The World's Largest Rewrite: Trepan Lizards, Grues, and Carving Up the Stone Theme

Last time: Grey Horse, Devil Swine, and Normal Humans

#109: Skeleton. We’ve already met our mummies, zombies, wraiths, and wights, so we have a pretty good number of undead knocking around the prison. Skeletons, like zombies, make sense as minions of more powerful undead. We placed the zombies with the mummies, but let's save the skeletons as servants of an as-yet unseen monster. I’m going to place them near the salt water aquatic monsters, to provide some variety there. The bones of failed escapees, animated by some dark magic, drifting up from the depths amidst the seaweed, to clutch at ankles, join us, join us... 

#132: Whale. Yeah, OK, let’s do this. OSE includes a killer whale (AKA orca), narwhal, and sperm whale. Rolling randomly gives us the sperm whale. A brief aside: It is interesting to see how OSE stats suggest a different concept of the world than a modern D&D system. A sperm whale has 36 HD and 162 (!) HP. A red dragon, the strongest of the chromatic dragons, has 10 HD and 45 HP in OSE. 

Compare that to 5E.2014, where an adult red dragon has 256 HP. The 5E MM doesn’t stat whales, but the roc is similar in HD to the whale in OSE, and that is about even with the dragon in 5E. It feels like modern designers felt it was important that the dragon was not only the most dangerous monster, but also (among) the biggest monsters. No such compunction in OSE.

The sperm whale is so big that it is hard to place it in the dungeon; it is very much intended as a wilderness monster. So let’s embrace that idea and put it outside the dungeon, but strongly associated with it. The sperm whale’s signature abilities are to ram ships and to swallow creatures whole. The prison builders enchanted this creature to crush ships approaching the island prison, and swallow any survivors in the water. Their intent was to prevent both prisoner escapes and rescue attempts from the outside. The whale itself can serve as a mini-dungeon, with survivors living inside the whale; that’s a classic trope. 

#41: Gargoyle. The flow of water is already a consistent theme within this dungeon, so let's include gargoyles that honor the origin of the term. Historically gargoyles were not just statues (or “grotesques”) but specifically rainspouts on the buildings they adorned. Our gargoyles are bound by the dungeon builders to channel water through the dungeon, for hydration, drainage, and sanitation purposes. They may be cruel and twisted monsters otherwise, but their commitment to the routing of water takes priority over monstrous behavior. Two gargoyles can form a 1’ diameter “pipe” through solid stone between their locations. PCs who parley with them could use their help to bypass various hazards and obstacles. Beware their betrayal, however, as they could easily send you trouble instead.

#115: Stegosaurus. The OSE entry for the stego is minimal, but the Monster Overhaul, as usual, proves its value with supplemental random tables. Rolling on a few of those we get “fleeing a predator” as a response to the prompt “why are they charging?” and “we steal and devour their eggs” to answer “what do we use them for?” We’ll place these guys near the sabre-tooth tiger, since they also fit the lost world theme. But we also want to build up the idea that the dungeon’s divisions are breaking down, and that creatures from different regions are interacting with each other. So we’ll say these stegosauruses are frightened by the chimera, and that the noble and his retinue are stealing and eating their eggs.

#23: Cockatrice. It is interesting that cockatrices and basilisks, at 5 and 6 HD respectively, are quite a bit bulkier monsters in old-school D&D, while modern D&D presents them as relatively junior monsters, compared to the medusa and gorgon in the Family Stone. The cockatrice is described by OSE as “small,” but is about as tough as an ogre. The monster overhaul notes they are “rarely viscous, though their feeble minds and tendency to panic can lead to disasters” (this also serves as a good description of my cat). As discussed when we placed our medusa, petrification is an incredibly useful ability for the fantasy prison warden. Petrification can be a way of safely transporting a prisoner; putting them in “solitary”; managing a shortage of supplies by temporarily (unless…?) reducing the percentage of the prison population that needs to, you know, eat; and even creating statues that can fit together to build walls. You know, like in Unico in the Island of Magic.

Uh, you know, Unico in the Island of Magic? OK, maybe not everyone has seen Unico in the Island of Magic. You can watch it on Youtube! The petrified-people-used-for-building stuff starts around the 30 minute mark. I watched this when I was very, very small, and for years afterward, I had a very deep and paralyzing fear of being turned into a statue and used as building material. Fortunately, this never happened to me in real life, and I eventually got over that fear. But this probably does explain why I have written about gorgons so much on this blog. These blog posts don’t typically include processing childhood fears, but there ya go.

What were we talking about? Oh yeah, the cockatrices are here for the same reason the gorgon (medusa) is – petrification is really useful to the prison builders. For the same reason I don’t want to cluster all the undead together, I’m not going to put the cocatrices next to the other petrification creatures. I’ll drop them near the center of the map, next to the doppelgangers, who were the second-ever monster we placed during this exercise. If the cockatrices petrified the doppelgangers, they might have formed a feral colony near that area. The zombies are unaffected or indifferent to their presence, and the bull sharks are protected from the stony birds by water. 

#4: Basilisk. What a serendipitous result. We have our oops-all-rocks enemies showing up back to back. In Soviet Russia, rocks pet you. 

Combined with our gargoyle in this entry, and our gorgon (medusa) lore, I think we have some interesting ideas of how petrification played a major role in the formation and maintenance of the dungeon. I like the idea that the prison-builders started with cockatrices (relatively easy to control, insofar as they have to make physical contact to petrify). Later, as the prison grew both larger and more difficult to control, they added basilisks, in a desperation move (capable of petrifying at range, but a real double-edged sword for other sighted creatures).

Rolling randomly, I’m gonna put the basilisk on the left side of the map, and it lands right next to the… rock baboons… which feels like the dice trying to tell me something. We’ll keep our scream-oriented baboons but also make them literal rock baboons. Why not? 

#15: Bugbear. Another minimally described goblinoid in OSE. Hairy, ungainly, attacks with surprise (this also serves as a good description of my cat). Going to the Monster Overhaul directs us to the… ogre entry, surprisingly, but I respect it. The Overhaul is not afraid to create Weird Problem Creatures, and the bugbear is definitely one of them. Its abilities are as follows: “Harmed by innocence. Attacks against the [bugbear] take a penalty equal to the attacker’s HD or level. Bonuses to damage become penalties. Toy swords cut as steel ones.”

That’s kinda mental, right? Certainly more interesting than “stronk goblin with +2 to surprise, move along.” You can imagine why a creature that fundamentally subverts violence would end up in our prison. However, by imprisoning this freak amongst mostly guilty inmates, the jailers have inadvertently made him powerful. None of the twisted reprobates in the prison would be able to harm this guy. I’m not trapped in here with you, you’re trapped in here with me. This is a cool puzzle monster. Clearly our bugbear is a dungeon boss, like the dragon and the devil swine, controlling other monsters/NPCs, because they cannot harm him. We’ll drop him near the bottom of our diagram; he’s probably bossing around some combination of hobgoblins and humans. Maybe the poor cyclops too.

#69: Lizard Man. The most evocative bit from OSE is obviously “Man-eaters: Kidnap humans and demihumans, whose flesh they regard as a delicacy” (thankfully, this does not serve as a good description of my cat). 

To my surprise, there is no entry for lizardmen in The Monster Overhaul. Not even an index entry pointing to another monster. I guess lizardfolk fall somewhere between kobolds and troglodytes in the Overhaul taxonomy. We’ll use troglodytes. With four separate random tables, they have enough going on in their entry that there will be material left for the actual troglodytes when we eventually roll them up.

Rolling a few times, we get “Genetic offshoot blessed with strange powers but cursed with mind-rot” and “Trepanning drill. Improves divine access” as a treasure. These two hooks together really sell these guys. The basic OSE beastiary does not have a direct mind flayer equivalent, so for this dungeon, our lizardfolk fill that roll.

#105: Shadow. “Intelligent, incorporeal (but not undead) monsters that look like shadows. Able to slightly change their shape.” I like the non-undead shadow concept. The Overhaul’s index suggests “Grue” as an interpretation, which isn’t on our OSE list, so we’ll use that to embellish these fellows. The grue is a highly mobile, durable, deadly creature in darkness, with light as its only weakness. It is easy to imagine our prison wardens trapping a grue/shadow here with redundant “walls” of light. As the prison keepers' authority slips away, the lights begin to go out, and the grue probes its containment, restless to be free. We’ll probably create a “high security” block for creatures like this, that require special methods to imprison. 

#45: Gnoll. Gnolls have a little bit more going on than some of the other humanoids, but not much; the OSE description is just a winking reference to their post-Dunsany presentation as a portmanteau of gnome and troll. The Overhaul has no gnoll entry, but refers to the hyena entry, which has a nice “legendary abilities” table. “Hyenas only eat the corpses of unholy people. The souls of those they consume stay inside the hyena.” This gives them a pretty clear reason to show up in our dungeon. The gnolls have broken into the prison in search of "unholy" souls to devour. We’ll assume they’ve entered the prison from the top and are trying to get toward the lower middle, where the humans are.

