesA sublimely beautiful intersex human, they were wooed by each of the trinity.
6Regwena
NG
Hero /
torture
Bow with arrowOriginally a princess, she singlehandedly defended her brothers from an orc attack and then found and freed her kidnapped husband. Mother of Coronosej.
7Coronosej
LN
Ruler / frivolityCrownShe created an alliance of races, defeated an invasion of monsters, and became the first empress of Cedreg.
8Pentwer
NE
Explorer / setting trapsFlamesA dwarf, he went into the wilderness and blazed a trail for settlers. (Caring little for who already lived in the wild.)
9Ogmalbo
LE
Sage /
deception
E-shaped runeAn elf, she compiled lore on the lands opened up by Pentwer.
10Zeipots
LG
Everyman / luxury goodsBumblebeeOriginally a human farmer, he was the first to tame bees and cultivate honey. 
11Mehupots
CN
Jester /
books & scrolls
Ancient vaseA halfling bard, she stole honey and invented mead.
12Tolcu-tolcu
CG
Innocent / bigotryCity on a hillShe taught cosmopolitanism and radical inclusiveness to the citizens of the empire, became widely worshipped and achieved apotheosis on her own.

Iconography

The archetypes for the gods came from The Hero and the Outlaw. The icons were all sourced from The Noun Project (for which I have an annual license).

Hexedland's 12 gods

Using “Old School Principles for Players” to Tweak O5R

I know O5R is seen as an oxymoron by many, but let’s go through the “Old School Principles for Players” section of Principia Apocrypha and determine what’s the least we have to change about 5e to live this philosophy. Its principles:

  1. Learn When to Run
  2. Combat as War, Not Sport
  3. Don’t Be Limited by Your Character Sheet
  4. Live Your Backstory
  5. Power Is Earned, Heroism Proven
  6. Scrutinize the World
  7. Interrogate the Fiction
  8. The Only Dead End Is Death
  9. Let Your Creativity Flow
  10. Play to Win, Savor Loss

I would argue most of the rule changes have to happen “behind the screen” rather than for the players. For the players, though:

  1. Learn When to Run – In my O5R campaign, the players know that encounters aren’t balanced. The fear is so strong they now sometimes run from threats they can defeat.
  2. Combat as War, Not Sport – They’re better than in our straight 5e campaign but have room to improve. After swearing they would hate a session without combat, we had a session without combat, and they all enjoyed it.
  3. Don’t Be Limited by Your Character Sheet – This one has taken coaching. “I approach the leader and do a Persuasion check.” “OK, but first what are you saying?”
  4. Live Your Backstory – Our Session 0 had them describe their relationship with the PC of the player next to them. (Next campaign I want to approach this with a DCC-style Level 0 funnel.)
  5. Power Is Earned, Heroism Proven – They recognize this principle and are already planning ahead. “We’ll attack that goblin tribe after we’ve learnt more about their lair and progressed further. Maybe we can convince the Wolf Claw orcs to attack with us.”
  6. Scrutinize the World – This is why we abandoned Forgotten Realms (the most well-documented campaign setting in history) in favor of a homebrew hex crawl. (Thanks to Hex Describe.) I still can’t get anyone to make maps, though. But they are deciphering the runic alphabet I’m using!
  7. Interrogate the Fiction – This is the hardest for them. There is a retreat to mechanics. “I do a Perception check.” “First, what are you looking at and for?”
  8. The Only Dead End Is Death – I’d argue removing resurrection-style spells is the main change needed to 5e. This group was resurrection-happy in Forgotten Realms. (Though Holmes wrote, “A seventh level cleric can raise the dead, if you can find one! Also, of course, wish rings and other magic can restore the deceased adventurer to his comrades and friends!”)
  9. Let Your Creativity Flow – They often come up with very creative approaches to situations and problems. Last campaign (which wasn’t O5R) they used a soul-stealing sword to capture the soul of the lich, when they couldn’t find her phylactery.
  10. Play to Win, Savor Loss – They do not savor loss. Three PCs were aged 10 to 40 years each due to a ghost’s Horrifying Visage and one of the players (25ish IRL) is so angry IRL that her character is now 38 years old that I’m starting to feel old indeed! But at least that has inspired an interesting quest to find the ingredients for a potion of lesser restoration (which will only roll back the clock for these three).

