31
\$\begingroup\$

With many published adventures from different systems, and also with adventures I myself have created for my groups, I noticed that the dungeons in them feel static. You have 1 room with beasts that prey on anyone who enters by falling from above. Another room with wolves and a third one with a lone kobold.

And almost regardless what the players do, it is expected that none of the other inhabitants of the dungeons react to what happens around them... or only very very few of them (in the above case the kobold).

Now, I have thought quite a lot on how I could manage it to get the dungeon to feel less static in these regards. But the sad part is with most systems I got a problem in regards to power levels.

If we take the above example as example (it's from a pathfinder adventure). If the players sneak through the dungeon and then encounter the beasts on the ceiling and fight with them...naturally they make quite a noise. If I now let the wolves and the kobold attack them (those are allies with each other), the players are as good as dead as the encounter would go from a normal one to an impossible one for them. At the same time, if I let only 1 wolf enter the fray the wolf encounter becomes too easy.

So thinking more and more about it I found, that making a dungeon feel less static is quite...hard. Especially in systems that have levels.

Now my question is: Are there any commonly used methods, or ways to make dungeons feel less static or avoid the staticness altogether?

\$\endgroup\$
2
38
\$\begingroup\$

It's not only difficult from a design point-of-view, it can also be unrealistic! How are bandits, owlbears, and a hag all living in harmony in neighbouring caverns?

Here are a couple of ideas. By no means is it meant to be an exhaustive list.

Keep 'em Separated

One simple solution is to add distance. Dungeons can be huge. If the swamp monster is a 15-minute walk away from the goblin lair, through meandering tunnels, there's very little risk of one responding to noises from the other.

The downside to this approach is that it makes mapping a little less convenient: it becomes harder to fit the whole dungeon on the one battle-map-grid. There will be large "buffer zones" needed.

Embrace the Interactions

Another solution is to embrace the way dungeon inhabitants react and design with that in mind. I've seen this adopted on the smaller scale (one room might notify another) and the larger scale (the whole dungeon might become alert).

Speaking to the larger scale, in a recent session I set up a small fort with guards stationed around the place. In a worst-case scenario -- where the guards could co-ordinate -- the PCs would be badly outmatched. I left it up to the players to figure out how they could divide and conquer the place.

I tried not to make the place air-tight: there were opportunities to sneak, bluff, or magic around threats. And I tried to roughly guess what the PCs might do in an attempt to make likely encounters balanced for party size and level.

This is less predictable, than the single-room-with-a-monster. And it also doesn't suit all types of dungeon: it's best for intelligent, co-operating creatures.

\$\endgroup\$
2
  • \$\begingroup\$ Hi Nichevo, welcome to rpg.se! Take the tour when you get a chance and visit the help center for more information. This is a really good first answer, thanks for participating and happy gaming! \$\endgroup\$
    – linksassin
    Apr 16 '20 at 5:30
  • \$\begingroup\$ Great answer. I would add Keep them flexible, i.e. those monsters might overwhelm the players if they just stuck to their jobs. So scale them accordingly: make them sick, hungry, or recovering; make it dark; or discombobulate them making them confused about where to go and "who the bad guys are" or even set up a fight between them. How on (Middle) earth did Bilbo and the Dwarves get out of the Goblin Kingdom; or the Fellowship get out of Moria? RPGs are about the story, not the dice and the grid paper. \$\endgroup\$
    – Rich
    Apr 16 '20 at 19:18
28
\$\begingroup\$

Undeniably, adding dynamic responses to a dungeon is hard, and when its a published adventure that doesn't account for it, it's doubly tricky. You're also correct in identifying the problem that, if all the inhabitants of a dungeon level work together against the party the CRs can get out of control quickly, especially at the lower levels.

A solution would be to find ways to reward or punish the players for their tactics in ways that don't necessarily result in overwhelming force slaughtering that players in the first room. I think the key is looking beyond having the monsters team up, and coming up with more varied responses

Don't think "coordinated response", think "every monster for himself"

In your example (a room with guardian monsters, a kobold in another room and two wolves in a third), you could decide that the wolves start barking if they hear fighting in the first room, alerting everyone in the complex that intruders have arrived. The kobold then grabs his loot and flees further into the dungeon.

Now when the PCs get to his room they don't get his loot or the XP for defeating him, and he has time to prepare some traps further down the hallway. Also, the denizens deeper in the dungeon have time to prepare, so when the party arrives they're armed with bows and fighting from behind overturned tables. The party suffers the consequences of not being stealthy with a few new traps and some slightly harder encounters, but not enough to end in a TPK.