Here’s our updated diagram. Getting a little more crowded, isn’t it? Remember, this is not the dungeon map; just a very simple diagram that can eventually inform regions, which can in turn serve as the basis for an actual map.

Oops-All-Backgrounds D&D

I have written before about the underrated design of the backgrounds in D&D 5E.2014. Backgrounds in 5E include a brief description and some “suggested characteristics,” but mechanically they boil down to the following pieces: 

• Two skill proficiencies
• Usually one or two tool proficiencies (although a few backgrounds don’t get any)
• Usually one language (although again, a few backgrounds don’t receive one)
• A brief starting equipment package
• A “feature” that doesn’t have an explicit mechanical application, but exists entirely as something to be leveraged by the player and adjudicated by the DM. Usually the feature grounds the character within the game’s social milieu.

This is a solid chassis for a starting character, even without ancestry and class. The problem with backgrounds in the context of 5E is that they are quickly swallowed up by class progression. Those background details may come up in the first few sessions, but nobody bothers to use the Rustic Hospitality feature by level 5, let alone level 20. 

But… what if they weren’t competing with the classes? What if instead, the classes were designed like backgrounds? What if... the fighter class just looked like this?

1. Skill proficiencies: Choose two, just as you normally would, from the 5E fighter skill list
2. Armor proficiency: Choose one from among light armor, medium armor, or heavy armor
3. Weapon proficiencies: All simple weapons and any three martial weapons
4. Equipment: Same options as the in the 5E fighter list
5. Feature: Choose either Action Surge or Fighting Style

And what if the dwarf ancestry just did this? 

• Weapon proficiencies: Battleaxe, handaxe, throwing hammer, and warhammer
• Tool proficiencies: Artisan’s tools of your choice (smith’s tools, brewer’s supplies, or mason’s tools)
• Languages: Dwarven
• Feature: Choose either Darkvision or Dwarven Resilience

Any one of these is an MVP (minimum viable PC). Where you go from there depends on what kind of game you want. If you want to hew closer to normal 5E, layer on additional features and mechanics as class advancements, but keep it much leaner than typical 5E. Use GLOG-style progression as a model, with maybe four levels, capping characters at a low/middle power level, just like you might when using a 5E chassis to run a megadungeon.

If you want to keep it simpler, instead model the game on a system that features little or no advancement, like Troika, or one of the Into the Odd-based games. Or assume that advancement comes mostly from exogenic acquisitions, as in an equipment-focused game like Knave.

Class, Ancestry, Background — Choose Two, Drop the Other

Old-school games sometimes leverage the concept of race-as-class (also known as ancestry-as-class). An elf or a dwarf could be a character choice functionally equivalent to (and exclusive of) a class like cleric or magic-user.

Later editions of the game separate ancestry and class as distinct categories on separate axes; classes are defined by advancement, while ancestry is (usually) inherent and unchanging. But ancestry-as-class maintains its appeal in some OSR systems, as it simplifies character creation and makes it easier to “play up” what is distinctive about a non-human character. And other games have tried to find a happy medium between these approaches.

In that spirit, here’s an alternative way to split the difference between modern and old-school; pick two out of three among ancestry, class, and background, and just drop the third.

Ancestry and class, but no background. This is already implicitly pretty common in D&D. D&D 5E’s backgrounds are one of its better game design structures, but many players pay them little heed. They choose a background at character creation in order to pick up an extra proficiency or two, then forget about them soon after. Ancestry plus class, with no background, just formalizes this implicit choice. Whatever this PC did before the dungeon, it isn’t relevant to their new life of adventure.   

Both are dwarves, but their backgrounds are very different.

Ancestry and background, but no class. This is something like ancestry-as-class in old-school play, but with a background to give the character some more texture. Part of the appeal of ancestry-as-class is that it can take powerful abilities like darkvision, underwater breathing, or even flight and cordon them off from complementary class choices. Adding a background helps distinguish one dwarf from another, and give them a bit more personality. It’s easier to put some more mechanical weight on backgrounds too; one can extrapolate from the flavor text and ribbon abilities of the 5E backgrounds and imagine ways they would be more prominent in play without classes sucking up all their oxygen.

Class and background, but no ancestry. Obviously a character still has an ancestry. This choice simply means it isn’t mechanically relevant. In old-school D&D, this is something like what a human fighter or human wizard was; it was just taken for granted that humans were a “blank” in terms of ancestry, and possessed no special powers. But in modern play (or in games that mix play styles), that is worth reconsidering, because those games have tended to give humanity some mechanical heft (for example, a bonus feat in 5E). It takes a little extra work to reason out what an elf with no elf powers or a dwarf with no dwarf powers looks like, but I think it is possible. For example, a class-plus-background character may nominally be a dwarf, yet does not possess signature abilities like darkvision because they grew up on the surface, or in a subgroup of dwarves who otherwise just don’t naturally have darkvision.

Stolen Board Game Mechanic: Add Value to the Unpicked Choice

Spots is a push-your-luck dice game. Players take turns choosing action cards, which dictate how they roll dice that turn. After they take their turn, the action card they chose is exhausted for the round. When all but one action card has been chosen, all actions are refreshed and the process begins again. This is arguably a form of Dutch auction, although with accruing value instead of decreasing price.

Pretty straightforward, right? But there’s a small rule that is secretly important to the game. The final action card that was left unselected gains a token that allows for a one-time reroll of the dice. Whoever picks that action card next will get to keep any tokens that have accumulated on the card, and can spend those tokens later. So if the players favor certain cards over others (and most players will), the unpopular card(s) will gradually accrue additional value as more tokens accumulate.

This mechanic is great because it automatically "balances" the perceived power level of the options available to the players. It doesn't matter if that’s an actual play design imbalance, where some choices are better than others, or simply a preference among the players for a certain style of play. The imbalances between the choices are self-correcting over time, as some quantity of tokens will eventually bring a less-popular choice into competition with the more popular ones.

Consider applying this mechanic to discrete, mutually exclusive choices in an RPG. For example, consider downtime for a party of three PCs in a fantasy adventure game like D&D. Each time they return to town, each PC can choose one of the following (in addition to resting): level up, gather information, carouse, research, or shop. Only one player may choose each action, and each player only gets one choice.

Player A chooses leveling, player B chooses research, and player C chooses shopping. Gather information and carouse go unchosen, so they each get a token that allows for one in-session reroll of the dice. Whoever chooses these options in the future gets to keep any tokens associated with that choice. Even if carousing or gathering information goes several downtimes without being chosen, the accumulation of tokens will eventually compel someone to take them.

The 5E Megadungeon: Death, Magic, and Cats

Last week: Running a Megadungeon Campaign in D&D 5E

Last week we covered darkvision and laid out the factors that make 5E insufficiently lethal for typical megadungeon play. Now lets discuss some solutions.

Tailor monsters to make them punch harder, but also die faster. Tweaking monster stats is categorically easier than getting PCs to buy into house rules that lower their own character’s powers. In addressing a common complaint with 5E that combat tends to drag on, I’ve found it helps a lot to make monsters hit harder, but be less tough. With some relatively modest adjustments, it is easy to cut an eight-round slogfest down to a tight three-round nailbiter. Monsters leave a bigger dent in PC HP totals, but PCs also have the satisfaction of taking them down before the battle gets boring. 

Cap leveling up somewhere between levels 8 and 12. D&D 5E is tuned around the first two to three tiers of play. The second tier ends at level 8 and the third ends at level 12. The game is fun in the fourth and fifth tiers, but parts of it break down, and it is certainly not suited to megadungeon play. I ran several hundred sessions of 5E in a game that went from level 1 to beyond level 20, and while that campaign worked for location-based play at low levels (including quite a bit of time in the Caverns of Thracia), it was essentially obligated to transition to scene-based play at high levels.

A megadungeon campaign really needs to stay location-focused for its duration, and the easiest way to make that happen is to agree at the outset to cap the PCs’ level. You can choose where you want to cap the level based on which dungeon-bypassing powers you really want to limit. A series of posts from the early days of my blog attempts to catalog which 5E powers bypass dungeon obstacles at various tiers.

This will take some buy-in from the PCs, but make the case that it is required to run a cool megadungeon in 5E. Capping progress doesn’t even require an explicitly old-school perspective. The idea of “E6” D&D (which caps progression at level 6) came out of the crunchy 3.5 D&D world all on its own. And maxed out PCs can become powerful figures in the local area, engaging in domain-level play. If the players still aren’t convinced, a megadungeon campaign isn’t right for them anyway.

Limit where the PCs can rest. The average modern-style party will gravitate toward a simple gameplan: Fully unload on any antagonists they encounter, then pass out on the spot for eight hours. There's some variance based on group composition – a minority of classes in 2014 5E are optimized around short rests – but most 5E groups will have a majority of long-rest-oriented PCs like wizards and paladins who want a solid eight hours of sleep so they can once again go supernova on the next monster that looks at them funny.