Behind the scenes, there’s a lot more work for the DM to use O5R:

  1. Learn When to Run – The 5e DMG says, “Choose the starting attitude of a creature the adventurers are interacting with: friendly, indifferent, or hostile.” I use the Holmes 2d6 rules: 2, attacks; 3-5, hostile; 6-8, uncertain; 9-11, friendly; 12, enthusiastic.
  2. Combat as War, Not Sport – They haven’t realized it yet, but every short rest I roll 2d6 and if a 6 comes up the rest is interrupted by a wandering monster. I’m trying to get them to better husband their resources.
  3. Don’t Be Limited by Your Character Sheet – I’ve kind of given up here. I tried to enforce Adventurers League rules of PHB+1 but that doesn’t help the DM when everyone picks a different book. I have characters from XGE, from Eberron, from Unearthed Arcana. But this is old school, too; Holmes wrote, “At the Dungeon Master’s discretion a character can be anything his or her player wants him to be… Thus, an expedition might include, in addition to the four basic classes and races (human, elven, dwarven, halflingish), a centaur, a lawful werebear, and a Japanese Samurai fighting man.” (Which is why I always laugh when OSR purists insist on 4 races and 4 classes.) I am definitely regretting the artificer, though, as I lose some control over the magic-item economy.
  4. Live Your Backstory – I’ve nudged the players out of town, as 5e had them expecting more stuff to happen in town, especially related to backstory.
  5. Power Is Earned, Heroism Proven – I’ve custom-created a magic-item table that offers limited upgrades. They’re all about leveling up, though, so that’s how they see their “power earned.” (Depending on how the session went, I award a third to half the XP they need to level up. This group has no interest in spending loot on manors, retainers,  political influence, etc. I tried repeatedly last year.)
  6. Scrutinize the World – The most work, by far. I spent a bit of the summer working on the campaign setting, so I wouldn’t have to prep much during the year. (Our last campaign ran from September to July.) It’s definitely cliché old school – they’re in one of the last bastions of human society, uncertain of what happened to the empire their city-state was part of 4 centuries ago or why their city-state was spared.
  7. Interrogate the Fiction – I’ve tried to be better about giving each location three interesting features to interact with (a la Sly Flourish).
  8. The Only Dead End Is Death – We use the three standard death saves. Too ingrained for the players to change, though we talked about it.
  9. Let Your Creativity Flow – I’m adding constraints to encourage creativity. Our last campaign, in Melvaunt, often involved players buying whatever magic item they wanted (from Thentia). Every player had a bag of holding; no bags of holding in this world, and we’re tracking encumbrance.
  10. Play to Win, Savor Loss – They really don’t want to start new characters at level 1 but want to introduce them at the same level as the dead character. This bums me out, but they were unanimous.

I have players that only know the core 5e mechanics of combat, attribute checks, and advantage/disadvantage, and do fine. And those mechanics are more than rich enough to hang an OSR game on. As for many of the other rules, they are often race and class and level specific, and those players who want to live off their 5e character sheet can and do. If I could, I’d limit the spell list more – many of the spells seem too powerful: my players use Mage Hand to pound the dungeon floor checking for traps, and use familiars to scout far ahead in the wilderness.

Still, my players want 5e and interesting sessions; I want OSR; so far O5R has made us all happy.

(I posted an earlier version of this on Reddit. Check out the comments.)

RPG Rules for Minecarts – Not!

Another bit of advice that I find particularly useful from Sly Flourish is to add three features to a fantastic location for players to interact with (see “What are three fantastic features of this location?” in Creative Mind Exercises for D&D). For a recent delve into a mine, for my 5e campaign, my three were a rusted minecart, rail track, and a trellis bridge over a deep chasm.

