D&D 5E Megadungeon delving as a campaign’s core; is it compatible with modern play?

Micah Sweet

Level Up & OSR Enthusiast
The recent changes to exhaustion is a good example. Yeah, a -1 cumulative penalty is easy to remember, but it's a lot less impactful than the old rule, and players don't worry as much about planning appropriately or risk assessment when the penalty isn't as bad for failing.
It's better than WotC's version, but not as good IMO as Level Up's version.

log in or register to remove this ad


Moderator Emeritus
But how do you play up the consequences of diminishing resources in WotC 5e when the rules radically weaken those consequences? I don't find pretending hardship is actually a problem when mechanically it isn't very satisfying.

Unless the characters literally have a way to deal with any and every potentiality I really don't see this as a problem. Every group does not have every spell or every piece of equipment (or enough of it) at any given time. I have had parties run out of rope, role-play the misery of having nothing to eat but goodberries, go days without anywhere safe to take a long rest, not have enough sacks or strength among them to carry off a hoard of treasure they found, or have to figure out how to get unequipped civilians back to safety. I have had them get separated or argue over the best way to cross a chasm when they worried that using spells to do so would leave them vulnerable in what they assumed was a big fight to come. I have had them assume something was "the big fight" when it wasn't and over-extend themselves unnecessarily. They have considered leaving the dungeon and worried about leading terrible monsters seeking revenge back to town. I have had a party where no one bothered to worry about light sources because they all had darkvision and then get surprised over and over because of the penalty to their passive perception. Etc. . Etc. . . It doesn't take much imagination or effort.

Currently one of my groups is scouting a submerged sahuagin lair and are there long enough to start carefully keeping track of their supply of potions of waterbreathing, esp. after the druid in the party decided to use his last remaining spell slot high enough to cast the spell himself.

And this is all without all the particular rules and rulings I make as DM depending on the environment and other factors. I mean, I am the kind of DM that thinks that the very notions of RAW and RAI are self-delusion, but even if I took a literalist approach, I would have to change nothing and still push the group in that way.
Last edited:


A Title Much Cooler Than Anything on the Old Site
I've been running the mega dungeon campaign Rappan Athuk by Frog God Games using 5e for over three years now. I do use some third-party supplements to the rules and some rule variants, so I'm not running bog standard 5e. But most of that related to downtime activity and leveling, not the actually exploration, combat, and skill mechanics.

Some things that have made it successful in my opinion include:

1. It is truly a MEGA dungeon.
Large enough for distinct regions and factions.

2. Home base.
This starts in a small trading settlement fairly close to the dungeon that serves as a home base. Plus there is a ruined castle with rules for fixing it up (after defeating/allying with the current goblin inhabitants). It is mean to be taken over while still in tier one but needing significant gold to fix up, requiring more adventuring). Also, I use Matt Coleville's Strongholds & Followers rules, so their "home base" grows as they level, eventually becoming a village, than a town, etc. Gives a reason to spend money, a reason to care about the threat of the nearby dungeon, and lots of opportunity for political intrigue if the players are up for that.

3. XP for GP.
I have used certain events or accomplishments as milestones, but I basically use a simple rule of rewarding XP for treasure removed from the dungeon. This encourages a different play style than XP for killing.

4. Reputation rules (Renown/Infamy faction points).
Using homebrew rules modified from an ENWorld article, the PCs earn renown and infamy that result in various good and bad events and determine how different groups interact with them. This is also a way to reward PCs for defeating certain enemies or overcoming certain challenges that don't lead to loot.

5. It is still part of a larger world.
The PCs, especially in lower levels, would have to travel long distances to larger cities to sell certain very expensive items or to buy rare items. So dungeon delving would be punctuated with some side travels and encounters, plus city exploration. As they get to higher levels they needed to find and interact with more powerful actors and factions to buy/sell items, get support, hire forces to secure areas they cleared, etc.

If they want a break from the mega dungeon there was always the option of doing a side quest or even abandoning Rappan Athuk and adventuring elsewhere. Their is a LOT of content for the Lost Lands. So the players never had to feel that they were "stuck" in the mega dungeon. If they get bored with it, the game can continue with the same PCs. But, interestingly and a bit surprisingly to me, other than making runs to a city or a powerful NPC for trade, research, and politicking, the game has taken place almost exclusively in the Mega Dungeon. An average of about 10-12 hours a month (we play an 8-hour session once a month, but we have fairly frequently played twice a month, rarely missing a month, so I'm rough estimating about 10-12 hours). I'd have to check the calendar for the exact number of months, but I know it is over 3 years. Easily over 400 hours of play time in one mega dungeon, with a fairly slow level progression (most characters are at level 17 after over 3 years of play).


We have run (and continue to run) many mega dungeons in 5e. So yes, it works just fine - lots of fun. However, if you're interested in more old school crawl (ala Castle Greyhawk), you could adopt some or all of the following:

1. Eliminate or severely restrict darkvision.

2. Use slow natural healing rules.

3. Eliminate passive perception. Avoid reducing challenges like secret doors and traps to just skill checks; instead, have players describe how they are searching/opening a door or trying to disarm a trap.