You are going to have to disabuse them of the idea that it is OK to rest in the dungeon. For a lot of players, it’s going to take some convincing.

Out of character, tell the players that resting in the dungeon or the surrounding wilderness is highly dangerous. Have NPCs reiterate this in-character. Ultimately, the PCs will attempt it anyway, and you should adjudicate consequences firmly, demonstrating how hard it is to get a good long rest in the dungeon. Of course, if the players take clever precautions to secure a long rest in the dungeon, reward them.

Finally, if you don’t think this will be enough to motivate your players, discuss a house rule at session zero that long rests are simply impossible inside the dungeon. I’m trying to be conservative with the house rules here, but this one may be worth it. 

Leverage time and antagonists against long rests. If resting in the dungeon isn’t practical, most PCs versed in modern-style play will pragmatically come up with an obvious fallback solution; retreat quickly to safety after every combat encounter. This is not really a bad thing; smart OSR PCs will keep avenues of retreat open as well. 

But retreating after every fight will slow the game to a crawl. Fortunately, both old school games (strict time records and faction play) and story games (clocks and fronts) offer some tools to incentivize modern and trad players to play differently. 

Establish antagonist NPCs and factions early, and then telegraph to the players how they are advancing their agendas every time the PCs take a long rest. It may help to present the PCs with antagonists right from session zero. A good example is the conceit used in Electric Bastionland and other games; have the party start with a shared debt they have to pay off. It could be one of the powerful factions in the town, or in the dungeon itself. The important thing is that the debt-holding faction both has a reason to be  antagonistic toward the PCs and methods for creating time pressure. 

Magic is ridiculously abundant to the point where you solve most of the normal OSR challenges with cantrips that half the party have.

I agree with this in a general sense. Cantrips are one of my least favorite parts of 5E, and they trivialize many parts of the game that old-school play emphasizes. 

But the problem is really limited to a small subset of cantrips. The biggest use-case of cantrips is essentially providing a DPS floor for full spellcasters in combat. I don’t enjoy this design decision, but it fits with how modern play handles combat, and we don’t need to change combat much to empower the megadungeon experience.

The genuinely concerning cantrips are the ones that trivialize challenges outside of combat. If I was running a megadungeon in 5E, I would modify or rule out a few cantrips:

• Light and Dancing Lights would be the obvious ones to ban or nerf by “promoting” them to first level, per the discussion of visibility and darkvision in the previous post.
• Mage Hand should probably receive the same treatment, given how useful it is for manipulating traps and doors without risking oneself. It may still be worth the spell slot even if “nerfed” to first level. If a player is interested in the Arcane Trickster archtype (a rogue subclass), you may need to negotiate with them how to interpret this adjustment, as Mage Hand is baked into that archetype’s core powers.
• Guidance is not causing a problem for a megadungeon specifically, but it is bad game design, so I would probably ban it if I was cutting other spells anyway. 
• Minor Illusion is a consideration, although strictly adjudicating it can denude it of its worst applications. 

The rest of the cantrips in the 5E.2014 PHB are not really disruptive to megadungeon play. Non-cantrip spells are a resource expenditure question, and are essentially covered by the time pressure tools discussed above. Sure, having access to Fly or Dimension Door can subvert some dungeon challenges; but these are precious spell slots if we cap the PC's level somewhere between 8 and 12. 

So yeah, a 5E group is going to move through the megadungeon more quickly and suffer fewer casualties than an equivalent old-school group. But they're not going to trivialize a well-run megadungeon. 

Cat People???

Running a Megadungeon Campaign in D&D 5E

A post on the 3d6 Down the Line Discord expressed skepticism that the Arden Vul megadungeon could be run in D&D 5E. 

And I get it. D&D 5E would not be my first choice for a megadungeon either. 

But I have run a lot of 5E with an OSR mindset, and I believe I could run a megadungeon in 5E with (relatively) modest house rules and campaign assumptions, if for some reason I decided to do so.

All points below are in the context of the 2014 5E rules. I do not own the new 5E books, but understand they do not deviate far from the 2014 rules, so I expect this approach would broadly work there as well. I also anticipate that most of the same ideas would apply to the 5E-compatible systems that have come out since 5E was released into the Creative Commons.

Below are the four issues that the post identified, addressed in turn.

Everybody Has Darkvision

I agree that darkvision is over-prescribed in 5E, and that darkvision can undermine old-school exploration by removing the question of visibility. But a serious part of the problem is that a lot of players and DMs don’t even follow 5E’s rules as written, and assume that darkvision is a more potent ability than it actually is.

D&D 5E’s rules allow creatures with darkvision to see in darkness as if it were dim light. That means disadvantage on Wisdom (perception) checks and an inability to see colors. Darkvision is better than nothing, but it is no substitute for a proper light source, particularly when checking for traps or keeping an eye out for secrets and treasure. I also remind players that whenever they are within the area of a source of light – whether from an ally, the environment, or an NPC or monster – that light prevents the use of their darkvision until they move out of the light. When I explain all of this to 5E players, they often choose to use light, even if they don’t “need” to, treating darkvision as more of a plan B, or an option for stealthing apart from the group.

Monsters face the same limitation. Whether an intelligent monster decides to rely on darkvision or use light typically reflects its level of confidence in its place in the dungeon hierarchy. Those confident in their control of the space use light. Those fearful of discovery favor the darkness.

Of course, an intelligent monster with 120’ darkvision will rely on the darkness more often than a creature possessing typical 60' darkvision, because it expects that it will have an edge. Creatures with tremorsense, blindsight, and similar abilities actually can functionally “see” in darkness as well as they can in light, so they do work in the way that many players think darkvision works. Creatures with those senses actually will completely skip light, for the most part. This makes them significantly scarier opponents in their native environment than creatures with mere darkvision.

If I was going to go further in houseruling this issue, I would take darkvision away from elves, and leave it to just the gnomes, dwarves, and tieflings. But even without altering the ancestry rules, darkvision can be brought into check simply by following the rules as written strictly.

It’s Impossible to Die in 5E

Let’s start with two easy caveats. First, simply removing 5E’s playculture presumption of level-appropriate encounters solves part of this problem. Even the most optimally constructed low-level 5E characters are not going to last long if they arrive at Arden Vul and beeline for the lair of Craastonistorex, the old and powerful green dragon. Once the players realize that difficulty is dictated by where they go and what they do – not what is appropriate for their current level – they will act more prudently.

Second, at very low levels, the problem isn’t really that pronounced anyway. Low-level PCs in 5E are much tougher than B/X or OSE characters, but they can still go down after just a few hits. The death saves system usually gives them a few chances to survive, but a deadly dungeon can kill many PCs outright through massive damage from falls and traps. In combat, monsters can also opportunistically focus on downed characters and quickly finish them off. Remember that any source of damage to a character on death's door equals a failed death save. Lowly goblins or kobolds become much scarier when they drag a downed PC away into the darkness, rather than “fighting fair” and engaging the PCs who are still standing.

But beyond those two caveats, I acknowledge the issue OP raises. At about level 5, 5E PCs get a lot stronger, and they don’t slow down from there. PCs in 5E at middle to high levels create a series of interrelated issues for DMs who want to run a megadungeon game that cares about exploration, time, and resources. I believe there are at least four interrelated issues that cause problems here:

• 5E PCs have massive amounts of HP
• 5E PCs have a lot of resources to replenish HP
• Both play culture and player powers make it unlikely that PCs will get lost, captured, or otherwise separated from safe locations where they can rest and recover 
• Modern play culture presumes little or no time pressure, so choosing to rest does not come with an inherent cost

A few different adjudication techniques and house rules can solve a lot of these problems. We'll cover those next week.

Next week: Death, Magic, and Cats

Rolling for Shoes and Quantum Gaming

I greatly enjoy "quantum" game mechanics. In this context, all that means is some aspect of the game that would typically be predefined is left undefined until its definition is specifically required by the events of the game. Various games use quantum mechanics without naming them, but one of my favorite articulations is this one by Luka Rejec. 

In a pinch, I would probably use Roll For Shoes by Ben Wray, or freeD6 down the line by Liche’s Libram for a quantum chassis. I particularly like the idea that basically anything can be a character skill (allowing for quantum character construction and a negotiated understanding of the game scenario), but that the randomness of the dice ensures that characters have choices and risk/reward calculations; they can’t just add skills indiscriminately, or they’ll end up with low ones that the GM can force them to test. These games also require nomic negotiation between the players to decide how to interrupt unclear overlaps and edge cases.

I do want to try running Roll For Shoes as-is, but I have also put some thought into a slight variation on it that makes it less of a d6 die pool system and lets it tack closer to D&D abilities and d20 resolution. All you need are 3d6 and 2d20, as well as a way to take notes. So you can run this game entirely from your phone, if needed. The game requires at least one GM and at least one player, although it could be played solo with an "oracle" replacing the GM.