I didn’t create any rules for minecarts, nor do I think you should. Instead, improvise rules for features using your RPG system’s mechanics. For instance:

  1. The artificer took a turn to oil the wheels of the minecart, which I ruled he automatically succeeded at. (Where’s the fun in a minecart that doesn’t move?!)
  2. When players decided to get in and later get out of the minecart, I ruled they had half their movement range on a turn when they did that.
  3. The track sloped down, so when players entered or exited the cart, then the player at the front of the minecart had to make a strength DC check (which increased as more players entered, from 15 to 18).
  4. The minecart became portable cover for ranged attacks against troglodytes entering the mine. Treated as regular cover. (The wizard minor-illusioned a cover over the cart!)
  5. When one of the players hopped out, the cart started to move, so another player slammed his foot on a brake outside the minecart (which I had never described, but the players assumed would be there, so there it was). I decided on the fly that the brake was rusty and that I’d reverse the strength check for the fun of it – a high roll would break the brake; he rolled a 3, so I ruled he engaged the brake successfully.
  6. The half-orc missed the players talking about the narrow bridge over the chasm and decided to shoot the cart over the chasm – she failed a dexterity check (on trying to push the cart a certain way) and ended up going on a roller coaster ride! Fortunately for her, the dwarves had engineered the track to incline back up (improvised at the table) and come to a gentle stop outside the new shafts the dwarves had been digging before abandoning the mine.

Now “low prep” doesn’t mean “no prep” – you should prep what you enjoy prepping and what you find hard to improvise at the table. Keeping in mind that players are by their nature unpredictable.

In this case, my players had already had a session in this mine system and had found the loot while purposefully avoiding the dragon they had all told me back in Session 0 they wanted to fight. The troglodytes were just supposed to provide a strong start to the session – I expected the players to use the minecart to escape from the troglodytes and find another way out. Instead, of course, the strong “start” became almost the entire session as they magicked, battled, and then drove away the dozen troglodytes.

Sketching out rule subsystems for each feature of a fantastic location, such as the minecart, would have just been too much prep, in my case. And had I done it they would have probably spent all their time crawling along the trellis instead!

Image credit: Gzzz

Most RPG Products Sell Fewer than 50 Copies

DriveThruRPG and DMsGuild (same corporate parent) publish the number of their products that sell at different “metal tiers”. For instance, DriveThruRPG has had 10,719 products (as of today) sell 50 to 99 copies (Copper tier), compared to 1,792 products at that level for DMsGuild.

Extrapolating from these, you find that 72% of DriveThruRPG products haven’t even sold 50 copies, and 90% of DMsGuild products haven’t either. So if you’re looking for a quick market forecast for the RPG product you’re planning to sell through these channels, “under 50” is a good estimate (probably “under 10”).

It’s good to be the aggregator, though: DriveThruRPG has sold perhaps 8.8 million products, while DMsGuild has sold perhaps 1.5 million. (Those estimates will be off by an order of magnitude if non-metal products sell more than 4 copies apiece – my best guess, averaged across the two sites.)

DriveThruRPG DMsGuild
Metal Tier Threshold Products per Tier % Catalog Unit Sales Products per Tier % Catalog Unit Sales
No metal 0 86,454 72% 6 40,145 90% 2
Copper 50 10,719 9% 76 1,792 4% 73
Silver 100 12,061 10% 150 1,519 3% 145
Electrum 250 6,045 5% 330 647 1.4% 332
Gold 500 2,866 2% 647 316 0.7% 732
Platinum 1,000 1,198 1% 1,096 274 0.6% 1,253
Mithril 2,000 127 0.106% 2,542 93 0.208% 2,393
Adamantine 5,000 28 0.023% 6,500 14 0.031% 6,500
(extrapolated) 10,000 12 0.010% 12,000 6 0.013% 12,500
(extrapolated) 20,000 3 0.003% 25,000 2 0.004% 26,667
(extrapolated) 40,000 1 0.001% 40,000 1 0.002% 40,000

Note: Thanks to Tory for pointing out that the thresholds are off by 1. Minor effect on the math. Can’t trust everything you read! 😉

Attributes/Stats for Characters in Old School RPGs

One comment I got about Tunnels & Traps was surprise that I used Wisdom instead of Intelligence. That was definitely atypical of 1970s RPGs (which were my influences in 1980).