There are dozens of responses a monster might have to an intruder in their dungeon that don't necessarily mean teaming up with the other monsters nearby- they might flee, they might hide, they might bargain with the intruders- heck, they might decide to set some furniture on fire to slow the intruders down.

Also, you should always be balancing on the fly

As DM, you should never feel confined by what's written- the author didn't know your players or the composition of their party, and you need to feel free to change encounters on the fly as necessary. Again, in your example, if you decide one of the wolves joins the battle alongside the ceiling creature, when the party gets to the wolves' den you can decide there were always three wolves waiting there, or one, or none, depending on what keeps the game flowing and everyone at the table enjoying themselves. If the party is at full strength when they reach the kobold's room, make it hold six kobolds- if the party staggered out of the previous room dragging the corpses of two of their friends, have the lone, elderly kobold surrender right away and turn it into a skill challenge to get him to talk.

\$\endgroup\$
0
9
\$\begingroup\$

One way to avoid static dungeons is with random dungeons

This is very much a matter of taste, but we had a number of dungeons back in the very old days where most of the rooms were empty; there were very few rooms with designated things in them. Most of what the party encountered were traps and random monsters. But wait, some of those "random monsters" could become allies, or enemies, based on a parley.

And by random I do mean that the DM, or GM when I ran Empire of the Petal Throne(EPT), didn't know what was coming until the dice were rolled.

Every unit of time (in-game time, be it turn, hour, half day, whatever) the DM rolled a d6 (or other sized die) and a 1 indicated a monster. Then, the DM rolled on a table and whatever came up was what the players encountered. (There are encounter tables in a lot of versions of D&D, not sure how many other games have them. EPT had them. You can make your own tables based on CR easily).

The random monsters were presumed to be "just passing through" or "taking a look at the old dungeon to see what snacks (wait, that's adventurers, darnit!) were passing through". The NPC/Monsters' operating attitude was something like "What are they serving at the cafeteria tonight?" - or they were checking out this abandoned underground area, or this old castle, or this cave complex to see if they wanted to homestead. Squatters rights, and all that.

This made for a very different dungeoneering experience than the more story / scripted adventures that are so common these days. The players had to decide to "fight or flee" on the spot and the decision mattered.

Yes, TPK was a thing, but players adapted quickly and in my experience it was quite rare.

We had a blast.

I can only recommend that you and your group try this. You may like it, and you may not, but it is worth a try if what you find published isn't doing it for you.

\$\endgroup\$
0
6
\$\begingroup\$

What I do

I basically run my dungeons by what I believe "would" happen, given the nature of the creatures in it. I make concessions at times, but I generally treat my enemy creatures as any other NPC, with its own thoughts and prerogative. Below are the common examples of situations I've GM'd.

  • If the enemies between rooms are intelligent allies, they may run for reinforcements
    • If the party becomes overwhelmed (and I balance the response based on current circumstances; the party doesn't necessarily know what other enemies are available to respond) then they should be making the decision to control enemies or retreat.
  • If the enemies between rooms can hear combat, I use a couple considerations for whether they respond. I find that these questions usually bring about a logical, dynamic conclusion for the party to have to respond to. I think about things like:
    • Is it this creature's duty to stay put and guard its area, and do they care about their duty?
    • Would this creature care if the next room over's occupants die, and more importantly, do they have any reason to expect that they wouldn't be able to handle the disturbance? (as the GM, I know the party will probably kill the mooks, but the mooks tend to be used to surviving so far)
    • Does the creature next door have a thirst to kill things, or is it more likely just going to defend its cave/pack/loot?
  • Is adding this creature going to be fun for me and the players?
    • At the end of the day, realistic encounters are usually fun, but fun encounters are always fun. Your job isn't generally to kill the party or tell the story that someone put on paper; it's to challenge the players with achievable struggles. If having the castle respond with proper coordination will absolutely TPK the party, I won't do that. There are plenty of reasons they wouldn't be able, such as not being prepared for an invasion, or having other orders. On the other hand, it takes time for creatures to respond. Sending waves of enemies at the party can create interesting scenarios.

Its those interesting scenarios that I search for in my own dungeon design and my players have responded well. Some sessions have their tone set by one player's audible response to a creature retreating. There isn't a right answer, but you'll know what your players think 80% of the time and can adopt future encounters to suit them. Some people are more than happy to hit a static dungeon like a hack-n-slash while others prefer to feel like they're in a real world. I walk that line of realism, with the intent of allowing the players to succeed unless they make themselves fail.