The procedure is as follows:

1. The group chooses a genre. The GM should be at least as familiar with the genre as the player or players are. 
2. Choose a goal. It should be very general, but within the confines of the genre. Don’t think too much about who the characters are yet. A goal could be “defeat the evil overlord” or “win the reality TV show” or “steal the huge diamond from the gallery.”
3. Each player chooses a very minimal character concept. This should be as bare-bones as possible. It could be an occupation, a background, or something else. Basically a one-sentence premise, just enough to explain why this person is present in the opening scene.
4. The GM begins the game. Play starts with an opening scene, usually something that will bring the characters together (if there is more than one) and either establish how they know each other or allow them to meet for the first time.
5. When a PC has to do something uncertain, they roll 3d6. This becomes their permanent ability stat (or skill, or whatever term you prefer) for that action.
6. Whenever a player needs to test that stat (including immediately after that first 3d6 roll), they roll d20. A result equal to or lower than their stat is a success. The GM may give them advantage or disadvantage on the roll.
7. If a player rolls something really low on 3d6, they can choose not to test it and can try to approach the situation in a different way. But the result of that 3d6 roll stays on their character sheet.
8. Failed d20 rolls create new antagonists, hazards, obstacles, complications, or other threats. This could be anything in the story that the GM can use to trip up the players. The GM writes it down, along with the number that the player rolled that resulted in the failure. So if the player has an Argue stat of 13, and then rolls a 16 on a test when trying to convince the studio boss to greenlight their movie, the GM notes “Studio Boss: 16.” 
9. Whenever a number associated with a threat is rolled by anyone, that threat can reappear. It doesn’t matter if a different player rolls the number, or if the PC whose failure led to the creation of the threat is present or not, or even if the roll in this instance was a success (because it is rolling against a higher stat). If the GM thinks it makes sense for the threat to appear, it can happen.
10. The story ends whenever everyone thinks it has reached a natural ending. Or…
11. Alternately, the story can end when the game reaches a predetermined number of one or more of the following criteria:
1. Successes. X successful checks against stats are enough to complete the goal.
2. Failures. X failed rolls on stats are enough for the goal to fail.
3. Threats. X threats created are enough to ensure the PCs will fail to complete the goal.
4. Stats. Each character can have only X stats, maximum. Once they’ve reached this maximum, their next failure will knock them out of the story in some fashion or another. If all the PCs are knocked out, they fail to complete the goal.

I'm keeping this in my back pocket in case I need a quick, improvised RPG. If I have an opportunity to try this, I will report back.

The Rival Taxonomies of D&D Magic

D&D’s origins were full of contradictions. The game was defined by strict rules, but open to liberal interpretation. Top-down design versus a strong DIY ethos; high fantasy versus science fantasy; Appendix N versus Hammer horror. It shows up in so many aspects of the game. 

Consider for example how spells are named in D&D. They can be grouped into two broad categories. 

Detect Magic, Locate Object, Comprehend Languages, and similar spell names have a technical nomenclature. The spell names quite literally explain what the spells do, in plain language. This is magic as technology. This is one of the two taxonomies of the game that goes all the way back to the beginning, appearing in the context of magic items as well. 

Now consider names like Hellish Rebuke, Crown of Madness, Eyebite, and almost all of the spells that include proper names, like Tasha's Hideous Laughter. These have a mythic nomenclature. The name still relates to what the spell does, but in a much more evocative, figurative, or culturally mediated way.

Now imagine that these rival naming conventions aren’t just an oddity of the game’s development. What if we infer there is an in-universe reason behind this distinction? Perhaps the technical nomenclature came from magic-as-science aliens, while the mythic nomenclature descended from primordial progenitors at the dawn of the world.

You could even group the spells accordingly and associate them with a new alignment axis, to replace the good-evil axis. Is your character lawful-technical, neutral-technical, chaotic technical, lawful-neutral, true neutral, chaotic neutral, lawful-mythic, neutral-mythic, or chaotic-mythic? Replacing good-evil with a different alignment axis to complement law-chaos can make for a much interesting milieu. 

Want to go even further? You live in a world of rationalist wizards and faithful priests. Wizards get all the technical language spells. Priests get all the mythic ones. That’s right, Regenerate and Remove Curse are technical names. Those are wizard spells now. Burning Hands and Cloudkill go in the other direction. These more embellished or ornate names are now cleric spells.

The World's Largest Rewrite: Grey Horse, Devil Swine, and Normal Humans

Last time: The World's Largest Rewrite: Floating Heads, Mother Fungus, Cellipedes

#60: Horse. OK, yes, I jumped into this exercise without really thinking about how to handle the various mundane “monsters” in the OSE bestiary. I don’t want to overlap with mules and camels, so I’m going to dive deeper into the Monster Overhaul, my manual of choice, to populate this one.

The book includes an entry for a “grey horse,” a strange constructed thing in the shape of a horse that challenges travelers to make clever rhymes, eating their rations as punishment if they fail. The grey horse seems too benign to be a prisoner, and too capricious to be one of the jailers. I’ll treat it as an invasive species, but without explaining how it got here (the grey horse just shows up in places where it is not wanted).

#84: Normal Human. Ah yes, the most dangerous monster of all; the Normal Human. OSE defines them as “Non-adventuring humans without a character class. Artists, beggars, children, craftspeople, farmers, fishermen, housewives, scholars, slaves.” But they are also implicitly not bandits, or pirates, or nobles, or any of the other “monster” types in the OSE bestiary that are obviously humans, but have their own entry. Essentially this is what the modern game would call a commoner.

I think we have some Normal Humans here who are guilty of Abnormal Crimes. They’re probably not individually dangerous to adventurers; even the typical serial killer is more of an opportunistic-but-ordinary person, rather than someone with high levels as an assassin or something. These Normal Humans are like the prisoners in Con Air or the mooks in the Batman Arkham games. They are weak, but numerous. And they likely have some powerful leaders among them as prison bosses; maybe NPCs with classes, maybe actual monsters.

#106: Shark. Shark!! The shark brings us back to the submerged section of our dungeon. We want to distinguish a bit from the other monsters that have flooded (literally and figuratively) into the prison. Picking randomly among OSE’s three sharks, we get the bull shark, which can ram and stun prey for three rounds. This is a nice twist that you don’t really see from beasts in modern D&D. I like the idea of the bull sharks ramming prey as they pass through a transitory space like a submerged hallway. The hallways are navigable for the sharks but too narrow for the sea serpent, who is the alpha predator in the seawater sector. Stunned swimmers sink deeper into the depths, so attempting to rescue them presents further risk for their allies. The crabs clean up what the bull sharks don't eat, at the bottom of the halls, amongst the bones of failed prison escapees.

#70: Lycanthrope. I usually choose a subtype randomly, but in this case I am going to simply pick the devil swine, because (a. they’re much more evocative than the other, more standard lycanthropes, and (b. they’re evil, so they’re the easiest to explain as prisoners of celestials. OSE describes them as follows: “Corpulent humans who can change into huge swine. Love to eat human flesh. Lurk in isolated human settlements close to forests or marshes.”

A devil swine has 9 HD (!) and a charm ability. So these guys are not minor brutes, but instead dangerous bosses, and with their charm ability, probably a powerful faction in their own right. I imagine they’ve been strategically charming other prisoners to take over the prison and eventually try to escape. Relative to some of the other very archetypal monsters we have featured so far, “shapeshifting mind-control pigs” could really surprise players.

Another nice detail on lycanthropes is as follows: “Horses and some other animals can smell lycanthropes and will become afraid.” The grey horse and the mules are both aware of the devil swine and could help the players avoid them, or at least anticipate their presence.

#22: Chimera. Another folklore classic. OSE doesn’t provide any suggestions beyond a visual description. The Overhaul gives us more to work with, including a roll table that produces a chimera with a goat for the left head and hindquarters, a leopard for the center head and forequarters, and a newt for the right head and tail. It breathes poison gas and has no wings. Created by a wizard who is also probably interred here.

#26: Cyclops. It’s interesting to compare the OSE cyclops to one from a more modern-style monster manual. The OSE version hews close to the Odyssey; it raises sheep, is slow-witted, and possesses the ability to curse people. All straight out of the Greek lore.

The 2014 5E monster manual, by comparison, shunts this information into the flavor text, abstracting it away from the source myths. Consequently, aside from its poor depth perception, the 5E cyclops has almost nothing to distinguish it from the statistically similar hill giant, which is a shame, particularly because 5E has an abundance of interchangeable brutes like this taking up space in the book.

The Overhaul parsimoniously groups the cyclops with the giants, so we’ll roll there to get some more of an idea of what to do with this dude. The “Why fight these giants?” table produces “They keep growing larger. Soon it won’t be possible to harm them.” So this cyclops was getting bigger and bigger with no end in sight, and the magic of the prison keeps that magical growth in check. The cyclops may even be a willing prisoner here, worried that the prison’s weakening structural integrity will reboot their uncontrollable embiggening.

