Gloamhold: Down with the 15-Minute Adventuring Day

I hate dungeons that are a seemingly unending grind of combat. The party do two rooms, blow all their spells and retreat to rest overnight.



Sure they make progress but it is slow and frustrating progress. I’d rather the party explore 10-20 rooms, actually map some of the complex and have to cleverly manage their resources before resting. Such game play rids us of the much reviled 15-minute adventuring day and provides the party with a sense of actual progress and achievement.

Achieving this is, however, hard. “Optimisation” and “character build” have seemingly become the norm in the hobby. It’s hard to tell who started the optimisation race , but it’s inevitable result is that as the PCs get tougher, so do the monsters (or vice versa). In this scenario, most of the encounters have an CR equal or greater than the party’s average level (as most GM like to at least challenge their players) and the party rarely complete more than four or five encounters before resting.

One inevitable consequence of this is that combats take longer to run and feature more complex opponents. The party rarely have an easy fight and consequently every battle is a life and death struggle. Game play is a grind and the party spend a lot of time resting (or they often retreat from the dungeon). Weirdly, the other result of optimisation and the CR system is smaller dungeons as designers create dungeons the PCs can (theoretically) get through without resting.

For Gloamhold, I want something different. I want to embrace the feel of Old School exploration. I want the PCs to wander the gloomy, dusty halls for hours. I want them to poke about in empty or abandoned rooms and spend time actually experiencing Gloamhold instead of simply whacking a long succession of monsters.

To achieve that, I need a plan. I particularly need a plan if other people help me design Gloamhold, in the future.

The following thoughts on CR ranges and frequency of encounters come from several sources.

CR Range

Obviously, the CR of encounters within Gloamhold will have a direct affect on the party’s ability to explore extensively between resting.

For encounter design, the Pathfinder Roleplaying Game Core Rulebook suggests the following CR categories:

  • Easy: CR = APL -1
  • Average: CR = APL
  • Challenging: CR = APL +1
  • Hard: 15% CR – APL +2
  • Epic: 5% CR = APL +3

Similarly, 3rd edition breaks down the CR spread of encounters thusly:

  • Easy: CR is lower than the party level. Easy encounters should comprise 10% of total encounters.
  • Easy*: These encounters are easy if handled correctly; if not handled properly, this encounter becomes challenging or very difficult. Easy* encounters should comprise 20% of total encounters.
  • Challenging: CR equals the party’s level. Challenging encounters should comprise 50% of total encounters.
  • Very Difficult:CR is 1-4 higher than the party’s level. Very difficult encounters should comprise 15% of total encounters.
  • Overwhelming: CR is 5 or more higher than the party’ s level. Overwhelming encounters should comprise 5% of total encounters.

Slow Advancement

(For more on the slow advancement track, check out why I love it and how I suggest using it in your campaign).

Using 1st-level characters as an example, a group of four characters using the slow advancement track need 30 CR 1 encounters to reach 2nd-level (assuming no bonus XP for role-playing or story awards). We already know not all encounters will have a CR equal to the party’s level. This means, based on the breakdown of encounters 3rd edition gives us, the CR of any 30 encounters should roughly be:

  • Easy: 3
  • Easy*: 6
  • Challenging: 15
  • Very Difficult: 5
  • Overwhelming: 1

Using this breakdown of CRs, 24 of the 30 encounters have a CR equal or lower than the party’s CR. This means they’ll be able to handle more encounters and explore more of the complex before they rest.

How Many Rooms Should Be “Empty”? 

One of the features of Old School play, is unoccupied rooms. Unoccupied rooms mean larger dungeons, which promotes the exploration style of play.