\$\endgroup\$
1
  • \$\begingroup\$ +1 Good points. In the dungeon I'm currently designing, the goblin inhabitants in the rooms are separated by long twisty corridors & are either very loud (eating, drinking, gambling, etc.) or else are sleeping, so won't hear or respond to a fight in other parts of the dungeon unless warned. Plus like you said some monsters will naturally guard their nest (the giant centipedes in the rubbish heap) or else are behind locked doors (the lair of the giant spiders which the goblins prefer to avoid). \$\endgroup\$
    – RobertF
    Apr 18 '20 at 16:52
3
\$\begingroup\$

I also have this problem with most published adventures. From my point of view...

You need a story!

Why are the monsters there, indeed? Why don't they work together to defeat the PCs and totally mess up the CR? These questions need answers...and I answer them by making sure my encounters have a coherent story behind them. Yes, it takes longer to write such a scenario, but I think the extra time is worth it (even if my players just mindlessly murderhobo through the whole thing...if they had paid attention to the story, they would have gotten more XP and loot.)

One level I made recently was an eccentric inventor's workshop. The traps were defense mechanisms, set to repel anyone that broke in. Most of the monsters were constructs that he had been repairing. The poltergeist in the kitchen was a chicken from the coop that had been killed for food; letting all the chickens go free banished the haunting. (Yes, it was a poultrygeist.) The elegant lady the PCs met was a clockwork android; she can do household chores, and doesn't converse above the level of a chatbot, but she doesn't understand death, so she's been tending the inventor's corpse, keeping him from rotting much.

I wish I could find the Reddit post where someone asked for an adventure involving animals disappearing from the zoo, but I offered that some of the animals had become awakened, and were helping their fellow animals escape their prison. Some fey had infiltrated some tunnels below the zoo, and were fighting against what they considered to be an abomination against nature. Most of the fey were content to quietly awaken and free animals, but others wanted to confront the zookeepers openly. And so on. As the PCs explored, they would unravel more and more of this story, and may even decide to form an alliance with some of the fey they encounter (instead of just mindlessly slaughtering them...they'd get the same XP, but conserve their resources better).

Here is another example from Reddit that I was able to find; the only adventure seed was an underwater cave, and after a bit of work, I wrote this story.

In summary...if you're not writing a story, then your scenarios will necessarily feel static.

\$\endgroup\$
3
\$\begingroup\$

Barriers Hinder Cooperation

A barrier can delay or bar a response to a disturbance.

Doors

A sturdy door might be sufficient to prevent entry to fairly strong opponents that cannot open doors. A barred door will delay opponents that can open doors. If the party can briefly spare one person to close the door after hearing animal noises from beyond it, then dealing with that threat can be delayed (and the party can choose to retreat safely for rest).

If the party does not think of closing/barring a door, the DM might decide that the animals flee rather than attack the party. This will warn others, who might flee, set up defenses, or marshal their resources for an attack. Prepared defenders/attackers will be more challenging, but the party will also have more time to defeat the current opponent and to decide whether to retreat or proceed with caution, knowing that the animals are not far away.

Do Not Enter the Fire Swamp

If the shortest path between the party and the main force includes a dangerous area, the main force might be substantially slowed, use an slower but safer alternate path, or choose to let the intruder pass through the dangerous area first.

  • A passageway that recently suffered a partial collapse might be considered safer to avoid.
  • A large area carpeted with patches of fungal growths that spew nauseous spores (very unpleasant and slowing but not longer-term harmful) would be avoided and traversed slowly by denizens even in an emergency.
  • A path crowded with traps will not be rushed through to attack the party.

The alerted denizens might be waiting just outside the Fire Swamp, making that encounter more difficult. However, the party will be given the opportunity to recover from the previous encounter before entering the dangerous area, and the dangerous area might both induce caution in approaching the waiting enemy and provide information about a future encounter (e.g., day-old goblin vomit found by a patch of fungus).

The perceived danger could also be an outdated taboo or something much easier for the party to manage than the local residents. Even something as trivial as the light sources used by the party might reduce the danger — e.g., being able to see discolored stones or weak-sighted cave scorpions finding even lantern light uncomfortably bright.

Constricted Passage

If the shortest path to the main force is constricted, a response might be delayed in reaching the party. In addition larger creatures may not be able to use that path or be extremely slowed. If the one alerted by a disturbance has to traverse the constricted passage (audible alarms warn the attacker and the creatures might prefer ambushes) and the response then has to traverse that area, enough time might pass to allow the party to finish the earlier combat.