#7: Beetle, Giant. The fire beetle (a fantasy firefly) and the oil beetle (a fantasy bombardier beetle) are the famous ones here, but rolling randomly tilts me toward the Tiger Beetle (a fantasy… uh, tiger beetle). OSE tiger beetles “hunt robber flies, but sometimes eat humans.” The bit about robber flies is useful, as we haven’t placed those guys yet.

The real-life tiger beetle has a number of gameable features we can steal, including antlion-like larvae that burrow into the sand to trap prey; an ability to charge very quickly toward prey, but with the need to stop and visually reorient; and the ability to mimic the sounds of toxic moths so that bats won’t eat them. We can tie these guys to both the robber flies and the bats when we get those results. 

#100: Rock Baboon. Once again I’m charmed by old-school D&D’s “animal, but slightly weird” approach, contra modern D&D’s harder division between mundane animals (lumped together in the back of the manual) and fantastic monsters. The rock baboon is a pretty straightforward monster per the OSE entry, but I do enjoy that they “communicate with screams.” Same, rock baboon, same. How far can we take that? 

Perhaps relative to other creatures in the dungeon, the rock baboons are particularly good at communicating important information over relatively far distances. The primary danger when encountering a single baboon or a small group is that they will alert the rest of their troop, even if they are far away. The baboons could even be useful allies if befriended, facilitating long-distance communication (filtered through baboon-speak, of course).

#134: Wolf. The most interesting bit about wolves in the OSE entry is that they can be trained, and that goblins ride dire wolves. So we have two possible routes here; wolves trained by the wardens to police the prison, and wolves ridden by the goblins we haven’t seen yet. The next entry better serves the prior option, so I’m going to go with the latter and assume these are goblin-affiliated wolves. We’ll put the wolves near the hobgoblins and leave the door open for a greater goblin zone in the prison.

#10: Blink Dog. Apparently we’re in the dog block. In addition to their signature teleport ability, blink dogs are lawful and hate warp beasts. I think it makes sense to consider them servants of the jailers. Their blink ability would make them well-suited to capture, corral, or pursue prisoners without the prison’s physical barriers limiting their movement. Perhaps they’ve been left to their own devices since the prison has gone to rot. A first encounter with the blink dogs will probably involve them shadowing PCs or observing them from afar to take their measure. They could be powerful allies for PCs who earn their trust by containing monsters or stopping escapes.

Frieren and What It Means to Play a Truly Ancient PC

Frieren, the manga and anime about the titular elf, is in many ways classic fantasy, using a lot of common fantasy vernacular to tell its story. A dragon is a dragon, a wizard is a wizard, and even a mimic can be used as a sight gag without explaining what a mimic is, and why it resembles a treasure chest. 

But Frieren also does several things that most classic fantasy does not, including taking seriously the idea that elves with extremely long lifespans would live fundamentally different lives than humans.

Of course, the concept of elves as long-lived beings is itself part of that common fantasy vernacular, and is drawn from Tolkien, who originated the idea of an "elf" as something distinct from the way the term was used in fairy tales, interchangeably with “gnome” or "fairy" and other depictions of magical fey folk. 

In The Fellowship of the Ring, when the members of the fellowship are talking about what is (to them) ancient history, Elrond can weigh in and say he was actually there. He witnessed those events firsthand.  This is a powerful way of illustrating the difference in how these people live. Frieren does the same thing, to great effect. 

But a lot of other fantasy derived from Tolkien tends to copy the aesthetics without incorporating the underlying worldbuilding. D&D rarely presents its elves as truly long-lived creatures, because elves are first and foremost PCs, and it is difficult to embed a PC in ancient lore, or think about what they might have been doing 300 years prior to the game's start.

But is there a way we could we create a more Tolkien/Frieren treatment for elvish PCs? I do not mean the “penalty to strength, bonus to charisma" kind of mechanics that D&D has sometimes applied to its ancestries. Instead, a simple layer of situational advantages and disadvantages could provide characters with hooks for understanding how their character’s age affects their place in the story. These could be literal advantages and disadvantages (applied to d20 rolls) or more abstracted tools for resolving situations.

Normal Lifespan

These creatures rarely live far beyond 100 years. The category includes humans, halflings, dragonborn, tieflings, and many other sapient creatures not otherwise known for long lifespans.

Disadvantage: Ignorant of history. You aren’t familiar with much of the world beyond your own experience. Unless you are a scholar or other specialist who has had a specific reason to learn something about the history of a person, place, or thing, you just don’t know it. You may know the history of your family over the past several generations, or the last century or so of events in the community where you grew up, but that’s about it.

Advantage: Unencumbered by the past. Longer-lived creatures do not expect you to know or adhere to customs, traditions, or obligations not expressly presented to you in any kind of formal social situation. A human among dwarves or elves can get away with a lot of behavior that those peoples would consider rude or even offensive in a venerable peer who should “know better.” A human or other normal-lifespan person can always throw off the obligations of country and clan if they choose to do so, and longer-lived peoples will simply view that as being the natural way of such short-lived people.

Extended Lifespan

These creatures can live to be several hundred years old. Dwarves and gnomes are the most well-known members of this category. Creatures of the land, of rock and stone, often belong to this category.

Disadvantage: Unforgotten Feuds. These people have long memories. Their lives are too long to allow for the quick passage of time from generation to generation to wash away disputes; but they are not so long-lived that such disputes will ever seem inherently trivial. A creature with an extended lifespan likely has at least one unsettled feud with a member of any large community they visit.

Advantage: Appeal to the Old Ways. In their dealings with other creatures as old or older than them, people with extended lifespans can always appeal to an alternate system of resolution to resolve a problem. Depending on the culture or polity where the dispute takes place, it could be trial by combat, an appeal to the gods, or something more esoteric. The important thing is that the alternative definitively predates whatever the normal, contemporary resolution would be to a dispute. 

Long Lifespan

These creatures can live hundreds of years, approaching 1000. Elves, of course, fit into this category. Other people who are not immortal, but whose infusion of magic lends them to greatly expanded lifespans, can fit in this category (for example, the druid’s high-level Timeless Body ability in D&D 5E fits this fiction well). 

Disadvantage: People are like leaves in the wind. Like Frieren herself, people with long lifespans struggle to form lasting relationships with others. When they travel to a place they haven’t been recently, at least one person, organization, or institution has changed since they were last here. Someone has died. The government has changed. The customs and culture are radically different. This will always take the long-lived person by surprise, no matter how many times it happens. 

Advantage: I was there. There is always a chance that something that seems ancient, secret, and powerful to the younger peoples is recent, obvious, and mundane to a long-lived person. Even if they didn’t personally witness an event or know a historical figure, they always have a chance of knowing things that no one else remembers.  

The World's Largest Rewrite: Floating Heads, Mother Fungus, Cellipedes

Last time: The World's Largest Rewrite: Dungeon Is Wet, Tortoise Foreshadowing, and 30-50 Feral Hogs

 #51: Green Slime. This could again be interpreted as a connection to the dragon, but I don’t want to make an entire section of the dungeon oops-all-slimes. I’m instead going to put the green slime(s) below the pixies – in both senses of the word. We haven’t been very specific about verticality yet, besides some broad strokes, but I’m thinking that the upper left corner (where the fresh water comes in from the "roof") is near the top of the dungeon, while the lower right corner (where seawater comes in from the ocean) is near the bottom. I like the idea that the green slimes are forming on the ceiling on the level below where the dragon is, as runoff from its breath attack, and moving laterally through some point of connectivity that could potentially be exploited by explorers. 

#116: Stirge. I’ve got a bit of an order of operations going for responding to these random rolls. Some of these immediately suggest a fit based on what we’ve already established about the dungeon. Others become clear once I check the OSE entry and notice some evocative bit of flavor, or an ability I didn't know about.

If I need more beyond those two steps, I’m going back to the Monster Overhaul once again. The stirge (“skeeter” in Overhaul terminology) has a table for reskins that are mechanically identical, but very different in aesthetics. I rolled and got “Rotting floating head. Lank hair, no eyes.” This immediately sparked some ideas. Perhaps some number of the prisoners in this prison were executed by beheading, and now those severed heads are mindless blood drinkers, wandering the halls. I’m going to place them near the wights, on the assumption that the execution chamber would be near those undead wardens.

#137 Yellow Mould. No entry in the Overhaul, possibly because yellow mould is more of a dungeon hazard than a “creature” per se. It is also possible that its functionality is captured by the lavish two-page spread for myconids. OSE doesn’t appear to have myconids, so we can liberally use the Overhaul tables to figure out what is up with this yellow mould.

Rolling on the “spore attack” table, we get “fungal curse,” which means that a creature failing the save will eventually sprout a “mother fungus,” forming a new colony. I like the idea that this fate befell a prisoner who was interred here, and a new colony formed in the dungeon.

Riffing off this idea further, I imagine these fungi are somewhat like the mycorrhizal fungi that allow trees to communicate in some forests. Rather than a purely parasitic dungeon hazard, I like the idea that the fungi are symbiotic, and are probably a major source of food, exchanging (non-toxic) edible mushroom growth with other dungeon denizens for things they want (especially water and fertilizer). Because tiny fungal filaments connect many parts of the dungeon, they can provide information or facilitate communication. We’ll place the mother fungus near the fresh water, but assume that their filaments have spread to many parts of the dungeon where there is at least some moisture and not too much heat.

#21: Centipede, Giant. A classic, flexible dungeon monster that can go anywhere. OSE notes they favor damp areas, but we need to narrow it down further than that (remember, dungeon is wet). The Monster Overhaul includes both a monstrous vermin category, and another section for ancient anthropods. Going with the latter because it has an intriguing “why fight these ancient anthropods” table, we roll a prompt that “one of them ate the key to this chest.”

I’m going to tweak that and combine it with the other tables in the entry that generate weird head-shapes for these bugs. These “cellipedes,” known colloquially as the prisoner’s best friend, have evolved key-like protuberances on their heads. They are drawn to places like prisons; the more locked doors, the better it is as a breeding habitat for these sickos. They can be used to open some doors, and particularly rare specimens have a skeleton key ability, and are able to open many locked doors. 

#99: Roc. I love that OSE has giant roc, large roc, and small roc. Given that the roc’s brand is “very large bird,” I'm kinda skeptical that three categories were required. We don’t have manual entries for “tall halfling” or “non-animated skeleton.” 

OSE notes that rocs are lawful creatures who react negatively to non-lawful creatures, and can also be trained as mounts. So let’s associate them with the prison builders. We haven’t yet decided pinned down the builders’ whole deal, but for the prison to make sense as an adventuring site, it helps to presume that their authority and control has partially or completely lapsed. 

To put a twist on the roc here, let’s make it a big egg. Not every monster has to appear in its fully grown adult form. And finding an egg is a classic sort of unusual “treasure” for PCs; a player in one game I ran took a deep interest in the unhatched egg of a giant carnivorous parrot, which became a focus of downtime work for many sessions afterward. We’ll drop it near the wights, on the assumption it has or had something to do with the builders/jailers.

#78: Mule. Another mundane animal. My first thought was to make them a population descended from working animals when the prison was built, like the wild burros of the southwestern United States, who descended from domesticated donkeys brought to the area by prospectors. Then I remembered that mules are, uh, by definition not the type of animals you’re going to find breeding in the wild.

So we’ll go with a more ordinary explanation, and say that mules are survivors of adventuring parties that have entered the dungeon. Some of them have gathered here to dwell among their own kind. I like the idea that mules regularly appear on the dungeon's random encounter table as well, with each mule encountered giving hints as to the status (or final fate) of the adventurers who brought that mule into the dungeon. Mules also have a few useful sundry items on them ("found a mule" is local dungeon slang for a stroke of good fortune; more dungeons should have custom slang). An amusing recurring motif is mule-as-evidence of a TPK. Somehow the lowly pack animal always survives.

#65: Kobold. Our first humanoid! OSE uses the old-school characterization of kobolds, noting they are “canine,” while the Monster Overhaul goes with a modern take, describing them as "reptilian" and dropping them in the “dragon” section of the book. I’m sympathetic to Skerples on this taxonomic decision, given how much more prevalent that portrayal is these days, and I'll go the same route since I'm sticking to OSE where I can. But I cannot continue without mentioning that the very good boy Kuro makes a strong case for the canine kobold. I love the Dungeon Meshi portrayal, especially because “dogfolk” never really clicked in D&D the way tabaxi did. It’s a minor gripe, but I always thought the dogfolk in Thracia were the least interesting of the beastmen faction members.

…What were we talking about? Oh yeah, the World’s Largest Dungeon. Kobolds, regardless of aesthetics, are known for being numerous and individually weak, so I don’t think they make much sense as prisoners. Instead, let’s imagine they have entered the dungeon while delving underground. Did they get here intentionally or accidentally? Rolling on the Overhaul tables for prompts, we get “geckotian” kobolds (“sticky pads, marbled eyes”) with a current activity of “prodding a corpse” and “bucolic mushroom farms” as a current scheme. I think this is already more interesting than just making them dragon servants and calling it a day. 

Let’s say they entered the dungeon seeking the yellow mould mother fungus. Prodding a corpse suggests they are corpse retrievers (and possibly even grave robbers) because they’re gathering fertilizer to bring to the fungus. There may be some tension with the mother fungus; the kobolds ideally would like to domesticate it, while the mother fungus wants to infect them. So they want different things and are in tension, which is a good scenario for the PCs to crash into.

#59: Hobgoblin. The humanoid hits continue. Modern D&D treats hobgoblins as martial warriors, something like how orcs were originally portrayed. OSE reverts hobgoblins back to their earlier presentation, but that doesn’t give us much to work with, as they are just “bigger goblins,” a trait they share with bugbears. At least bugbears have the element of surprise. The Monster Overhaul (correctly) just folds hobgoblins into the orc category, which we’ll save for when orcs come up in this dungeon. So that’s no help. Hobgoblins, hobgoblins, what do you do with those hobgoblins?

It’s a thin sliver of lore, but the OSE hobgoblin entry does note that thouls sometimes serve as bodyguards to hobgoblin kings. If you’re not familiar, thouls are an infamous monster from early D&D that combines aspects of hobgoblin, ghoul, and troll, and probably originated as a typesetting mistake.

Perhaps we can make our hobgoblins more interesting by playing up the connection to goblins, ghouls, and trolls. Let’s say that hobgoblins are themselves goblins who are particularly susceptible to mutation, something already implied in other treatments of goblins. Mutation has made them bigger, for starters, but some of them have also been able to mutate into traits from other creatures. We’ll hold further specifics of their mutative powers for a future monster that hasn’t been placed yet. I’m also not sure whether they are prisoners, trespassers, or something else. We’ll revisit that later.

#32: Driver Ant. Part of the fun of this exercise is looking at the stripped-down presentation of creatures in OSE. There’s a less-is-more vibe to these monsters. At first glance, driver ants are giant bugs with a standard bite attack and not much else to distinguish them. But on second glance…

• Omnivorous and rapacious: “Consume everything in their path, when hungry” – I feel seen.
• Morale: “Attack relentlessly, once they are engaged in melee (morale 12). Will even pursue through flames.”
• Gold: “30% chance of 1d10 × 1,000gp worth of gold nuggets, mined by the ants.”

So there are a couple of adventure vectors here. The ants are driven by hunger and can eat a lot of different things. We can imagine them chewing through barriers made from organic material, invading and connecting different regions of the dungeon.

The morale aspect is compelling for games that use morale rigorously. One of my biggest complaints with modern-style play is the strong presumption that every fight only ends when all the monsters are dead. When I run games, I stress how advantageous it is to compel monsters to flee or surrender, rather than slaughtering all of them because of video game logic.

Contrasted with our expectation of how morale may quickly end a fight against hobgoblins or kobolds or mules (please, do not fight the mules), a monster that goes into a 12-morale frenzy when you melee with it is a big problem. PCs who study their behavior could distract them with food, pelt them with arrows from a distance, or trick other monsters into fighting them. But also… the ants may have gold in their lairs. So there’s also a strong incentive to risk engaging with them further.

#135: Wraith. This is our first incorporeal undead. OSE notes that they “Dwell in deserted regions or in the homes of former victims.” I think this suggests that they are prisoners, perhaps murders or other capital criminals who persist after death, but are trapped in the "deserted region" of the prison. 

We had previously decided that our wights could be guards. Perhaps part of their role is to guard the wraith prisoners? The OSE SRD description of energy drain doesn’t specifically state that it wouldn’t work on undead, but it follows logically from the flavor of the power to say that wights would be immune. 

Next time: The World's Largest Rewrite: Grey Horse, Devil Swine, and Normal Humans

The World's Largest Rewrite: Dungeon Is Wet, Tortoise Foreshadowing, and 30-50 Feral Hogs

Last time: The World’s Largest Rewrite: Salvaging the Core Idea From a Megadungeon Disaster

#104: Sea Serpent (Lesser). Why “lesser”? There is no “greater” sea serpent in the OSE bestiary. Regardless, it is a chance to carve out a water area separate from the giant catfish's pond. One of the big problems with the original WLD, relative to megadungeons like Arden Vul or Thracia, is its flatness; most good megadungeons are three-dimensional, and defined as much by their depth as their breadth. 

We won't worry too much about specifics just yet -- the diagram I posted at the end of the last post is for abstracted, relative positions. But lets assume the lower parts of this dungeon are flooded with seawater, and this was an Alcatraz-style island prison before it became an adventuring site. I like the idea that this part of the dungeon is a possible escape route, but made dangerous by sea creatures that have made their lairs here. The serpent is tough enough to dissuade most of the nearby prisoners from sneaking out this way. The black dragon could probably kill the serpent, but it is too big to fit through the underwater tunnels (and the seawater dilutes the acid too much for it to corrode its way out).