If every room in the dungeon is stuffed full of opponents, it’s hard for the PCs to make meaningful progress during a delve. If opening every door results in a fight, the party quickly realise this and make their preparations accordingly. Adding empty rooms give the party room to breathe and manoeuvre. It breaks up the style of play by providing options for non-combat dice rolls and even (gasp) roleplaying

1st edition (Appendix A, Random Dungeon Generation) gives us the following frequencies for occupied and non-occupied rooms in a dungeon:

  • 60% rooms empty
  • 25% monster (with or without treasure)
  • 5% special feature
  • 5% trapped
  • 5% hidden treasure

Assuming the PCs can potentially earn XP in any room that isn’t empty, 40% of the rooms they discover will hold some form of challenge (and be a potential source of XP).

This means, on average the party must explore 75 rooms before they reach 2nd-level. Thus, Gloamhold needs to be huge — with lots of empty or abandoned space — to support multi-level play.

Published by


Creighton is the publisher at Raging Swan Press and the designer of the award winning adventure Madness at Gardmore Abbey. He has designed many critically acclaimed modules such as Retribution and Shadowed Keep on the Borderlands and worked with Wizards of the Coast, Paizo, Expeditious Retreat Press, Rite Publishing and Kobold Press.

15 thoughts on “Gloamhold: Down with the 15-Minute Adventuring Day”

  1. Very good points. I’d add that the addition of various point pools and limited use character abilities starting with 3.X become destabilizing in 15 minute dungeons. When the party rests after every other encounter, these pools no longer function as resource constraints but almost as virtual character levels. The example I have in mind is the Monk’s ki pool in Pathfinder adding extra attacks at full bonus.

  2. I definitely sympathize, but frequent low CR fights can easily turn into dull attrition that fails to add tension or add to the narrative. As Rod suggests, a significant part of the issue is ingrained in the resource cycles that make up games like Pathfinder and even 5th edition D&D. This is compounded by an almost nonsensical set of encounter guidelines and Challenge Ratings. The 15mwd will persist so long as character effectiveness is heavily dependent on radically fluctuating resources and CR seems largely divorced from actual threat.

  3. I’m missing the roleplay encounters. The room that holds no monsters, traps, or treasure, but a kitty. Possibly lost, possibly caged (ogre snack), probably hungry and scared.
    Or maybe the bones and especially the ghost of a dog, that fled combat and died here and needs to be reunited with its master (who was killed several rooms away). Of course attempts hostility will result in loud barking as the dog calls for its master (alerting the locals), while trying to ignore/leave it will make it howl in loneliness (same result).
    And of course Old Faithfull: the very expensive cabinet with exquisite woodwork and goldleaf decorations and hidden drawers: -and the dreadfull problem of how to get it home…

    Perhaps these are supposed to be covered under “5% special feature”, but in that case, the 5% is really low imo.

    1. Roleplaying encounters definitely have a place in Gloamhold – and in any dungeon hopefully. Probably some of these fall into the Easy* category in that talking to the NPCs can resolve the situation instead of just hacking away.

      I also think I need better terminology having read this and other posts. Empty room implies just that. In my mind an empty room just means a room in which there is no immediate challenge. Such rooms could hold prisoners, non-hostile dungeon natives and so on.

      And, of course, we must not forget wandering monsters. A ripe source of roleplaying and problem solving!

      1. Yea, to me, an empty room means one with nothing interesting in it except maybe some broken furniture or a fraying rug, and of course 3 inches of dust.

        Also, while most roleplaying encounters likely will be fairly easy, there’s absolutely no reason they all must be. Perhaps the encounter with a young blue dragon that instead of combat accepts a riddling duel for the PC’s prize posessions will eat less daily powers and other resources than the 3 hostile orcs a few rooms earlier, but which do you think will make a bigger impression (and should reward the most *experience* points) ?

  4. I think that you also need to factor in a few RP encounters, perhaps reminiscent of the sort of things found in rpg games, where you can give the party a reason to go to location a or b through the use of npc-given mini-missions. I’d also agree with some of the comments that at low levels, many combats will likely feel very grindy, and therefore slow. Although I understand why you like the slow progression, I do wonder if that it may make advancement a little too slow, unless enough variation can be created to keep things feeling fresh, especially as many of the encounters at level 2 will be very similar to those already faced. I think that this will be quite a challenge for your designers.