4. Track equipment and resource consumption, including food, water, ammo, etc.

5. Track encumbrance. Do this in an easy/intuitive way though - make it common sense, not a calculator exercise.

5. Have the smarter monsters use hit and run tactics (not fight to the death), perhaps making alliances in self defense. Sticky enemies that keep coming back are often the most memorable.

6. Use "random" encounters. Using quotes here in that I try to use judgement so random encounters don't dominate, but they also don't let the party get too comfortable in the dungeon.

7. Allow the PC's to have henchmen and retainers.

There are many, many other tips you can find on these boards and elsewhere. These are just a few that I tend to employ.
Last edited:


I have a setting I hope to eventually run that entirely takes place in an interplaner megadungeon, but it's a place where people live, so the characters are basically natives of the dungeon. The dungeon isn't static, so new traps and rooms and threats appear and old ones change or disappear, so it's not something that's been thoroughly mapped out. And I know that at least some of the game would be inter-settlement conflicts and diplomacy, as well as exploration and trade. I can't imagine why it wouldn't work.


Victoria Rules
All you need is sufficient challenge for the PCs power level, playing up the consequences of diminishing resources of all kinds,
I suspect that's where people might have trouble, given 5e's significant downplaying of the resource-management side of the game - hit points, in particular, being so easy to recover.
use wandering monsters (either randomly or narratively), and players willing to buy in to the narratives that emerge from play.
This last is important, though speaking personally if they're not willing to buy in to emerging narratives there's a severe expectations problem.
To be honest, I don't understand the need for specific mechanics for exploration. The PCs explore and uncover things using reason and skills. What else do you need? Exploration has been a big part of some games I've run and I never had specific mechanics for it beyond description and reaction.🤷‍♀️
Exploration mechanics around time spent and distance covered - and mapped - are relevant. Granularity is also relevant - boiling an adventuring day's worth of exploration down to a couple of 4e-like skill challenges, for example, rather defeats the whole purpose.


Victoria Rules
The shift from old-school to trad/modern play has far less to do IMO with D&D moving "out of the dungeon," or even from challenge and skill driven play to story and performative thespianism, than the shift from open to dedicated tables. Everything that makes trad play different from old-school follows from the player-to-character attachment fostered by running the same one PC every session for a whole campaign. So to ask the question, "Can a megadungeon support modern play?" is to ask, "Can a megadungeon be interesting to a fixed group of players playing a fixed cast of player characters?" I think yes, but it would likely require some very deliberate dungeon design.
Along with much lower lethality that a typical megadungeon tends to offer.

The 3e Rappan Athuk, for example, leads with a killer trap in the first few rooms and the write-up even calls this out.


Victoria Rules
I imagine you’re referring to spells & hp more than food, water, ammunition & light? Those consequences are still very much there and things can still be as deadly. There just tends to be less of “we go back to town/barricade the dungeon door and rest 3 days to heal, let’s move on” and “okay, long rest over - everyone reset and let’s move on.”
Except going back to town or holing up for three days mean that time is passing in the setting; far more time than a simple overnight rest; and either of these options can present their own dangers even if only a much greater chance of wandering monsters. Further, more time passing in the setting gives the setting and its occupants more time to react to what the PCs have done. (or, in cases where going back to town then returning to the dungeon takes weeks, more time to stand down from alert)

And if the party doesn't have the ability to cast Create Food and Water that extra time can become huge.


Follower of the Way
Is it possible? Yes, absolutely. But you're going to have to accept some significant changes to one side, or both of them.

I'm a big fan of the (so-called) "modern," character-driven approach, where things are more like Babylon 5 or Star Trek: Deep Space Nine than they are like Star Trek: The Original Series or Blake's 7. That is, before B5, the vast majority of such shows did not have long-running, intentional character arcs and such; they were written one season at a time, with the occasional hook thrown out that could be referenced back later on. Now, most D&D games aren't as fully written-out as B5 was, and shouldn't be--I am, after all, a bit fan of "play to find out what happens"--but there's a certain analogy to draw here.

Players in older-school adventures were almost entirely "Story After"* and largely avoided "Story Before"; that is, "story" was only something you cobbled together from reflecting back on the events you lived through, and connecting them together into some kind of narrative, even if it is (almost surely) a bit disjointed and liable to sudden lurches and dead-ends. Then the Dragonlance kind of stuff hit, and suddenly Story Before (you decide what story will be told, and then bring that story about; this often involves railroading) was a big deal. The "modern" take, IME, straddles lines but favors something between "Story Before" and "Story Now" (actively and intentionally posing and resolving conflicts through play): backstory and understood interpersonal connections matter a lot, but there's a great emphasis on extemporaneous portrayal and allowing characters to "grow naturally" rather than following anything predetermined.

Either you're going to need to get the "modern" player on board with some of the structures of old-school play (e.g. high lethality, rapid character turnover, "Story After," etc.), or you're going to need to relent on some of those old-school approaches, or some mix of the two. I certainly think it can be done, but it would require group buy-in and an understanding that there will likely be points of difficulty that will need to be worked out as they arise.

*I know some folks are allergic to Forge terms, but these ones are actually really useful IMO.

Voidrunner's Codex

Remove ads