A section inhabited by small creatures might be impassable to a large creature. A somewhat playful use: a low-intelligence head warrior might use a growth potion ("Me giant now. Crush puny men.") only to be trapped in the more spacious areas or forced to crawl/worm through small passages. By the time this warrior is encountered, the potion's effect may be nearly over (and the DM can intentionally time it so that this happens). Even if the party encounters this warrior before the potion is quaffed, the party could still escape (a gigantic goblin crawling through a narrow passage will block the passage).

A wide crevasse/pit with planks set aside in another room might significantly delay the response to a disturbance. If the party scouted ahead, the scout might move the planks so that a response coming from that direction would be significantly slowed; leaving behind one sabotaged plank might be even more effective.

Variable Responsiveness

Fools Rush In

If the creatures alerted are intelligent, they may choose to meet the party in a more defensible position. They may also attempt to flank the party by sending a significant portion of their strength around behind the party. While Flanking is often good tactics, it also presents the possibility of easier defeat from divided forces.

Ambush predators will not rush to potential food and some humanoids prefer traps and ambushes.

It's Just a Rabbit

Even if a community fights as one man, bringing out the holy hand grenade any time one sees a rabbit is not normal behavior — most rabbits are not Killer Rabbits. For intelligent creatures, the initial response to the sound of combat would likely be alerting the community to give time for preparation and sending a significant but not especially powerful scouting force to handle the threat or at least determine its nature.

The expected intruders would typically be incidental wanderers (easy kill), conquering forces (organized response), or raiding parties (defend valuables). Commando raids are not the normal threat.

This is not as effective when the allied inhabitants are few (as the first response party would be most of the creatures anyway) and is not normal for creatures with a pack response (e.g., ants, wolves) or when all are relatively close and don't have to grab weapons much less don armor.

A Strong Defense Is the Best Defense

With an intruder, there is less expectation of spoil and more concern with being despoiled. This inclines more intelligent creatures to marshal resources, secure valued possessions, and man defensible positions (or get the heck out of Dodge). This can give the party time to recover from an initial encounter, to reassess the situation, and to prepare for danger.

This can also lead to concentrations of force away from incidental dangers (whether wandering or stationary), distributing the risk over time, and away from treasure valuable to the party but not the inhabitants. Goblins will generally not set up defenses around a library — even if they have discovered the wonder of toilet paper. The grandson of the village elder the party is sent to rescue may have fallen into a pit or been caught in a hunting trap rather than be held captive by dungeon residents or may use the diversion to attempt escape and so be found away from concentrations of defenders.

Creatures seeing the party as food would be an exception, but many of these would not rush in concert to possible food even if they are not ambush predators. Different creature types might have different territories and the strongest member of a group may get the first chance at tasty morsels with the others coming later to scavenge leftovers.

'Gik the Wimpy No Fight Tarks'

Fight is only one of the standard responses to a threat. Flight and freeze (hide) are other options that may be attractive to weaker or less invested creatures. It is not unreasonable for a young goblin to hide under his bed rather than fight alongside his skeleton friend (Nobody Likes a Goblin).

Undetected Activity

Partying Like It's the Last Day of Your Life

If a group of intelligent creatures are celebrating a successful raid or even just having a noisy coming of age ceremony, the noise of combat in moderately distant area might not be noticed. Even if guards are positioned to detect such a disturbance, they might be resentful of being excluded from the celebration and less attentive to their duties or less eager to raise an alarm ("You get Urg! I like my head attached.").

The Sound of Silence

If an intervening area has permanent magical silence, a guard might be placed just outside of the silent zone. This guard would only see the party when they enter the silent zone, so no rushed attack while the party handles its first encounter. If the guard attacks (after sounding the alarm) when the party is fully within the silent zone, the party may be able to defeat the guard before reinforcements arrive or choose to retreat and the guard decide to let the reinforcements pursue the party.

The locals could exploit this silent zone, particularly in attacking a spell-casting party. Alternatively, the locals might avoid the area because it is just too spooky.

The silence might have been set up to facilitate mediation by novice monks, to isolate a wizard's study/workshop, or just as a prank by some crazy wizard ("Bongzar the Inscrutable Was Here").

The Sound of Thundering Waters

Natural noise like waterfalls can mask combat even if nearby. More moderate noise might be annoying enough to the inhabitants that doors are kept closed; the clamor of combat mixed with moderate noise and muffled by a door may not be clear enough to draw a response.