#24: Crab, Giant. Conveniently showing up right after Sea Serpent, we’ll build out our ocean depths a bit more here. We’ll put the crabs a little closer to the core. They scavenge what the sea serpent doesn't eat itself, and are also occasionally consumed by it.

#18: Cat, Great. Rolling randomly for the sub-types once again we get… sabre-toothed tiger. Hell yeah. “Normally only found in Lost World regions.” OK. Maybe part of the purpose of this prison isn’t just to house criminals but also to preserve lost wildlife that no longer exists out in the world. I’m going to put the tiger near the nobles and the catfish – he’s been drawn to the fresh water and stalks the area nearby. There may be a supernatural zoo sub-theme we can explore with other entries.

#81: Nixie. Yeah, nixies are going to require some more dungeon infrastructure and background to explain. I’m beginning to think that entire prison is not just an island, but also overgrown and covered in natural growth on the top, including a large body of fresh water. The nixies were washed in here when the water eroded through the dungeon's ceiling and flooded several regions. The lake where the giant catfish lives is a terminus, but the nixies control the river flowing into it. They’d like to claim control of the lake, but the giant catfish is too big for them to deal with directly. They’ve probably charmed other humanoids, including a few of the noble’s retinue, and perhaps some others we haven’t placed yet.

#62: Insect Swarm. This could go anywhere, couldn’t it? We don’t need to explain why an insect swarm is in the prison, because insects just like to show up in places where they’re not supposed to be. I like how the OSE bestiary has such extensive procedures for encountering them. I’ll draw inspiration from one of those – the rules for escaping the swarm by “diving into water.” Putting them near the water gives the PCs an “out of the frying pan, into the fire” option that might send some of them into the arms of the nixies. I don’t want more bees, so we’re going to go with beetles instead. I think they’re actually plant-eaters and just want to consume the PCs clothes and other textiles, but adventurers won't know that, and their bites through the clothes still hurt!

#133: Wight. It’s interesting that OSE says these are “Corpses of humans or demihumans, possessed by malevolent spirits.” The 2014 5E Manual suggests they are more conventional undead, i.e., the evil spirit is animating the same body it occupied before death. The Monster Overhaul, my current go-to bestiary, emphasizes that “A Wight’s un-life is tied to an oath, a strong emotion, or the simple will to endure.” It has a nice random table of wight types. I rolled on it and got “Avenger.” Perhaps these are enforcers who swore an oath to the prison builders to hunt those who escaped their cells. The oath extends into un-life (oops) and they’re now doing this forever. I think the builders of this prison may be jerks. Per OSE, wights that drain someone of all levels create more wights, so these wights may be “recruiting” more hunters.

I don’t want a whole undead zone where they’re all clustered together, so I’m going to separate these guys from the mummy-zombie zone. We’ll place them in as-yet unexplored territory south of the ochre jelly zone. Mundane acid doesn’t harm them, so they’re safe from the jellies. Presumably they roam around looking for escapees, but their barracks are down there. 

#92: Pixie. OSE treats pixies and sprites as separate things, and while the latter has a bit of a hook to it, pixies are quite boring. The Monster Overhaul lumps them together, but does include some extra flavor we can tap. They are often invisible and have a mercurial, forgetful nature. I like the idea that these invisible troublemakers were accidentally captured when some larger, more important prisoner was detained. That could place them almost anywhere, but the bigger the monster, the more plausible there presence here. I believe they are kind of Tinkerbelling or Jiminycricketing the dragon. The dragon probably finds them annoying, but hasn’t dissolved them yet, because their polymorph ability might come in handy at some point.

#36: Elemental. Picking randomly, we get fire elemental. OSE emphasizes they are summoned servants. Of the prison builder perhaps? I need more detail, so checking the Monster Overhaul, we get some excellent flavor and tables. The “who summoned this elemental?” table suggests tortoise tsar, a Monster Overhaul original, who has some fire-based powers, so fire elemental fits. The tortoise tsar isn’t part of my original conceit of using the OSE bestiary, but I can merge him with the dragon turtle entry. We’ll plan to revisit this situation when we roll up dragon turtle / tortoise tsar and figure out what is going on here.

#15: Caecilia. It’s OK, I had to look it up too. It’s an amphibian that looks like a snake or worm, although OSE’s are 30’ long. To take stock here, all of our monsters so far fit into one of the following categories:

• Prisoners or "zoo" animals
• Invasive species or other intruders
• Guardians or servants of the prison builders
• Creations of other creatures in the dungeon

I want to avoid putting all the monstrous animals in the second category. The prison should still feel prison-like, and not be completely overrun by creatures from outside. I think we’ll say these are prisoners, like our sabre-toothed tiger. Like the big cat, they’re extinct in the outside world (probably for the best – 30’ long, yikes!) but they live on here in the prison. 

#11: Boar. As I said, there’s a lot of beasts in this bestiary. I’m going to tap the Monster Overhaul for inspiration again. It has a table for “local boar crimes,” which is too good to pass up. I rolled “ransacked a granary.” And I note that the Overhaul suggests boars are “as smart as most people.” I like the idea that the prison builders decided these 30-50 feral hogs were smart enough to stand trial for their crimes, just like people would. So they’re prisoners here, recently escaped from their cells, but still trapped within the larger prison. This could go in a sort of Silent Titans direction.

Next time: The World's Largest Rewrite: Floating Heads, Mother Fungus, Cellipedes

The World’s Largest Rewrite: Salvaging the Core Idea From a Megadungeon Disaster

I recently read an exhaustive review from 2013 by oriongates of the World’s Largest Dungeon (WLD for short, hereforth), a megadungeon released for D&D in 2004, at the height of the d20 OGL publishing era.

The product’s gimmick was that it was bigger (and more expensive) than any other dungeon of the time, and that it included every type of monster in the game. This is good advertising and bad design. Reading oriongates’ exhaustive review, it is clear that much of the size is wasted on repetitive, purposeless rooms, and that many of the monsters are shoehorned in to meet the product’s central conceit, rather than appearing because they present interesting challenges and opportunities to tell the story of the place and give the PCs compelling choices for exploration. Compare this to something like Arden Vul (more modern in publication date, more old-school in approach) where the size, shape, and populace of the dungeon is very deliberately communicating something, rather than trying to hit an external, artificial objective.

The WLD would be a mess in any system, but it is particularly ill-suited to D&D 3.5E, where monsters and magic have particularly detailed and complex abilities and mechanics. At least a B/X game could reduce this kind of enterprise to minimalist keys, in the style of Palace of the Vampire Queen. Anything in the D&D-3.5E world has to deal with huge stat blocks and include contingencies for super-powered characters.

Trying to fix this abomination is a fool’s errand. In their review, oriongates mentions trying to do so, but later giving up, and I think that’s the right call. If neither the map nor the lore nor the factions nor the NPCs are really anything to write home about, why would anyone think it was worthwhile to fix? 

I mean, these people somehow delude themselves into thinking it might, but ... But it might work for us.

Update (June 2025): I started writing this post (and subsequent posts about the WLD) in early 2025, and it was mostly done by the spring. It turned out to be a topical exercise, as I later learned that a crowdfunded overhaul of the the WLD is in the works, and as of writing this update, has clocked nearly $700,000. The Backerkit page is extensive, and features lavish production design and a mountain of extras. It also features a long list of contributors with extensive RPG credits. 

While the product certainly looks nice, I am suspicious of language like "Massive in Every Way That Matters." The marketing still sounds very "woooah, Guinness Book of World Records!!" rather than "here's how and why this would be a compelling roleplaying experience." Sheer page count and taxonomic exhaustiveness cannot make up for what really defines a great megadungeon: a specific, gameable concept for an intriguing adventure scenario underpinned by an evocative milieu. 

That said, perhaps the team behind this effort decided that such nuances of dungeon design were simply not right for what is basically get-hype marketing copy, and they will quietly fix the original WLD's many flaws behind the scenes. It will be interesting to monitor the reviews when this beast comes out and see how it does. End of Update.

So, if we were going to try to make our own WLD, we could make life easier by first starting with a smaller monster list. D&D’s bestiary in the 3.5E days was huge, and featured some really weird, niche monsters that could not logically be packed into a megadungeon. Go back to something more fundamental. Start with a classic and basic bestiary, like the OSE monster bestiary. There are 138 monsters here. That is pretty manageable. 

Here’s a list of all those monsters. Look at 'em go!