      1. ‘Encounters’ can be softened by conditions that unbalance the situation (you catch the orcs on the way back from a raid, low on hit points and spells, for those that have them; the ogre is asleep, you can have at least one free round of attacks if you don’t make a lot of noise or might completely avoid the fight if you don’t wake him; the goblins are already embroiled in a fight with another threat).

        Other ‘encounters’ could be things that are completely safe if handled correctly, or completely avoidable. I tend to put non-critical extras behind avoidable threats. The goals can still be met without that fight, but if you choose to fight (or to try for the loot you expect or can see right behind the ogre) it’s on your head.

        Player agency allows bad decisions, after all.

        1. Excellent point, Keith, about softening encounters with “conditions.”

          The classic example of the “other” encounters for me is the roper in Forge of Fury. It’s not essential, the players don’t have to do it, but it’s rewarding if they do.

    1. I think you are right: this will be a challenge to do well. I think there are a couple of things working in our favour though:

      1. Low CR fights (fights with a CR lower than the party’s) will probably be over pretty quick. They are unlikely to go on for more than a round or two.

      2. In a megadungeon setting, the party can go where they like. So, if they get bored by the encounters in (say) Rivengate they can try somewhere else. I think megadungeon play works well in an exploration/sandbox style of game play and in such a game the players have tremendous freedom of choice to go where they will and fight what they want.

  5. Reinforcements, my friend, and tactical adjustment.

    If the PCs want to pick away at an adventure site, they’re going to find it harder going the more times they go back.

    Nova strike followed by retreat means the PCs are likely to find the positions not only reinforced (the inhabitants are on high alert) but better prepared for the tactics used. “PCs fireballed the place” might lead to barricades and improvised baffles, and broader placement of guards so they don’t all get caught at once. Anything of value that might have been easily reached gets pulled back to a more defensible position. Other ‘defensive’ (remember the best defense) options might be brought into play because it’s obvious something is going on.

    Or, if the initial attack is fierce enough, the targets bug out entirely. Sorry about the prisoners the PCs were hoping to rescue, they’re unneeded baggage and left behind (dead, of course). Or they’re a portable larder and the PCs have to try to catch the targets on the road before all the prisoners are eaten.

    Having alternate routes in and out so the PCs don’t have to — and perhaps cannot — go deep, hit hard, and pull back out might mean they have to hold onto some of their big mojo for the retreat.

    Depending on the nature of the scenario, “sneak in and avoid notice, and avoid especially major confrontations unless absolutely necessary” might be the only path to success, and nova strikes are not only inefficient, but possible counterproductive.

  6. Other tips to break up the 15 minute work-day in a megadungeon:

    1) Random encounters. This takes the decision of when to have an encounter away from the players and demands a response from them when they are not in top form.

    2) Get them lost. It can be as grand as a teleportation trap or as simple as a shifting wall. Take away their ability to simply say “We go back to town” after one or two encounters.

    3) Rivals. Add a competitive spice to the game with rival adventuring parties. These can be NPCs, or actual other groups of adventurers running the same dungeon. If one party doesn’t get a given treasure, another party will.

    4) Deadlines. While megadungeons do not necessarily have overarching plots, some subplots or quests could come with deadlines attached. A deadline they’re not going to meet if they plod through the dungeon at a snail’s pace.

    5) Logical consequences. Basically, this means that the dungeon denizens become aware that adventurers are slowly but surely working their way through the dungeon. Given enough time to prepare, intelligent monsters may devise ambushes, build traps, or even organize into a mutual defense society of sorts.

    1. Keep on the Borderlands was an awesome adventure for the roleplaying possibilities. You could do tons with that, for a year or more, if you wanted. Chisel at this group by allying with this band over here, whittle down the trust of those guys over there. You could do a lot with that little module. It taught countless DM’s how-to effectively have a world that would react and plan to player characters. And in the character’s absence, they could try to mend fences in their back yard, but don’t do that trick so often.

      Players get wise to it.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.