Inspired by D&D, the typical attribute-based RPG had Strength, Dexterity, and Intelligence. A majority had Constitution and Charisma. And they all had 6 to 8 stats:

D&D EotPT T&T B&B C&S Traveller RuneQuest
1974 1974 1975 1976 1977 1977 1978
Attributes 6 6 6 8 7 6 7
Strength 100% X X X X X X X
Dexterity 100% X X X X X X X
Intelligence 100% X X X X X X X
Constitution 86% X X X X X X
Charisma 57% X X X X
Wisdom 43% X X X
Appearance 14% X
Bardic Voice 14% X
Comeliness 14% X
Education 14% X
Endurance 14% X
Luck 14% X
Power 14% X
Psychic Ability 14% X
Size 14% X
Smell 14% X
Social Standing 14% X
Speed 14% X

OK, so here’s a system that’s Old School and evocative of every one of these: Bardic Voice, Luck, Psychic Ability, Smell, Social Standing, and Power. (Plus Strength, Intelligence, and Dexterity – of course.) Clearly it’s a game about singing, psionic ants.

Update: For comments, check out the Reddit thread, Attributes in 1970s RPGs.

Provide Your Players with Situations Rather than Scenes

The best book I’ve read on gamemastering so far is Matt Shea’s The Return of the Lazy Dungeon Master. The key philosophy is to prep only those things that you find hard to improv. Session prep typically follows this template:

  1. Characters
  2. Strong Start
  3. Scenes
  4. Secrets and Clues
  5. Fantastic Locations
  6. Important NPCs
  7. Potential Monsters
  8. Potential Treasure

While still following Matt Shea’s advice, instead of “Scenes”, I’ve relabeled that section in my notes “Situations”. Here’s a good example of why.

Last night my PCs encountered a band of 19 orcs: I had prepped interesting terrain for a battle or a staged retreat, plus I had written a sentence about the leader’s motivation. The players could engage with this situation in any of the Three Pillars of roleplaying: exploration, social, or combat.

One of the PCs had ended up in prison at the end of the prior session (he had been possessed by a ghost and then, possessed, tried to commit regicide). So his player introduced a new PC, whom I knew nothing about beforehand: a 16-charisma dragonborn archeologist.

Heck, I didn’t even know there were dragonborn in this world!

That provided me with a good reminder that roleplaying is co-creation between players and the DM, even in a homebrew setting.

While I never would have expected an archeologist, the overall campaign arc is about exploring ancient ruins to learn why the empire fell, so that was a good call on the player’s part.

As mentioned, an hour or two into play, 19 orcs appear on the horizon. Selma, our naïve half-orc PC, wants to approach the orcs but gets talked out of it. The dragonborn player, whose regular PC has by now been broken out of prison, decides to make a point of how evil the orcs are and sacrifice his new character by marching him off to the orcs all alone.

Off the dragonborn goes. But he rolls a 24 on a Persuasion check!

All I knew about the orc leader, besides his name, Jomongen, was that it was his motivation to study the old empire to establish a new one. So now Jomongen wants to bring the architect back to his citadel to explain his land’s role in the old empire!

With this unexpected new ally, suddenly secrets and clues (again from Sly Flourish’s guide for session prep) now end up being volunteered by their new orc ally instead of discovered in other ways!

And I accidentally have the orc leader talk about the dragonborn’s tail, so now dragonborn have tails in this world. (That’s canon now.)

So, yeah. The DM’s job is to provide *situations*. The players’ decisions will create the scenes.