Uncertain Activity

Was 'Alarm!' a Long Flat Trump or an Ascending Trump?

An inexperienced guard might use an 'All Clear' signal (close enough to the normal time) rather than 'Alarm!'. This implies that if the guard is defeated, a limited time would be available before the lack of 'All Clear' would raise an alarm. An individual might be sent to check on (and reprimand) the guard, but a wrong or unclear signal may not draw a strong urgent response.

We Know You're Out There, Somewhere

If the party is approaching from one of several paths connecting to a large echoing chamber, the reasonable response may be to gather a defensive force at that chamber rather than guess which direction to attack. Anyone hearing the combat may run back for reinforcements rather than try to determine where the party is.

The Orc That Cried 'Elf!'

If false alarms or drills have been common (or recent), the reaction to an alarm might be slow and half-hearted.

Ignored Activity

No One Escapes Shelob

If the initial encounter is with a formidable, commensal (not domesticated) guard creature, the "allied" denizens might not respond urgently to sounds of combat — they might even just laugh that they won't have to feed her for a few more days. A small scouting party — quite possibly the weakest, lowest status members of the group — might be sent to check on the result (and possibly gather any spoil) substantially later (while the intruders are being digested). Although the scouts would be cautious when approaching the area of danger, they would not be expecting the party and might be less cautious outside of the danger area.

If the greatest threat encountered in living memory was a lone hunter seeking shelter or a brown bear, the other occupants of the complex would not be expecting anything to survive. The adventurers might be tougher, more cautious, more prepared, or just luckier than the other denizens expect. Even level 1 adventurers are pretty formidable compared to a single commoner or a wild animal.

Remnants from previous meals might indicate the proximity (but not close proximity) of other, intelligent occupants. Humanoid bones without tools or clothing or bones with butchering marks could indicate a fed guard monster. Remains from unfortunate intruders may provide equipment. An adventurer's remains might give (potentially out-of-date) information about what threats the adventurer was expecting.

Somebody Else's Business

A fellow resident of an area may not be an ally. Fleeing or hiding and letting the others deal with the incoming threat is reasonable behavior for weaker creatures (even if their cooperation could have tipped the balance). Letting the mooks handle the intruders is reasonable behavior for a stronger creature or group of creatures.

Bound By Duty

A guard might be ordered not to leave his post. While the guard would sound the alarm when a threat is recognized (e.g., the sound of nearby combat), the guard might be much closer to the threat than to reinforcements. By the time reinforcements arrive, the party may well have completed the earlier combat and had time to evaluate their condition.

Similarly, a guard — who may be one of the more powerful combatants in a group — may not be sent to deal with an ordinary intrusion. The object of the guard's duty might be considered more important than avoiding a few combat losses, especially if the attack might be a feint.

Encouraging Forethought

Encourage the players/party to think about how to avoid having to fight every creature in the place at once.

Leaving Behind Special Gifts

If the party places distractions (spoil/food) or hazards behind it, a flanking response may be delayed, divided, or weakened.

  • Some gaudy trinkets or fresh meat may distract creatures seeking to attack the party from behind. The creatures might be split between those rushing to attack and those getting the goods, contend over the spoil, or be delayed by having to enforce order.
  • Caltrops in the direct path would typically slow (and weaken) a flanking group but do little to hinder the party's flight as they would know to go around the danger.
  • A steepish slope presents an opportunity for greater delay. Oil or cave slime may be applied to make it slippery. Ball bearings at the top may lead to a pile-up if a creature falls down on ascending comrades. With orc-like personalities, this could lead to significant delay and potentially even a reduction in fighting strength. Since the party would be aware of the danger, their flight might not be slowed much (sliding down the slope might be faster than a normal careful descent).

Who Cares Who Brought the Beer, Let's Party

If the party earlier found a cache of nearly undrinkable (to humans) rum, they might think to leave a broken-down cart with barrels of rum on an ill-used path near the lair of goblins. Attacking while most of the goblins are drunk or hung-over is likely not only to delay or avoid sounding an alarm but also to make the response much more staggered. Drunk or hung-over opponents also act less intelligently, fight less effectively, and are less inclined to pursue aggressively.

This is just one example of the DM providing opportunities through ordinary low value resources ("Oh, what I wouldn't give for a holocaust cloak!") to make difficult challenges easier or more survivable. While the DM should not try to railroad the party into a specific plan (and not all players like strategizing or have a party that can be roleplayed that way), making resources available and even hinting at a possible uses (Player: "I dip a finger into the barrel and taste the contents." DM: "The barrel contains a rum so bad only an orc would drink it.") increases the odds of success or survival.