1. Acolyte 
2. Ape, White 
3. Bandit
4. Basilisk
5. Bat
6. Bear
7. Beetle, Giant
8. Berserker
9. Black Pudding
10. Blink Dog
11. Boar
12. Brigand
13. Buccaneer
14. Bugbear
15. Caecilia
16. Camel
17. Carcass Crawler
18. Cat, Great
19. Cave Locust
20. Centaur
21. Centipede, Giant
22. Chimera
23. Cockatrice
24. Crab, Giant
25. Crocodile
26. Cyclops
27. Dervish
28. Djinni (Lesser)
29. Doppelgänger
30. Dragon
31. Dragon Turtle
32. Driver Ant
33. Dryad
34. Dwarf (Monster)|Dwarf
35. Efreeti (Lesser)
36. Elemental
37. Elephant
38. Elf (Monster)|Elf
39. Ferret, Giant
40. Fish, Giant
41. Gargoyle
42. Gelatinous Cube
43. Ghoul
44. Giant
45. Gnoll
46. Gnome
47. Goblin
48. Golem
49. Gorgon
50. Grey Ooze
51. Green Slime
52. Griffon
53. Halfling (Monster)|Halfling
54. Harpy
55. Hawk
56. Hellhound
57. Herd Animal
58. Hippogriff
59. Hobgoblin
60. Horse
61. Hydra
62. Insect Swarm
63. Invisible Stalker (Monster)|Invisible Stalker
64. Killer Bee
65. Kobold
66. Leech, Giant
67. Living Statue
68. Lizard, Giant
69. Lizard Man
70. Lycanthrope
71. Manticore
72. Mastodon
73. Medium
74. Medusa
75. Merchant
76. Merman
77. Minotaur
78. Mule
79. Mummy
80. Neanderthal (Caveman)
81. Nixie
82. Noble
83. Nomad
84. Normal Human
85. Ochre Jelly
86. Octopus, Giant
87. Ogre
88. Orc
89. Owl Bear
90. Pegasus
91. Pirate
92. Pixie
93. Pterosaur
94. Purple Worm
95. Rat
96. Rhagodessa
97. Rhinoceros
98. Robber Fly
99. Roc
100. Rock Baboon
101. Rust Monster
102. Salamander
103. Scorpion, Giant
104. Sea Serpent (Lesser)
105. Shadow
106. Shark
107. Shrew, Giant
108. Shrieker
109. Skeleton
110. Snake
111. Spectre
112. Spider, Giant
113. Sprite
114. Squid, Giant
115. Stegosaurus
116. Stirge
117. Thoul
118. Titanothere
119. Toad, Giant
120. Trader
121. Treant
122. Triceratops
123. Troglodyte
124. Troll
125. Tyrannosaurus Rex
126. Unicorn
127. Vampire
128. Veteran
129. Warp Beast
130. Water Termite
131. Weasel, Giant
132. Whale
133. Wight
134. Wolf
135. Wraith
136. Wyvern
137. Yellow Mould
138. Zombie

I’m going to randomly pick from this list and put it together as I go. There’s no reason to do this randomly. Randomizing it just makes this exercise more interesting. I mean I certainly don’t want to do this in alphabetical order. No “A is for ape, white as the snow; B is for bandit, after your dough…” No, no, no, absolutely not. Random it is.

Droning Bee by Eric Poslinski

#64: Killer bee. Let’s keep as much of the WLD conceit as possible and assume that this place is (or at some point in the past was) a prison. The OSE killer bees “build hives underground” (interesting) so something about this place attracted them when they were seeking a nesting site. We’ll decide what that was later. But we can definitely imagine a structure of beeswax and propolis that has repurposed and displaced parts of the original dungeon structure. This will give us a good opportunity to add some texture to the dungeon, right from the start; from oriongates’ review, it’s clear that too much of the WLD was plain gray stone.

#29: Doppelganger. The doppelgangers might be prisoners. Thracia has a good encounter with doppelgangers who have been imprisoned behind a sealed-up wall for centuries, which I suppose implied that doppelgangers are immortal and can't starve to death. Perhaps doppelgangers go into a hibernation state when no other sapient bipeds are nearby. Their cells have been covered in hive structure, but they may wake up if PCs or other people get close enough.  

#82: Noble. “Powerful humans with noble titles (e.g. Count, Duke, Knight, etc.)... Squire and retainers: Accompanied by a 2nd level fighter (a squire) and up to ten 1st level fighters (retainers).” OK, I like the idea that this guy and his followers are all prisoners together. Let’s suppose our noble is a Qin Shi Huang type who has done some pretty terrible stuff. The PCs can talk to him, and he will argue that his methods were necessary to unite a kingdom that will last long after his death. 

I like the idea that rather than this being a prison of unambiguously evil creatures like demons and liches, it is more like a real prison, with degrees of culpability, moral gray areas, and judgment calls on the part of the celestial jailers. The PCs will probably not find this noble to be particularly sympathetic… but they might! Or at least they may see him as someone worthy of a temporary alliance or truce. 

He’s here with a cadre of true believers who volunteered to go with him into captivity. Since I already established that the doppelgangers were not near other humanoids, I’m going to put this noble and his retinue on the other side of the bees. The noble and his retinue have been stealing honey to supplement whatever sustenance is otherwise available (we’ll decide later how creatures are getting enough food  to survive here). The killer bee hive provides a nice risk/reward opportunity for NPCs and players alike, since the honey can heal wounds… but the bees also, uh, kill, so it is a dangerous place as well.

#30: Dragon. We’re not messing around. Let's go right to the big guns. I agree with the WLD’s decision not to include every variety within a category of monster, so when we get these monster entries with multiple sub-types, we'll just choose one. I’ll pick randomly and get a black dragon. 

This particularly rapacious dragon has been using its acid breath to slowly burrow out of its cell. The prison was designed to resist this kind of escape attempt, but its structural integrity depended in part on maintenance and monitoring from the jailers, which has since lapsed. The dragon is patient, and it has linked up a number of cells around its primary domain. Like the bees, the dragon provides a reason for the dungeon’s structure to deviate from the sensible, dull, repetitive layout one would expect of a prison. The dragon limits the killer bees’ expansion in this direction, as its scales protect it from their stings, and its acid can easily destroy their hiveworks.

#138: Zombie. There is no real reason to imprison zombies. Another complaint from oriongates' review is that there are too many monsters – particularly the undead – that celestials would just destroy, not imprison. 

So we'll say the zombies were created by something else that is imprisoned here; something more dangerous. Some or perhaps all of them are former members of the noble’s nearby entourage who were killed and then zombified. We’ll figure out later who or what caused that to happen. For now, we'll place our zombies just to the east of the killer bees. They’re indifferent to the stinging insects, so they make a good buffer. 

#16: Camel. From dragons and zombies to… camels. Not all the entries are going to be easy. The camels are not prisoners, and they’re also unlikely to have migrated here intentionally, like the bees did. Let’s say they were brought here by some group of prisoners – possibly our noble and his retinue. We’ll put them adjacent to the dragon’s territory. The dragon has been herding them to supplement whatever food it is getting elsewhere. Humans find them to be irascible, but they freeze like deer in the headlights when the dragon approaches.

#40: Fish, Giant. One of the oddities of old-school D&D is the extensive “unusual animal” entries. I hadn’t realized the OSE bestiary included five different types of giant fish. We’ll pick randomly again for our sub-type, and land on giant catfish. One of the issues with the WLD is that several areas of the dungeon are transparently excuses to cram in monsters who need a custom biome, and the WLD’s “water level” is one of the most conspicuous. We’ll instead presume a number of separate watery areas, several of which may also be connections between different parts of the dungeon, as good megadungeon design necessitates. Water can also help explain the breakdown of separations between dungeon areas. 

We’ll say that this fish was once an ordinary catfish that was sucked into the dungeon as part of a flooding event, and later grew to its abnormal size as a result of the powerful mana suffusing the water within the dungeon. Because it is submerged except when hunting, it is safe from the killer bees, so we’ll put it next to them to form another buffer area. The noble and his entourage probably come here for water; they know to avoid the catfish.

#85: Ochre Jelly. A classic dungeon denizen that can be placed just about anywhere. Since the jelly is acid-themed, we’ll place it near the black dragon’s lair. Perhaps the jellies even originated with the dragon, gradually gaining mobility through latent dungeon magic?

#79: Mummy. We’ll put the mummy near the zombies, and posit that the mummy (whether intentionally or ambiently through its aura of uneath) is what roused them. Obviously this is another prisoner – perhaps the magic that allows it to respawn is particularly pernicious, and the celestials decided to imprison it after failing to find the canopic jar that powered its resurrection cycle. We’ll learn more about the mummy after we place a few more monsters nearby.

#74: Medusa. Gorgons are a favorite of mine. It would be easy enough to just assume this one is a prisoner, but I want to subvert expectations here. Perhaps they were contracted by the builders to help build the dungeon; after all, turning living matter into stone is a good way to supplement whatever stone you’re quarrying. This gorgon was either betrayed by the builders, or trapped here by accident. The gorgon is immortal and at least as willing to negotiate as the noble, if not more so. They can’t turn off their gaze, so they’re a dangerous ally even when attempting to work with diplomatic PCs.

Next time: The World's Largest Rewrite: Dungeon Is Wet, Tortoise Foreshadowing, and Feral Hogs on Trial