I Know Where You Live

If the party researches or scouts the area, the approach taken can account for possible cooperation and possibly even exploit conflict, hazards, bottlenecks, and timing of activity.

Part of this will often be simply part of the motivation for the adventure or rumors about the area. If more inhabitants/dangers are expected than can be handled at once, the party should be expected to be prepared for retreat and to use unusual means to limit coordinated responses.

E.g., knowing that wolves are one of the threats, the party might decide to approach in the later afternoon and enter in the early night when the wolves could be expected to be out hunting. Wolves remaining in the den will not be likely to leave the den to attack the party but will alert the other inhabitants (and the party).

If one likely opponent is too fast to run from, avoiding (going around, blocking, or distracting) or prioritizing (fighting while freshest) that opponent may be good tactics.

Even discoveries made while exploring can provide information about occupants and their interactions well before alerting them to the party's presence. E.g., a goblin skeleton at the head of one passage out of a chamber might tell the party that something down that passage does not want to be bothered by goblins and that goblins are probably elsewhere in the area.

Failure Is an Option

The party should be aware that becoming goblin poop is even less heroic than running away.

Noises like wolves howling or the sounding of a gong may give the party sufficient warning that they need to decide whether to retreat, use precious resources to finish an encounter more quickly, or prepare to face additional opponents while still dealing with the current opponent.

Planning for retreat can include disabling rather than moving around traps (playing hopscotch to avoid dart traps while fleeing the goblin horde is not wise, though putting a plank — that can be quickly moved off — over the triggers may be clever), having the means to slow pursuit (caltrops and ball bearings are somewhat affordable and light), marking the way to the exit (avoiding wrong turns or having to think about which way to go, though this can also alert patrols to the party's presence).

I Don't Know, I Just Live Here

Residents of an area are not necessarily familiar with all the paths through it. If the party discovers a secret or less observable passage, such might be used to hide from pursuit after a nearly exhausting encounter with the sound of reinforcements approaching. A party that explores — rather than hurrying from one combat to another — may find resources helpful later in the adventure.

Dynamically Adjusting Difficulty

Even if the party previously scouted the area and took a head count, adding or removing creatures is not necessarily unrealistic. The return of a raiding party or the visiting of a potential ally could add to the count; the opposite could reduce the count. The condition of the creatures can also be adjusted; sick or drunk/hung-over creatures are less capable and less likely to flee or act intelligently. Communicable illnesses might be a double-edged sword, helping the party succeed but keeping the party from returning immediately to civilization.

The usefulness of spoils can be adjusted dynamically. E.g., an unidentified potion can be mere treasure or a critical resource. Quaffing a believed safe but unknown effect potion can be used to give the party a chance to survive or escape an encounter without enabling the party to curbstomp their opponents. Using the potion to buff a character during a normal encounter might result in a useless effect (fire resistance might not be very useful against an ogre with a club), but used in desperation the potion might give just enough help to make survival likely.

Avoiding single entrance designs can facilitate escape (of residents or the party), allow the party to come from an unexpected or uncommon direction (initial disturbances might not even be recognized as signs of intruders), and provide an unexpected path for reinforcements (if the party is having too easy a time). This can also help to distribute dangers (both mobile and stationary) to reduce the risk of all the threats appearing at once.

Conclusion

Some of the above (particularly in "Undetected Activity" and "Uncertain Activity") should not be normal features of play but may not break the suspension of disbelief if used sparingly and sensibly (and perhaps with a little lampshading). Some of the above merely says that not all creature rush to attack intruders even if allies are in danger or an immediate pack response would be more effective.

Access to information (particularly sensory warning about coming danger and indications of the degree and type of dangers) can greatly help the party make good decisions about handling risk. In-dungeon resources such as defensible positions, places to hide, easily obstructed paths, useful equipment, and information about the nature and location of dangers can make a challenging task more survivable.

Restriction of information can allow the DM to dynamically adjust difficulty based on the circumstances. Mild, infrequent Deus Ex Machina coincidences like having the least encumbered and highest Dexterity member crossing the rope bridge when it breaks, having the dwarf hit by the poison dart, or having a falling block trap get stuck (so only tall characters take moderate damage on a failing saving throw rather than characters being crushed beneath tons of granite) — these can alert the party to danger if they are not being sufficiently cautious or nerf accidental failures without removing the in-story danger. Having the party saved at the last moment by Alexander Smugbottom the Magnificent may be a fate almost as bad as death, especially if the tale is recounted in taverns for miles around; such may add flavor to future activity ("You're the schmucks Smugbottom had to save from an elderly kobold?!" or "You're also a victim of Smugbutt the Magnifable's lies?").

\$\endgroup\$
2
\$\begingroup\$

Make the dungeon a character

I don't know about you, but I find that in any kind of RPG, whether tabletop or video game (my experience is mostly with video games, but the same rules can apply to both), having the dungeon just be a place, a passive location to go around and find monsters to kill, is just boring and repetitive. You need to make each dungeon unique. Treat them the same as you would treat a major NPC. A dungeon should have a back story, it should have a "personality". No two dungeons should be alike.

I had a DM back in the day who had a problem with this - we started to notice that every dungeon started with 40" and a 4-way intersection, and it became a running gag. We even brought it to his attention, and the most variation he ever made to it was one where it was 30" and a 4-way intersection. But generally speaking, each and every dungeon had only minor differences in their layout, and the only thing you could ever do in them was find a room, fight some monsters, and repeat.

Here's just a few examples of what you could do to add some variety:

  • Add traps. There's more hazards than just monsters.
  • Add secret doors, hidden passages, etc.
  • Add puzzles. This can be anything from literal puzzle mini-games (push levers until the doors open) to logical physics-type puzzles - think games like the Zelda series. Less of the "Where am I?" and more of the "How do I get over there?".
  • Think 3-dimensionally. Add height to rooms, allow attacks from above, below, across chasms, from balconies, etc. The entire dungeon could be sinking into a swamp and the floors are all at a 30 degree angle and covered in slippery mud. Hope you guys brought ropes and grappling hooks.
  • Get outside of the box. If your dungeon was designed as a building, sure, it's okay to have the place fit perfectly on grid lines. But if it's a natural cave, it's ridiculous to have all the walls defined by straight lines and right angles.
  • Turn out the lights. A cave should not be lit up like a castle. A torch or magical light source should be required in some places, and both of these can be extinguished. Make the players fumble around and get lost in the dark for a bit. If the players are making a map based on your description and it's 100% accurate to your map, you might be doing it wrong.
  • Make the environment change around your players. Walls move, tunnels collapse, entire rooms rotate, etc. Just because the players have been in this room before doesn't mean it's going to look the same.
  • Split the party. Drop a gate between them and force them to go separate directions, make the rope bridge break after only some have crossed, or put in obstacles that not everybody can cross based on their abilities. You could even have things that need to be triggered from two distant locations simultaneously. It can really change the dynamic having people have to explore in smaller groups, where you don't have all of the party's abilities at hand. It forces you to change your battle tactics when you don't have your healer nearby or the like.

This is just a sampling of the things you can do to make your dungeons unique and provide variety to each location your players have to explore.

\$\endgroup\$
3
  • \$\begingroup\$ These are all excellent recommendations, but I don't feel like they apply to OP's (admittedly vague) definition of what is "static" about a location \$\endgroup\$
    – Ifusaso
    Apr 17 '20 at 11:37
  • \$\begingroup\$ It sort of appears that you've misunderstood the question, which revolves primarily around NPCs interacting (or not interacting) with fights in other rooms. \$\endgroup\$ Apr 18 '20 at 21:34
  • \$\begingroup\$ @MooingDuck No, I reframed the question. I saw that the OP was putting too much emphasis on just the NPC encounters and wondering why their dungeons seemed too static, so I offered ways to make them more dynamic beyond simply going room to room fighting monsters. \$\endgroup\$ Apr 20 '20 at 13:14
1
\$\begingroup\$

GM and player play styles and house rules can make a huge difference, and that session-to-session, players vary from sharp and switched on to distracted or not tracking well. You absolutely have to judge on the fly. I consciously adjust monster HPs, some attack rolls or saves, etc. as necessary to deal with 'The A+ Team' showing up, or the other way if "The D- Team" shows up.

Sandbox adventuring (you setup locales and major actors with their plans, the players decide how to interact without any predetermined story arc) requires this sort of adaptability all the time, but even the plotted adventures need a lot of flexibility on delivery from a GM.

TSR's B1: Keep on the Borderlands has a cave complex filled with quite a varied mix of humanoids. What's in there could, fighting smart, take out patrols from the Keep and easily wipe the floor with neophyte adventuring parties.

The counter-balance comes in these respects:

Players need to learn to think a lot about the situations they get themselves into. Hire some extra muscle (mercenaries). Scout as much as you can. Try to ambush a humanoid raiding party and take a hostage to gain intelligence. Discover the relationships between the humanoid groups and figure out if there is a chance to develop that friction (perhaps pick a faction to aid, like helping the weaker ones rise up against the stronger factions). Infiltrate and try to form a connection that you can use to direct one of the factions while gathering good intelligence. Know what you are getting into and how deep. And watch for the first sign of things about to turn on you and have a prepared exit strategy with pursuit deterrents. Try to get the powers in the Keep to support some strikes, ambushes, etc. (talk to the Castellan, local churches, etc).

From a DM perspective, the humanoids have a pecking order, but it is quite Darwinian and is subject to change. If one group of humanoids gets raided and hammered, they might take out or brutalize the human party, but then their competitors (or weaker humanoids who now are less intimidated by the number of the stronger ones left) may rise up against them. Maybe the Orcs should ally with the Bugbears. Treachery isn't out of the question - say you're with them and either dessert the Bugbears once engaged or attack them (assuming that will put the humans in a mood to accept that).

The DM can have humanoids that will parley and will give the players (with some good prodding) information the players can use to choose sides or to manipulate factions. A good trading relationship with Golbins might give you inside info to ambush Orcs or to get warning when Bugbears are going out on a night raid. Also have some of the humanoids worried enough about their own hard won space that they won't rush to aid 'allies' who might turn on them or who will abuse them in any case.

You can have some sort of ideas of the different leaders and seconds and see if there are internal strifes within a faction too that could lead to alliances or at least neutral status or might even let a party get some help (even if it is not responding during an attack on one part of a faction).

With a varied bunch of personalities and a bunch of different motivations and weighing consequences of rushing to help (which might not be all sunshine and roses), you can justify not dogpiling. And with players learning not just to fight, but to negotiate, bluff, confuse, manipulate, interrogate, etc, you will have the party looking out for its own wellbeing more often.

I tell you one Old-School Lesson: If you kill of a party once because they weren't bright and you point out five ways they could have had a better outcome, you bet that the next time, they'll keep those in mind. Even a good near death could generate that revised degree of care and planning.

\$\endgroup\$
0
\$\begingroup\$

Dynamic does not mean a simple, quick pile-on.

I like to have sensible organization by the intelligent monsters. It should feel kind of real, and that means it is dangerous to take on a big, organized faction if it has time to respond by plan. So both you and the players should be thinking about tempo and risk level as it plays out.

An organized group over several linked rooms should do things like have some sentries out at the edge of their territory whose main task it to raise the alarm and retreat, so the adventurers end up having to deal with them quickly and stealthily or have to face big and accumulating threats as the faction gets mobilized. And defensive strongpoints intended to hold them up while things get organized for the counterattack. It should become really dangerous if they lead with their chin against organized foes of a comparable threat level.

When they do get a fight a tough room, picking off that guy that is trying to get away to warn the neighbors should be a thing of importance to the party, as well as a way of dropping the immediate pressure since he isn't contributing to the combat.

But mobilization shouldn't be instant or frictionless, you can avoid a too extreme snowball by having some of the nearby defenders rush in half equipped, some more take their time to equip, some others cowering, panicking or retreating to stash loot in more secure rooms. Details help. The time it takes to put on armor is a lot longer than the typical combat encounter. So if some defenders have to do that, it meters either the pace or the strength of the response. A mix of ferocity and hesitancy or panic among the defenders can meter the response rate. You can adjust based on the quality of the opposition using morale dice of both the group they are already fighting as casualties pile up and of the neighbors considering piling in, and on the fly by GM fiat to make it tough, but not impossible. Things like tactical retreat by the PCs or monsters, and bottlenecks & chokepoints in the area should all come into play.

Rival groups

The snowball effect can also be leavened by having multiple factions within the dungeon with a mix of relationships, in part so they don't all just rally to each other's defense and in part so some groups might even try by diplomacy or stratagem to get the PCs fighting their rivals instead of themselves, to mix up the kind and pace of encounters. Some monsters that join a fight might arrive to hit a rival faction in the back while the PCs are pressuring their front, cutting down the threat level of a situation that was looking overwhelming and giving a possibility for a role playing encounter with the second group after the fight. Unintelligent large monsters can arrive, grab a quick meal from among the combatants, and take off with it, giving forewarning of their presence, and adding an urgent new issue if it is a PC rather than a defender that is being carried off to be eaten.

\$\endgroup\$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .