Friday, May 17, 2024

D&D is a Problem-Solving Game

The argument that D&D is combat game goes like this: A game is "about" what it has rules for. Because D&D (specifically 5e) has far more rules for combat than for anything else, it's a combat game. 

I don't really have or care to have an opinion as to whether "system matters", aside from the simple fact that one should play a system they enjoy, and that yes, there are other games besides D&D which facilitate an experience that is not like D&D. D&D is not the all-encompassing system that some people treat it like it is. I don't want to play a 5e-compatible farming simulator or whatever, but I also don't care much if someone else decides to till that barren soil.

I do think things are a bit more nuanced than "If most of a game's rules pertain to X, then the game is about X." So, I wanted to try to articulate what kind of game I think D&D is.

Is D&D a combat game?

I see it broadly like this: D&D is first and foremost a fantasy roleplaying game. More specifically: D&D is a fantasy problem-solving game, with a greater or lesser focus towards combat as the primary problem-solving method, depending on the edition, DM, and players. Different editions have greater or fewer rules for things other than combat. Different DMs prepare greater or fewer scenarios which they anticipate the players will resolve using combat. Different players are more or less inclined towards fighting everything their characters encounter.

5e, allegedly, specifically leans towards combat because it has a lot of DM- and player-facing rules for it. There are a lot of monsters, for example, and most of the statistics for those monsters are oriented towards fighting them. Most character abilities are combat abilities. The DMG goes into great detail as to how to create a combat encounter.

The very specific guidelines for building combat encounters (plus online tools to assist in doing so) makes it relatively easy to quickly create a combat scenario that is "balanced" for the player characters' level to the DM's taste (i.e. easy, medium, hard, or deadly).

If the DM has a session coming up, and nothing prepped, it's easy to open the Monster Manual, pick a monster for the player characters to fight that week, and build an appropriate encounter which will eat up a lot of game time. It's easy to force the player characters to fight something - the monster of the week is causing problems, and if the player characters go investigate, it attacks them and tries to kill them - now they have to kill it. It's a convenient crutch for the unprepared or inexperienced DM.

It's not so easy to instead create a non-hostile NPC to oppose the player characters socially, determine why they're opposed to the characters, what would convince them to leave the characters alone, and how hard it should be to convince them to do so. It's not so easy to design such an encounter to take up all or most of a session. It's not so easy to force the player characters into a social confrontation with the NPC.

But it's also not impossible. The DMG has guidelines for creating NPCs with ideals, bonds, flaws, and goals, and for adjudicating social interactions with those NPCs. They don't read as strongly as "rules" because, by comparison, combat has numbers and statistics and all sorts of things that appeal to people who think that numbers = rules, and combat in D&D is expected to have that structure. 

Social interaction is looser, and more nebulous, and it should be because it mostly involves person-to-person interaction and only the occasional rolling of dice, informed by predetermined NPC characteristics (again, provided by the DMG) and DM judgment. But because there are fewer hard numbers and because lifelike people in a fantasy world are somewhat hard to conceptualize, many DMs shy away from prepping these types of scenarios in the same way they prep combat scenarios.

So, inexperienced or underprepared DMs are going to tend to lean on combat a lot, prepping many encounters to which the only solution is fighting. Players under such DMs are going to learn to expect to fight most of their problems. If that is one's experience with D&D, it's easy to see why so many people see it as a "combat game".

Ancient Mechanisms

Compared to 5e, older editions have far more mechanics which support modes of play other than combat: rules for generating dungeons and wilderness environments, stocking them, and structured dungeon and wilderness exploration procedures make it easier to prep an exploration scenario; reaction rolls make it more likely that not every monster will immediately attack the player characters - some may even want to help them; morale rules make monsters less likely to fight to the death, which can turn combat encounters into chase scenes or social interactions; player characters gain most XP from acquiring treasure, not killing monsters. 

In the latter case, the goal is to acquire the treasure by any means necessary - if fighting can be avoided, then that's great, because characters can advance without undue risk. The problem to solve is getting the treasure, not killing the monster.

Pretty much all of these mechanics had been discarded by the time we got to 5e: the DMG has a pretty lackluster dungeon generator (with monsters in a whopping 50% of rooms) and an unsatisfying optional morale mechanic; reaction rolls appear only in Bigby Presents: Glory of the Giants, for some reason; experience points are "most often" rewarded for combat, not gold, though I argue that instead they should be rewarded for overcoming challenges of all kinds.

As a result, a 5e DM will probably only know about these mechanics from familiarity with past editions - either from having played those editions, from reading the old books, or from word of mouth. This is why I'm grateful for blogs and for the OSR: for turning me on to bits of knowledge that make the game far more enjoyable for me and my players, but which have been lost in modern editions of D&D. The blogs, in turn, have peaked my interest in reading about and running older editions of the game.

The problem is that as D&D develops, valuable tools from prior editions are dropped, and those tools are lost to new players who aren't familiar with the game's history. Most new DMs and players are unlikely to be familiar with any edition beyond the current, most popular one, and the current version of the game teaches new DMs - via the DMG - only how to prep combat encounters, and it teaches players - via their characters' abilities - to expect to solve their problems using combat.

Right?

Sort of. 

Goals of Play

It's definitely true that most of 5e's rules lean towards combat, and that the procedures for prepping combat encounters are very mathematically detailed. However, I want to look at the chapter in the DMG on creating adventures, specifically "Location-Based Adventures", on page 72. That section has tables for dungeon goals, wilderness goals, and "other" goals.

All of the goals presented by these tables can be viewed instead as problems to solve. The "Dungeon Goals" table has exactly one entry out of 19 which explicitly says, basically, "Fight a monster" ("Slay a dragon or some other challenging monster"). There are a few other entries that could be loosely interpreted as "fight a monster":

  • Stop the dungeon's monstrous inhabitants from raiding the surface world [by fighting them]
  • Foil a villain's evil scheme [by fighting them]
  • Destroy a magical effect inside the dungeon [by fighting it]
  • Pursue fleeing foes taking refuge in the dungeon [and fight them]
  • Clear a ruin so it can be rebuilt and reoccupied [by fighting everything inside]
Those are all assumptions, though. The player characters could find out why the dungeon's inhabitants are raiding the surface world, and convince them to stop by other means. The magical effect might not be a monster, but a complex trap to disarm or a restless spirit to placate. The fleeing foes might parley with the characters, or need to be captured (alive) and brought back to civilization to see justice. The ruin to be cleared might be filled with traps and hazards, rather than monsters. 

The other two tables are similarly diverse in the goals they present. What this indicates to me is that 5e is meant to be a problem-solving game, and these tables are prompts with which the DM can generate problems for the players to solve however they wish.

Achieving those goals might involve fighting monsters along the way, but combat is a part of the game (and often, a fun one), and one that both DMs and players anticipate engaging with from time to time.

True, the combat-orientation of D&D's rules might lead the DM and/or players to anticipate combat and prepare or consider no other solutions, leading to a bias towards combat, but that's a playstyle issue. It isn't being strictly enforced by the DMG itself. The fact that the first step in prepping a location-based adventure is to determine the party's goals, and that so few of those goals explicitly state that something has to be fought, suggests to me that player characters are supposed to experience a wide array of goals and problems during their adventuring career.

It's not like fighting is the only skill at the player characters' disposal. They have character knowledge, the ability to perceive and intuit things, social skills, equipment other than weapons, and magical abilities that allow them to understand languages, disguise themselves, charm others, and the like. They can do anything that a person in the real world can do, and more. The players have their own real-life skills and knowledge they can draw upon as well.

The rules for social interaction aren't robust, but they're sufficient; the players (via their characters) and DM (via their NPCs) have a conversation, and when the outcome of that conversation is in doubt, an ability check is rolled, based on the NPC's disposition towards the characters. I drew upon old-school reaction rolls and other materials to make my own social interaction mechanics more robust, because that was my preference, but doing so isn't strictly necessary for this aspect of the game to function. 


Well, that's just great - I added some mechanics that aren't in the rules and now my games are less combat-oriented. That doesn't mean the game itself isn't about combat.

But that's the thing - when I think about the statement "D&D is a combat game", I'm not just thinking about the current edition, or how "most" people play it. I'm thinking about D&D in its totality. It isn't one thing. Not even one specific edition is the same across groups - every table plays it differently, and the rules don't prescribe one type of play. I like my modern D&D to feel more like old-school D&D, and I modify things to suit that preference. Other DMs may modify it to feel more like Pathfinder, or Warhammer, or Dark Souls, or Animal Crossing, or whatever - whether they should be playing a different game is a conversation for another time.

So new players maybe play D&D more like a combat game than anything else, but we don't even know that for sure, and the way that most people seem to play the game isn't necessarily the same as the way the game designers are trying to indicate that one should play. It might be easier for a new DM to prepare a combat scenario, which might lead to new DMs preparing more combat scenarios, but the DMG isn't telling DMs to only ever prepare combat scenarios. 

Could the game be designed better to communicate this intent? Probably. But would people actually read the books and thus grasp this intended playstyle, or would they keep on playing their own version of D&D based on their idea of what the game is supposed to be? If the game is designed one way, but people play it another way, which game are we talking about when we say the game is a "combat game"?

So, is D&D a combat game? Yes, kind of, but also only to extent that the DM and players make it about that. I would probably be disappointed to play in a D&D game where I never got to fight a dragon, but I'd also be disappointed to play in one where I never got to talk to a dragon, or where the DM hadn't drawn up the dragon's lair so that we could explore it in detail. Some players and DMs may just want one of those three things, but the game's designers are clearly saying that there should be a mix of all three.

Wednesday, May 15, 2024

AD&D 2e Play Report: Session 19

We played the 19th session of our AD&D 2e campaign last night. The play report for our last session is here

To recap, the party was locked in battle with powerful aquatic zombies - we had started the fight the previous session, so this was a continuation of that. Corym Vadercast was unconscious, and the party was trying to extract him. They slugged it out and eventually won the day, but not before their thief henchman, Sheyla, was killed while attempting to retreat. The party cut their losses and decided to return to town to regroup. Along the way, they encountered some friendly wood elves, and agreed to return with them to their settlement. They recuperated there, recruited seven elf hirelings, and returned to the tomb.

There, Karven Stone scouted ahead and found the heretical priest, Themar, guarded by ten zombies, with another eight zombies excavating a pocket of necrotic energy to the south, across the river. The party set an ambush. Karven backstabbed the priest. Bernhardt Dalton turned the majority of the larger group of zombies. Corym, Orlina, and Yinvalur Sparkguard charged in to deal with the stragglers. Haymond Baler, Yuliana, and the elves launched a volley of ranged attacks from the south. That's where we ended the session.

Party Roster

  • Bernhardt Dalton, NG human cleric 4
    • Magicheart, NG pseudodragon
    • Orlina, CN human cleric 1
    • Yinvalur Sparkguard, NG elf fighter 3
  • Corym Vadercast, NG elf fighter 3/mage 3
  • Haymond Baler, LG human fighter 4
    • Yuliana, NG human cleric 2
  • Karven Stone, LE dwarf thief 5

This should be a pretty brief play report, since most the session was spent in combat, and I'm not going to do much play-by-play. We started the session by rolling initiative. 

Bernhardt continued turning the undead, who fled in the opposite direction and into the river, getting swept downstream (zombies are not smart). As the current dragged the zombies past, the elves fired their bows, killing the zombies over the next several rounds. 

The zombies in the pit climbed out and pressed through the narrow corridor towards Haymond and Yuliana. Haymond held them at bay while Yuliana turned a few. They were eventually overwhelmed, and Yuliana was badly injured (she had 1 hit point remaining) and had to retreat to heal herself. A few of the elves drew their swords and moved to reinforce Haymond, while the others rushed across the river to where the rest of the party fought.

Corym, Orlina, and Yinvalur battled the sole remaining unturned zombie on the north side of the river. Orlina was also badly wounded (she also had 1 hit point remaining) and had to retreat to heal. 

Karven went toe-to-toe with Themar. Themar is a zombie lord. Zombie lords can target living creatures with an animate dead spell, killing them instantly and turning them into zombies unless they save vs death. I tried to unleash this powerful ability on Karven, but Karven struck Themar with his sword before the spell could go off, so it was lost. Corym and Yinvalur had killed the last remaining unturned zombie on their side, so they and a few of the elves rushed to aid Karven.

Zombie lords also have an extremely potent stench:

The odor of death surrounding the zombie lord is so potent it causes horrible effects in those who breathe it. On the first round a character comes within 30 yards, he must save vs. poison or be affected in some way. The following results are possible:

1d6 Roll    Effect
1               Weakness (as the spell).
2               Cause disease (as the spell).
3               -1 point of Constitution.
4               Contagion (as the spell).
5               Character unable to act for 1d4 rounds due to nausea and vomiting.
6               Character dies instantly and becomes a zombie under control of the zombie lord.

When I prepped this dungeon, I rolled a 6. Uh oh.

The fact that this has a 30-yard range and affects anyone who simply catches wind of it was a bit extreme for my tastes. Looking at the map of the dungeon, this effect had a chance to instantly kill anyone who exited the entrance chamber. I instead decided that it was a special attack the zombie lord could utilize on his turn, and that it only affected people he could see. Now that Themar was being attacked by Corym, Karven, and Yinvalur, I decided to unleash this power.

Dark tendrils of necrotic energy began to exude out of the heretical priest. His vestments billowed, and the characters could see that his body was rotting beneath, infested with maggots and flies. The tendrils lashed out at those surrounding Themar - Corym, Karven, and Yinvalur would need to save vs poison.

Corym and Karven barely made their saves (i.e., they rolled the exact minimum number they needed to pass - Karven was helped by the fact that dwarves get a bonus to saving throws vs poison tied to their Constitution, or else he'd have failed). Yinvalur was not so lucky. The party watched in horror as their chad elf henchman - who had been with them since session 1 and bailed them out of several dicey battles - was struck dead. The next round, he rose as a zombie under Themar's control.

This was the players' first real encounter with a save or die effect. They had battled giant spiders with save or die poison, but none of their characters or henchmen had been bitten in that encounter. Some of them were a bit shocked. "He just dies?" Yep.

The rest of the fight was rather pedestrian. Themar had used his big dangerous abilities, and the rest of the zombies to the south were held at a chokepoint. Karven cut down one of the curtains partitioning Themar's vestibule, then worked with one of the elves to toss it over the priest's head. Themar protested the indignity while trying to crush his enemies with his fists, but he was no match for the player characters and elves surrounding him. He was slain. The remaining zombies, no longer under his sway, stood slack jawed. 

This combat was a bit of a slog once the bulk of the zombies were turned, Themar had used his powers, and the injured henchmen had managed to get out of danger. Still, the death of Yinvalur made a big impression, so the victory had a grave cost. Two other henchmen very nearly died, so it was tense for a few rounds. It was not quite as white-knuckled as the ogre fight from a few sessions ago.

Karven looted Themar's body and found a pouch of 30 gems. He skimmed a few off the top and shared the rest with the party. Bernhardt found Themar's journal, which was in Elvish, so he asked Corym to decipher it. 

Essentially, Themar had been corrupted by the evil god Malyk, who revealed to him the presence of a pocket of magical energy. Themar and his acolytes excavated the pocket, and the dark energy turned Themar into a zombie lord. He had his zombie acolytes continue to dig, and was planning to bring more elves to the tomb to be made into zombie lords like himself. They would then raise an undead army and lay siege to the lands of humankind, restoring elf supremacy in the region, albeit at the cost of all the elves being undead abominations. Typical evil zombie elf cleric stuff.

The party discussed what to do with Yinvalur's zombie, which was still "alive". Could they bring him back? Raise dead, apparently, doesn't work on elves:

When the priest casts a raise dead spell, he can restore life to a dwarf, gnome, half-elf, halfling, or human (other creatures may be allowed, at the DM's option).

 The omission of elves didn't even occur to me until I read the description of resurrection:

The priest is able to restore life and complete strength to any living creature, including elves, by bestowing the resurrection spell.

Resurrection is a 7th-level spell, meaning a priest must be 14th-level or more in order to cast it. I reasoned that powerful priests such as this could be found in the city, however, the DMG list of costs for NPC spellcasting doesn't even include resurrection, and states of raise dead, "This spell is normally cast only for those of similar faith or belief. Even then a payment or service may be requested." 

Resurrection is more powerful than raise dead, and casting it causes the caster to age three years, so one would imagine that this service would come at high cost. I told the players to imagine that they were able to find a 14th- to 20th-level priest in the city who would cast the spell for them. I then asked them to imagine what such a high-level priest would ask of a party in return, and whether or not their 4th- to 5th-level characters would be up to the task. They concluded, probably correctly, that they would not be able to procure this service. Teary eyed, the party put down the Yinvalur zombie, most likely for good.

The good news is, the party still had at least 140 years to bring Yinvalur back at a later date:

The creature can have been dead up to 10 years per level of the priest casting the spell. Thus, a 19th-level priest can resurrect the bones of a creature dead up to 190 years.

The party planned to escort the elf hirelings back to their settlement, then return to Mythshire. We ended the session there.

Bernhardt leveled up. He is now 5th-level, gains a d8 hit die, and a 2nd- and 3rd-level spell slot. Yuliana also leveled up. She is now 3rd-level, gains a d8 hit die, a 2nd-level spell slot, and a nonweapon proficiency.

Friday, May 10, 2024

On Hands (and what PCs can do with them)

Okay, weird post title. Bear with me.

This post was inspired by Warren D.'s post "WHAT'S INTERESTING ABOUT BASIC DUNGONEERING: And what is not" on the I Cast Light! blog. What got me thinking about hands specifically is Warren's thoughts on light and weapons (highlights mine):

Light: Due to torch cost and number per slot, it is easy to carry a lot of torches.

Not interesting: Carrying enough light to last 12-24 turns- easily done

Interesting: When torches extinguish-- like in the middle of a fight or when the goblins you are negotiating with get mad; how many hands in the party are occupied by torches

... 

Weapons: I've yet to find a really good way to do weapons simply outside of 1d6 damage for all types. I don't mind BX's variable weapon damage. And I do like some old rulesets sorta "first strike" if your weapon is larger than an opponents other wise smaller, lighter weapons strike first in subsequent arounds.

So here is what I have got so far: Using a weapon two-handed is a +1 to damage, using an off-hand weapon is +1 to-hit, and a shield is of course +1 AC. I do like that fighters with bows can shoot twice if they did not move and the "cleave" ability.

Not interesting: Weapon factors that are so extensive they require a separate rules discussion, trigger player obsession, and/or orient the whole of gameplay to combat

Interesting: What PCs chose to do with their hands: more armor, more weapons, or more light

What I'm interested in is how player characters choose to use their hands. It's a simple thing, but it's important - a hand can do any of the following things:

  • Cast a spell with material or somatic components
  • Hold a torch or lantern (providing light with which to see)
  • Increase weapon damage (per my rules for variable weapon damage, using two hands increases a weapon's damage die size by one) 
  • Increase a weapon's reach (most weapons with the reach property require two hands to use)
  • Increase a weapon's range (ranged weapons which require two hands, such as bows, usually have a longer range than those which require one hand, such as thrown weapons)
  • Interact with objects (open doors, throw levers, withdraw items from a bag or pouch)
  • Make an extra attack (if dual wielding or a monk)
  • Provide extra protection (a hand used to hold a shield provides +2 to the character's armor class)
  • Use items (drink a potion, read a scroll, activate a wand, etc.)

It's something many DMs might not even think about, and I suspect that most modern DMs view it as one of those things - like encumbrance - which is too onerous to keep track of, and simply don't, beyond "Is the character dual-wielding, two-handing, or using a shield?" I have certainly forgotten about it a good number of times. In fact, when thinking about this, I realized that in a recent AD&D 2e session, I forgot about something very important - light.

Who is Carrying the Light?

The player characters were exploring a network of subterranean tunnels made by a brood of ankhegs. When the party first entered the tunnels, I said, "It's dark, so you'll probably want a light source. Who is carrying the light?" Then, when the party had its climactic battle with the ankheg brood, I never once considered the issue of light. Characters were firing bows, dual wielding, two-handing, and wielding shields. Who was carrying the light? Could the characters even see? I don't know.

When not in combat, the question of who is carrying the light isn't incredibly important. One might as well assume that all of the characters are carrying light unless they state otherwise. Characters might place a torch in a sconce when exploring a room or put it on the ground while they check a chest for traps. True, monsters hidden far away in the dark might target the torchbearer specifically, but in that case, the DM can simply check with the players who exactly is carrying the light before springing the ambush, after which the specifics become more crucial.

Once combat begins, it becomes very important who is carrying the light, because that character can't use their light-carrying hand to cast spells, wield a weapon with both hands, interact with other objects, duel wield (aside from using the torch as a weapon), wield a shield, or use items. That character is making a choice to benefit the rest of the party (by providing light) at the expense of their own tactical versatility. It's part of what makes light-providing spells, weapons, torchbearers, and the like valuable - they can provide light without occupying a hand.

Furthermore, the light source typically moves with the torchbearer, unless the torchbearer drops their torch on the ground in order to free up their other hand. Monsters may attempt to attack the torchbearer and snuff out the light. So, keeping track of who is carrying the light in these scenarios is very important. 

Stow, Drop, and Draw

In D&D 5e, a character can interact with one object for free on their turn, as part of their attack or move action (PHB, page 190). This includes drawing a weapon as part of the action they use to make an attack, opening a door as they move through it, picking up a dropped item, stowing an item, or withdrawing a stowed item. Redditors will happily say that according to the rules, a spellcaster wielding a sword and shield can drop their sword (no action required), freeing up a hand to cast a spell, then immediately pick up the sword again (using their free object interaction).

I hate this. It's stupid. Changing what hand is doing what is a tactical choice. It should have benefits and drawbacks like any other.

Baldur's Gate 3 is the current hotness as far as video games based on 5e go, but another 5e-based game, Solasta: The Crown of the Magister, is much better at little rules minutia like this. In Solasta, each character has weapon sets - a primary weapon set, a secondary weapon set, and a light source set. A character can switch between sets once on their turn.

So if a character is wielding a sword and shield, but wants to make a ranged attack, they switch to their longbow. After attacking with their longbow, they cannot then switch back to their sword and shield to benefit from the improved armor class. If light is an issue, the character can switch to holding their sword and torch, but cannot then switch back to their sword and shield.

This is more or less how I do it in my games, although locking characters into using predefined "weapon sets" is not exactly lifelike. Instead, I implement a stow/drop/draw action, which allows a character to switch what they're holding in each hand once per turn. They can drop their torch and draw another weapon, stow their sword and draw their wand, stow two items to free up both hands, or any combination of those things, so long as the item being held or not is interacted with just once, and the hand switches between only two items per turn. That means no dropping an item to do something with the hand and then picking the item right back up again.

To Two-Hand or Not to Two-Hand

A character who chooses to wield a two-handed weapon is making a deliberate choice. They are doing more damage if they hit, but they are sacrificing increased armor class (from a shield) and number of chances to hit (a dual-wielding character is making an extra attack each round). 

Only, are they actually doing more damage than a dual-wielding character? Using my own rules for variant weapon damage, a character wielding a heavy martial weapon with two hands does d12 damage. A character wielding a light martial weapon in each hand does d6 damage with each. If they hit with both, they do an average of 7 damage, versus the two-hander's average 6.5. Not a huge difference, but it's there.

On top of that, normally, the player does not add their character's Strength or Dexterity modifier to the damage roll from their off-hand weapon, but the two-weapon fighting style allows them to do so, which can make the disparity greater. The dual-wielder also has an extra chance to score a critical hit.

But aha, two-handed weapons only require two hands when making an attack (PHB, page 146)! The second hand is free for the rest of the character's turn, allowing them to manipulate objects, draw items, and the like. If they can cast spells, they might also use their now-free hand for that (more on that later). If the dual-wielder wants to free up a hand, they have to use their stow/drop/draw action on that turn to do so, then use their stow/drop/draw action again on their next turn to switch back. Not a huge difference, but it's there.

Shields

Shields might be an exception to the stow/drop/draw action rule. In 5e, shields require a full action to don or doff (PHB, page 146). This feels overly punitive to me. I would rule that a character could switch their shield out for another item as if that hand was holding anything else. Maybe this isn't realistic (assuming the shield is strapped to the arm), but the gamey, mechanical tradeoff decision from round to round is what I'm interested in here, not realism.

This requires another change - in 5e, a character can benefit from a shield even if they're not proficient with shields. Here's the tradeoff: "If you wear armor that you lack proficiency with, you have disadvantage on any ability check, saving throw, or attack roll that involves Strength or Dexterity, and you can't cast spells" (PHB, page 144). 

That above rule doesn't matter (aside from making saving throws) if the character has already taken their action that turn, meaning a character without shield proficiency could take their action as normal, then whip out a shield to benefit from the extra AC. I don't want that.

So if we remove the action required to don or doff a shield, we have to go a bit further and say "You do not benefit from the bonus to Armor Class when wielding a shield with which you are not proficient." Easy.

The Buckler Shield

One thing I miss from AD&D is different types of shields. 2e has buckler shields, small shields, medium shields, and body shields. I've simplified this a bit in 5e, introducing light (buckler) shields, medium (standard) shields, and heavy (body) shields. 

A character no longer needs proficiency in shields to use one - they just need proficiency in the corresponding armor type. This means that classes which don't normally get to use shields (most bards, rogues, and warlocks) can now use some certain shields, which doesn't bother me.

Standard shields work the same as the typical 5e shield. Body shields are similar, but against ranged attacks from the front and flanks, the character is treated as having three-quarters cover (+5 to AC, instead of the +2 typical of shields).

The buckler is the important shield when it comes to the discussion of hands, because it's light and fastens to the forearm, allowing the hand of the shield arm to be used to do things like fire a bow or crossbow, cast a spell, hold a torch, open a door, or the like. So the character wielding a buckler never needs to worry about stowing, dropping, or drawing their shield.

The tradeoff is that the buckler shield only grants the +2 bonus to Armor Class against one attack per round, and only if the character uses their reaction. It comes in handy (haha), but only if the character's reaction isn't already spoken for, and it's still not as good as having a proper shield.

The Components of a Spellcaster

Finally, I want to talk about spellcasting components. There are three kinds of spell components in D&D 5e: verbal, somatic, and material (PHB, page 203). Every spell in D&D requires at least one of these components, but more often, some combination of two or all three components in order to cast. 

Verbal components are "mystic words" which are spoken aloud as the spell is cast. It's not enough to simply say the words - the combination of sounds requires "specific pitch and resonance" to "set the threads of magic in motion". It's why, in my opinion, a character cannot "quietly" or "stealthily" cast a spell. Uttering the verbal components is loud and showy - it's evident to everyone present that a spell is being cast, disgraced 5e designers be damned.

Somatic components "might include a forceful gesticulation or an intricate set of gestures". Material components are "particular objects" but can be substituted by a component pouch or spellcasting focus so long as the components has no cost and isn't consumed by the spell.

Here's where we get to hands: To cast a spell with somatic components, "the caster must have free use of at least one hand". To cast a spell with material components, "a spellcaster must have a hand free to access [the] material components". However, the hand used to access the material components "can be the same hand that [the caster] uses to perform somatic components".

It used to be that a caster's whole body had to be free to cast a spell. A spellcaster casting a spell with somatic components couldn't just wiggle their fingers. Their whole body had to move. They had to do a dumb little dance. A spellcaster could not move on the same turn they cast a spell. If they were on a ship in the middle of a storm, the rest of the party would have to brace them to hold them still so that they could cast without being thrown around the ship. This, along with proper Vancian spellcasting and simultaneous actions in combat - allowing spellcasters to be interrupted while casting - was a huge element of reigning in the power of spellcasters. 

5e has a much different conception of spellcasters. They're basically hyper-mobile artillery. They can run, ride, and fly around casting spells, and all the while, they only need one hand free. They can be restrained by a titanic boa constrictor and can still cast spells with somatic and material components, so long as they can wiggle their fingers.

I personally rule that a spellcaster can't cast spells with somatic components if they're, for example, grappled or restrained, and in that case, they also can't cast spells with a material component unless those materials are already in hand.

I'm inclined to take it a step farther. What if spellcasters needed one free hand for somatic components, and another for material components? Casting a spell which requires both is a full-body action, meaning the character has to stow or drop whatever other items are occupying their hands before they cast.

Unfortunately, the characters who would suffer most from this aren't the dedicated casters, who probably aren't holding much other than a spellcasting focus anyway - the true victims are those that straddle the line between fighting and spellcasting: artificers, clerics, druids, paladins, and rangers. These characters are spellcasters as well as characters that want to wield weapons and, often, shields.

Most of these classes have features which would mitigate the impact of this effect, allowing them to substitute some other item for a spellcasting focus. Artificers can use infused items as spellcasting foci, meaning that an artificer could infuse their weapon or shield, using it for their material components. Clerics and paladins can etch their holy symbol into a shield or, even better, simply wear it visibly on their person. Druids and - as of Tasha's Cauldron of Everything - rangers can use a druidic focus, which can be a staff, which is also a weapon (although maybe not the preferred one). 

In baseline 5e, this allows each of these character types to wield a useful item while still casting spells which require both somatic and material components (oddly, if the spell only requires a somatic component, they still need a free hand). If we change the rule to require one free hand for each of these components, then they at least need one less hand free.

These classes can also benefit from taking the War Caster feat, which eliminates the need for a free hand to use somatic components, so long as the character is wielding weapons or a shield in one or both hands (my reading of it is that the character couldn't cast a spell requiring somatic components while holding a torch in one hand and holding a potion in the other, for example).

Certain subclasses suffer more greatly - those classes which normally don't cast spells, but have subclasses which do, such as the eldritch knight fighter or the arcane trickster rogue, or those classes which normally don't wield weapons, but have subclasses which do, such as the swords and valor bards and the hexblade warlock. Since 5e is designed in such a way that casting spells with somatic components, material components, or both requires just one hand, this isn't normally a problem. If we require them to use both hands, they become much more limited. If I were to implement the house rule I'm proposing, I would likely give these subclasses some extra feature that allows them to use a weapon or shield as a focus.

Overall, I've not totally sold myself on making this change. It's not like changing a single spell or the way a class feature works, which is relatively easy - it's changing the base rules about how spellcasting as a whole works, which potentially has tons of knock-on effects. 

I'm not particularly interested in going through the whole spell list and analyzing the edge cases of which spells become more or less effective with different classes and subclasses depending on what they can and can't use as a spellcasting focus. It's the kind of thing that would require playtesting. Maybe next time I run a 5e game, I'll try it. Or maybe by then I'll have decided I don't really care.

The important take away is that I'm trying to pay more attention to what my players' characters are doing with their hands. Doing so generates interesting round-by-round decision points (holding a light source, choosing weapons, using a shield, and casting spells are all meaningful tradeoffs). This, and a healthy dose of common sense, reigns in more powerful character concepts, such as spellcasters (a wizard can't cast elaborate spells while being throttled by a tentacled beast). That at the very least is worth considering.

Wednesday, May 8, 2024

AD&D 2e Play Report: Session 18

We played the 18th session of our AD&D 2e campaign last night. The play report for our last session is here.

To recap briefly, the party had just defeated a clan of ogres at great cost - Derrell, their druid henchman, had died. The rest of them were low on hit points. They took two nights to recover in the empty den, then headed back to Grasshold. From there, they traveled to Mythshire to deliver the share of treasure from another fallen companion, Ash, to the elf enclave he called home. There, they learned that a nearby elf tomb was overrun with undead. The tomb's custodian, Themar, had turned to dark powers. The party recruited a new henchman, Orlina, a cleric, to join them, then headed out.

On the journey, the party evaded a warband of 50 orcs, who destroyed the party's campsite. The party decided to return to town to resupply, then set out again. Upon entering the tomb, they were ambushed by turn resistant, spellcasting zombies that dragged themselves out of the stream flowing through the complex. Corym Vadercast was brought within an inch of his life as the party attempted to make a fighting retreat.

Party Roster

  • Bernhardt Dalton, NG human cleric 4
    • Magicheart, NG pseudodragon
    • Orlina, CN human cleric 1
    • Yinvalur Sparkguard, NG elf fighter 3
  • Corym Vadercast, NG elf fighter 3/mage 3
  • Haymond Baler, LG human fighter 4
    • Yuliana, NG human cleric 2
  • Karven Stone, LE dwarf thief 5
    • Sheyla, LE human thief 3

A Watery Grave

For this session, I gave Corym's player control of Yinvalur, since otherwise the player would be sitting around doing nothing until the combat ended.

The session did not get off to an auspicious start. Sheyla was bloodied, so Karven's player wanted her to withdraw immediately. Since she was badly wounded, I made a morale check for her. If she passed, she would keep her cool and withdraw carefully, without provoking a free attack from the zombie she was in melee with. If she failed, she would turn her back and flee as quickly as possible, which would provoke an attack.

She failed quite handily. She turned and fled, and the zombie clubbed her in the back of the head, killing her.

The party continued to fight. They missed a lot of attacks, and when they did hit, they almost always did minimal damage. Haymond's player rolled minimum damage three rounds in a row, in fact. The party took a few more licks, but after some nice damage rolls from Yinvalur (again, the chad elf henchman saves the day), they eventually prevailed.

The party collected what little valuables the zombies had on them (2d4 gold pieces each), and Karven took Sheyla's hoarded wealth (about 450 gold in platinum pieces, plus a healing potion - if only I had remembered she had that!). The party dumped the zombies' bodies back into the water, lest they be discovered by whatever else lurked in this tomb.

The party was not enthusiastic about continuing to explore the place, so they headed back to town. It would take them the rest of the day (they had five hours of travel time left), plus the better part of the next day.

Encounter Fatigue

I am still trying to decide whether to continue using the AD&D 1e or 2e wilderness encounter tables. The 1e tables use d100 and have more entries as a result. The 2e tables use 2d10 and have fewer encounters. I'm trying to get a sense for which set of tables is better for my purposes. For now, when an encounter is rolled, I roll d2 to determine which set of tables I use. 

The greater issue is that I'm beginning to reconsider using these tables during a session at all. I'd be lying if I said I wasn't become fatigued by having to improvise up to seven random encounters per session, depending on how far the party is traveling and in what terrain. It's no problem if the party is traveling through settled farmland, but the forest (and dense forest in particular) is proving to be more of a problem.

For example, this session ended up having four random encounters, and I would say we spent no more than a third of our game time on travel. One of the encounters happened in the middle of the night and involved 60 human pilgrims traveling in the woods. Perplexed, I ruled that instead, there was no encounter (this is somewhat of an entertaining challenge - either I improvise something on the spot to the best of my ability, or the players luck out, and nothing happens).

I think I maybe could have made something of this encounter if I'd had time to prepare it in advance. Loosely interpreting the result, perhaps the pilgrims are actually elves on their pilgrimage to the afterlife. Maybe they're skeletons on a "pilgrimage" to the tomb. But I don't know, I shy away from twisting the original intent too much.

In an ideal sandbox, the DM creates custom encounter tables for the environments in their region. The problem is that my region doesn't have clearly delineated territories. Sure, there's a big forest and a big swamp and a big desert, but there are also smaller forests and bodies of water, and multiple mountain ranges. Does each of those need an encounter table? In my 5e game, I made tables for each terrain type, which worked well, but in that case, I might as well use the tables I'm using already.

Another option is to decide how many encounters I want to detail ahead of time, then roll in advance on the existing tables that many times to determine what those encounters are, then detail just that handful of encounters. This is maybe the best solution, and allows me to save table time by predetermining encounter distance, the number of monsters, their disposition, treasure, names, motives, and the like. 

That might be what I end up doing, but it's a lot of work, and I'm currently in a bit of a "prep less" mindset, so I'm not sure. Something to think about.

Encountering Elves

The party was a few hours out from the tomb when they encountered a scouting party of wood elves. The elves had the best possible reaction - helpful. I decided that they were related to the elves in Mythshire, but that a dispute had caused the two groups to fracture. These elves continued to live in the forest and monitor the tomb, and Themar's activities.

The elves offered to lead the party to their sanctuary, where they could recuperate before returning to the tomb. The party was about equidistant from town and from the elves' settlement, but the elves' settlement was closer to the tomb, so they agreed to follow. 

Along the way, they had to rest. While camping, the party spotted a pair of minotaurs prowling in the forest. The minotaurs were indifferent and, not wanting to fight minotaurs, the party let them go on their way. 

That night, they heard a huge creature crashing through the trees overhead, clumsily flying or leaping from tree to tree, making both animal sounds and speaking in a man's voice. It sounded like it was moving around their camp, not towards it, so the party kept a watchful eye, but didn't seek the creature out. (It was a manticore - again, indifferent.)

The next day, the party reached the elves' settlement without incident. Corym and Yinvalur were honored guests at the feast that night, and the local elves thought it cute that they had brought along their human followers and their nasty pet dwarf.

Karven paid an elf to deliver Sheyla's body to Mythshire and pay for a proper burial. The party also asked around and hired seven (!) elf hirelings to help them when they returned to the tomb.

The elf settlement had a population of 70. I determined that there were no henchmen there to recruit (my rule is that 1 in 100 people in the setting have class levels), and that one in ten elves there would be the fighting type. Haymond's Charisma alone is sufficient to have up to 10 followers, so I ruled that the party could recruit all of them, if they were willing to pay (2 gp per day for 2 days, so 4 gp per elf).

Now that I've had the chance to actually review the exhaustive Monstrous Manual entry for elves, I see that there is supposed to be one 2nd- or 3rd-level fighter for every 20 elves, and one 2nd- or 3rd-level mage for every 40 elves, meaning this group of elves would have three 2nd- or 3rd-level fighters and one 2nd- or 3rd-level mage. Furthermore, in a camp, for every 40 elves there are an additional 4th-level fighter, 4th-level cleric, 2nd-level fighter/mage/thief, a 4th-level fighter/7th-level mage, a 5th-level fighter, a 6th-level fighter, and a 7th-level cleric. Like I said, exhaustive.

Only Karven is high enough level to recruit new henchmen of 2nd-level, and he is Evil (and a dwarf, to boot), so the elves wouldn't have followed him anyway, making my in the moment oversight a non-issue.

I'm also seeing now that wood elves avoid contact with strangers 75% of the time. They also only speak their own language and the languages of forest animals and treants. They keep the locations of their camps secret from outsiders, going so far as to kill people who stumble upon them.

So uh, yeah. This encounter probably shouldn't have happened this way. All the more reason for me to prep this stuff ahead of time, I guess.

Seven Elves

The party spent a night recuperating among the elves. The clerics in the party burned some spell slots topping off their wounded members' hit points in the morning. Then, the party took their seven elves back to the tomb. One player named them: Inky, Blinky, Pinky, Stinky, Finky, Winky, and Clyde. Another named them: Inhale, Sociable, Awake, Sad, Cheerful, Alert, and Civilian.

The party arrived at the tomb without incident. Bernhardt cast invisibility to undead on Karven, who proceeded to scout ahead.

Karven found a cavern where ten zombies stood guard in front of an alcove blocked off by curtains. Beyond the curtains, he found an elf priest with sallow grey skin and black veins, a cloud of flies buzzing about his head. To the south, across the stream, was a smaller cavern where eight more zombies were digging a pit with shovels. Inky black, malicious-looking magical tendrils were snaking out of the cracks in the walls of the pit. All of these zombies were noticeably different from the ones they fought earlier. They looked less powerful - more like zombie rabble than especially potent zombies infused with malign, god-denying dark energy.

The party devised a plan. While invisible, Karven would get into position to backstab the priest. Bernhardt would also make himself invisible to the undead, then get into position in front of the north tunnel and prepare to reveal himself and turn the zombie rabble. Just behind him, Corym, Yinvalur, and Orlina would wait out of earshot, ready to charge in. Meanwhile, Haymond and Yuliana would take the seven elves around the south, across the stream and in front of the dig site, ready to either open fire on the zombies to the north or cut off the zombies to the south.

This took about 30 minutes for the players to devise, and I actually really enjoy these moments. It can feel like nothing is being accomplished as the players go back and forth with ideas and hemming and hawing, but they're actually playing the game by making strategic decisions based on the information they have. The fact that they have to take this time to figure out how to approach the situation makes me feel like I managed to devise an actually interesting, complex scenario. I'm more than happy to let the players talk it out, answer their questions, clarify their intentions, and help them implement their plan as they describe it.

Once everyone got into their positions, the party launched its attack. I gave them a surprise round as a reward for good planning. Karven successfully backstabbed the priest...for minimal damage, again. Bernhardt revealed himself, turning eight of the ten zombies. Yinvalur and Orlina positioned themselves in front of Bernhardt, with Corym bringing up the rear. Haymond and Yuliana moved in with their elves, launching a volley of ranged attacks at the unturned zombies - as well as the turned ones, who were fleeing in their direction.

All told, the party killed one of the two unturned zombies and two of the turned ones. Unfortunately, Karven's backstab did little to deter the priest commanding the undead, who turned angrily to face him.

That's where we wrapped up the session. Next time, the party must deal with the heretical priest, Themar. Haymond, Yuliana, and the elves will have half a dozen frightened zombies bearing down on them, and perhaps half a dozen more emerging from the pit to attack their flank. Themar has a few nasty tricks up his sleeve, and I wouldn't be at all surprised if we had a few casualties.

Friday, May 3, 2024

Can't Someone Else Do This?

I'm currently catching up on Blogs on Tape's backlog of excellent, well...blogs on tape. I haven't always been super tuned-in to the blogosphere, so there's plenty of quality stuff I've missed. I'm grateful for Blogs on Tape for giving me the opportunity to catch up on what I've missed (and reexperience old favorites) while I'm driving, cooking, playing Mariokart, etc.

Yesterday, I had the opportunity to listen to the reading of Nick LS Whelan's Questgivers are Evil from his Paper and Pencils blog. In this post, Nick writes:

Your PCs are not an army, they’re a handful of dudes with weapons who are willing to kill stuff for money. Who hires people like that?

People who want you to do some bad shit. People who want to deny knowledge of something. People who want to blame “those outsiders.”

In other words, questgivers are evil.

There is pretty much NEVER going to be a long chain of dudes who want to hold your hand while you become an icon of heroic heroism and noble nobility, all while making sure your endeavors remain profitable.

Accepting a quest should mean that you’re about to do some shady ass shit that will probably hurt some good people.

I really like this conclusion. It doesn't make a lot of sense that people (whether they be commoners, nobility, or the merchant class) would turn to crazy, heavily armed outsider mercenary warriors and mysterious sorcerers to solve their local problems. Commoners would expect their rulers to protect them. The nobility presumably have some sworn vassals or other means of asserting their authority, which they would also use to keep the goblins away from the homesteads at night. The merchant class's resources are more limited, but they would likely appeal to the authorities with their problems as well, rather than turn to unknown elements.

I don't necessarily think that questgivers should be Evil. I like to occasionally sprinkle an Evil questgiver into my campaigns. A wizard who offers boatloads of gold to any adventurers willing to butcher a nest of pegasi and steal their eggs for his weird experiments offers a hook for Evil characters, a dilemma for Neutral ones, and an inverse hook for Good ones - "We have to take out that wizard!" The occasional questgiver who willfully gets the player characters into trouble to accomplish their own insidious ends adds some spice to an otherwise textbook campaign - but only if used sparingly.

If all questgivers are Evil (or at least, all the ones offering money are), then players might simply stop doing quests (at least, not for the gold). This might be a good thing, as the players will learn to start doing things out of the Goodness of their hearts, rather than due to the lightness of their purses, and their characters will become true heroes in so doing. Alternatively, they can dance with the devil, getting in trouble from one settlement to the next, suspecting every patron of ulterior motives, getting the gold, burning bridges, and escaping just before the other shoe drops.

That does actually sound kind of fun. D&D should be more like a picaresque - the player characters are Cugel the Clever or Larry David-like protagonists (not heroes), who do things for their own benefit, alienating others and either getting their comeuppance or causing some hapless associate to take the blame instead. But that's not what I'm arguing for here.

What I'm coming to find that I prefer in my own games is instead this: the authorities - or those in power in the setting - are either Evil, incompetent, indifferent, or powerless. People in these tight-knit communities turn to strangers for help because their rulers or would-be protectors either actively prey upon them, approach the communities' problems ineffectively, don't care about them, or lack the means to address them directly.

Perhaps the rulers are Evil, motivated solely by greed or domination. Maybe the government, as it exists, is in a state of decline or negligent decadence. The bureaucracy functions poorly or not at all. The rulers are not interested in doing good work or getting things done. Their hands are tied by distant wars. Their ideology is one of self-reliance and hands-off rulership. The government is physically far away and unable to project power - which works especially well in a typical "frontier" campaign setting.

This also solves the issue that I refer to as "exceptional individual demographics". Surely, the 1st-level player characters are not the only exceptional individuals in the setting, or else where do replacement player characters and henchmen come from? Who trained the fighter, tutored the wizard, or initiated the cleric? Who rules the strongholds, churches, and thieves' guilds? My general rule is that 1% of people in the setting are such individuals with class levels. Most of them are low level, but 5% are very high level. Why don't these people solve the setting's problems?

Because they're Evil, or indifferent (it would probably be wrong to characterize 20th-level NPCs as incompetent or powerless). They may have bigger fish to fry or they may simply not care.

The crux of the issue is that the local power structure - and powerful individuals in the setting - are flawed in some critical way, unable or unwilling to affect change or help regular people. This allows for a variety of scenario set ups. The government and powerful NPCs may be antagonists, or allies with limited accessibility/utility. They may be accustomed to employing mercenaries and regularly post rewards for work that needs doing. The merchant class may be the ones driving progress, or communities may pool resources to attract entrepreneurial adventurer types. 

Of course, on occasion the questgiver is or should be Evil, but it is this general, dominating zeitgeist of anarchy, indifference, or self-interest which allows those Evildoers to flourish, and to recruit stooges (the player characters and/or their rivals and enemies) to do their dirty work and to take the fall.

When the Players Don't Care

Sometimes, the players just don't care to do something. They may have other priorities, the quest may not sound interesting to them, the reward may not be compelling enough, the patron may be too suspicious, or the players may simply be too afraid to fight the dragon. This is totally fine. The game is about players making choices, and while those choices should always have consequences, it's not exactly fun gaming for the fictional player characters and the setting to be punished over and over again because the real-world players simply weren't interested in the hook. 

Some hooks land with some groups, players, and characters, and some don't. It's why I throw out as many hooks as I can reasonably prep scenarios for. Eventually, something will interest them, and I can prep the next volley of hooks while they're busy pursuing the one they're currently invested in.

But what happens to the other scenarios? I just said that the powers-that-be in the setting are unwilling or unable to do anything about these problems, so don't they just get worse?

Maybe. The players have chosen to leave it to fate. I like to use a simple 2d6 roll to determine what happens when a scenario is neglected:

  • On a 2 or 3, the worst possible thing has happened. Another group of adventurers or the local ruler has finally gotten around to intervening, with disastrous results. The source of the problem has triumphed and strengthened, and is retaliating in some way. If the local power structure is Evil, perhaps they have resolved the scenario in a way which strengthens them instead. The Evil sorcerer-king has aligned with the orc warband and is now using them to enact their next nefarious scheme.
  • On a 4 or 5, a setback has occurred. Someone tried to do something about the problem and failed, and now the problem is worse in some minor way. Adventurers tried to defeat the necromancer, but died, and now the necromancer's dungeon is more dangerous (guarded by the reanimated corpses of said adventurers). The goblins are emboldened by their victory, and now are raiding homesteads closer to the local settlement.
  • On a 6, 7, or 8, the status quo is maintained. No meaningful action has been taken towards resolving the scenario.
  • On 9 or 10, something good has happened. The local rulers have finally gotten around to sending soldiers to guard the town against the nearby gnoll stronghold, or, another group of adventurers made a foray against the gnolls and managed to clear out a few rooms before retreating. The scenario is now easier, should the players choose to pursue it.
  • On an 11 or 12, the problem has been resolved in the best possible way for the setting (though this might not benefit the characters). A rival band of adventurers have solved the problem, and are hailed as heroes. The player characters have to listen to everyone constantly talk about how brave and strong these other heroes are.

Wednesday, May 1, 2024

AD&D 2e Play Report: Session 17

We played the 17th session of our AD&D 2e campaign last night (we skipped last week). The play report for our last session is here.

To recap briefly, the party battled a den of five ogres. Karven Stone attempted to kill the ogres' leader with a backstab, his dagger coated with lethal spider venom, but he missed his initial attempt, and the ogre matriarch eventually saved against the poison anyway. A dangerous melee ensued, and Derrell, the party's druid henchman, was killed. Sheyla, the henchman thief, ran off in a panic. Eventually, the party was victorious, slaying all the ogres, but not without their own grievous wounds.

Party Roster

  • Bernhardt Dalton, NG human cleric 4
    • Magicheart, NG pseudodragon
    • Yinvalur Sparkguard, NG elf fighter 3
  • Corym Vadercast, NG elf fighter 2/mage 2
  • Haymond Baler, LG human fighter 4
    • Yuliana, NG human cleric 2
  • Karven Stone, LE dwarf thief 5
    • Sheyla, LE human thief 3

Recuperation

One thing I noticed while doing my usual prep was that I had misinterpreted how natural healing is supposed to work in 2e. From the PHB:

Characters heal naturally at a rate of 1 hit point per day of rest. Rest is defined as low activity--nothing more strenuous than riding a horse or traveling from one place to another. Fighting, running in fear, lifting a heavy boulder, or any other physical activity prevents resting, since it strains old wounds and may even reopen them.

Characters recover hit points at a rate of 1 per day only if they don't fight that day. The healing proficiency (which Bernhardt Dalton has) says this (also from the PHB):

If a wounded character remains under the care of someone with healing proficiency, that character can recover lost hit points at the rate of 1 per day even when traveling or engaging in nonstrenuous activity. If the wounded character gets complete rest, he can recover 2 hit points per day while under such care. Only characters with both healing and herbalism proficiencies can help others recover at the rate of 3 hit points per day of rest. This care does not require a proficiency check, only the regular attention of the proficient character. Up to six patients can be cared for at any time.

And even now, reading it back again, I'm realizing that I've made another mistake. Bernhardt has the herbalism proficiency in addition to the healing proficiency, so my original interpretation was that characters he tended to recovered 3 hit points any time they completed what we would call a "long rest". I failed to realize that this only happens when the characters have complete rest.

My new interpretation, as I explained it to the players, was that with complete rest for a day (i.e., no traveling or fighting), Bernhardt could tend to their characters, and they would recover 3 hit points upon completing the day of rest (which I think is correct). If the characters are traveling and fighting, they instead gain 1 hit point per day under Bernhardt's care. The players mourned the loss of their 2 additional hit points per day, and we moved on. Now I'm realizing this is also wrong. I misread "nonstrenuous" in the first sentence of the description of the healing proficiency as "strenuous", somehow.

But it's strange. The default rules for natural healing say that characters regain 1 hit point per day so long as they don't strain themselves (i.e. by fighting). The rules for the healing proficiency then go on to say characters regain 1 hit point under these circumstances only if they're being tended to by a healer. I'm guessing that because the proficiency rules are optional, this redundancy is an oversight, or simply an unfortunate lack of clarification. I suppose the answer is that one uses the regular rules for natural healing if the optional proficiency system isn't used, and that one instead uses the rules for natural healing as described by the healing proficiency if the proficiency system is in use.

Even with the healing proficiency, no natural healing of any kind occurs on a day in which the characters strain themselves (i.e. fight). I guess I've got more bad news for the players next session.

In any case, with my temporary misinterpretation of the healing rules in mind, the players decided to rest in the ogre den for the remainder of the day, plus one additional day, to recover a bit before making the trip back to town. Bernhardt and Yuliana used their spell slots to spam cure wounds during that time. 

Karven poked his head outside to find Sheyla (she was hiding behind a rock - I made another morale check for her to see if she'd flee all the way back to town, and she succeeded on the check). He also set a rudimentary trap at the cave entrance in case anything else came sniffing around while the party was resting, but I made it clear to the players that they were unlikely to have random encounters while resting in a known, very recently-inhabited ogre den.

Packs, Laden with Loot

The players determined that between their two mules and cart, they could just barely carry the 37,000 copper coins they found in the ogres' treasure hoard, but they would be traveling at one-third their usual speed. The journey back to Grasshold (normally half a day) would instead take a day and a half. Satisfied with their hit point totals, they set out.

The party encountered a patrol from town on the second day, who escorted them the rest of the way. Back in town, they collected their reward, changed their money into more portable denominations, and spent the night in an inn. Karven bought another mule to ease the burden the next time they find thousands of pennies in a cave.

The players decided to travel next to Southreach, hoping to find an armorer who could make something useful of the ankheg shells they're carrying. Along the way, they'd stop in Mythshire. Ash, their fallen companion, was a member of an elf enclave there. The party wanted to bring his share of gold back to the enclave and inform them of his passing. 

They traveled for a day to Spiritbrook, then another three and a half days to Mythshire. The journey was uneventful.

The Death of Elves

In Mythshire, the party sought out Zylra Oakwhisper, the matriarch of the town's elf enclave. There was much melancholy about Ash's death. The demihumans of this world are, in generic fantasy fashion, dying out. The dwarves have been undone by their greed and hubris. The sylvan lands of the elves have become deforested, and their slow reproduction rate can't keep up with their losses. The halfling's shires are being despoiled by industrialization. And so on.

Zylra thanked the party for notifying her of Ash's demise. Since they were here, she asked if they might be willing to help with another matter. An elf burial ground, Ryl Themar, reserved for the honored dead, lay in the nearby forest. The priest there, Themar, who was charged with watching over the place, has been consorting with dark powers, and has raised the elves' magically-preserved dead as unliving abominations. Zylra offered a hefty reward, as well as the enclave's favor, in exchange for purging Themar's insidious influence.

The tomb was just over a day's travel from town, and the reward was significant, so the party agreed to look into it. I also know that Karven's player is fiending for some elven chain mail, so the promise of goodwill with the local elves would be too good for Karven to pass up. With a new understanding of the rules for healing, and knowing that they'd be up against undead, the party sought out and recruited Orlina, a Lawful Neutral 1st-level cleric (she is now a henchman of Bernhardt). With half a day left of traveling time, the party set out immediately.

Encounter Tables

Sometime before I started this campaign, I purchased a PDF of the 2e Monstrous Compendium. The Monstrous Manual (the reprinted version of the Compendium), for whatever reason, doesn't have encounter tables (just advice for the DM to make their own), but the Compendium does (which is why I bought it). For some reason, I had looked at the tables in the Compendium and decided to use those in the AD&D 1e DMG instead. I think it was probably because the 1e tables are more robust (they are d100 tables, whereas the Compendium ones are 2d10, which is the standard in 2e).

In any case, for this session I decided to use the Compendium tables. 

The party traveled for half a day and made camp in the forest. This was still close enough to town to be considered "inhabited lands", so the players felt safe to camp there. Otherwise, they would have rested at the edge of the forest and entered the woods the following day.

I rolled a nighttime encounter and got orcs. I rolled d100 and got a 90, indicating that this was to be a high-level encounter. Uh oh. Corym was on watch. With his infravision, he was able to see a group of about 50 orcs (!) moving through the forest towards the party's camp. The orcs would be on them in less than a minute. Corym roused the party, who quickly gathered as many of their things as they could - including the mules - and made a run for it.

I used OSE's evasion rules (2e has no such thing) and determined that the party was able to successfully scatter and hide. The orcs sent about a dozen scouts out looking for them while the rest tore apart their campsite. Eventually, they wandered off, and the party returned to salvage what they could and finish the night.

In hindsight, I wish I'd handled this differently. It doesn't sit quite right with me that the party could gather everything and escape in such a short time. At the very least, they should probably have lost their mules. And the orcs probably should have left some scouts behind to see if the party returned. As is, the party only lost some tents and such, which is a minor inconvenience. In the future, I'll probably impose some penalty on the evasion roll if the party takes the time to gather their things before fleeing, especially since most of them were asleep when this occurred.

My reaction roll determined that the orcs were unfriendly, so they wouldn't have attacked on sight anyway. My plan was for them to simply bully and rob the party. It's somewhat of a boon that the party didn't interact with them, though, as this gives me an opportunity to figure out what a sizeable orc warband is doing in the forest, so close to town. Mythshire's ruler is an evil wizard, so maybe they're in league somehow, but who knows what they're up to.

In the morning, the party returned to Mythshire to resupply and warn the militia about the orcs. Then, they went right back out. On their second night camping in the woods, they encountered a patrol from Mythshire. I determined that this was probably some sort of fantasy special ops team - rangers or the like well-versed in dangerous nighttime forest excursions. They were seeing about those orcs. They warned the party to be cautious and went on their way.

Dead Water

The party arrived at Ryl Themar without further incident. It was a rudimentary burial ground built into a natural cave, with some hastily constructed, crude monoliths atop it. To the elves in the party, it was only a shadow of the grand mausoleums from the height of their peoples' past. A tributary from the nearby river flowed through the cave, with a narrow stone ledge alongside the stream allowing access to the cavern's depths. An intolerable stench emanated from within.

For this small adventure site, I used Dyson Logos's Tombs of the Throl Tribe.

Karven snuck inside to scout ahead. He found a large open cavern alongside the stream. There was a pile of bones on one side of the cave, next to an ascending tunnel where the stench intensified. At the far side of the cavern was a plank bridge across the river to another cavern with a semicircular arrangement of carved wooden doors. The rest of the party followed closely behind Karven, who decided to sneak up the ascending tunnel while the others watched the river.

Before long, bloated, stinking corpses began to haul themselves up out of the river. Despite their suspicions, Haymond and Karven were surprised. Upon laying eyes on the water-logged zombies, the party became nauseated (everyone besides Orlina failed their saving throw vs poison). The sickening effect lasts for 2d4 rounds and imposes a -1 penalty to attack rolls and armor class as the character retches and such.

Bernhardt, Orlina, and Yuliana all attempted to turn the zombies. Unfortunately, these are sea zombies (er...river zombies), which are immune to turn undead. This is because of some baseline D&D lore stuff that I should have probably disregarded, but I wanted to run the monsters as is to stay true to their intended threat level.

The zombies closed in to attack - one went after Orlina, another after Yuliana (these two were now blocking the party's escape), and three (!) after Corym, who happened to be closest to all three of them. Yuliana was quickly bloodied. Corym was overwhelmed and brought to 0 hit points - he began dying.

The party concluded that a fighting retreat was in order, but they'd have to save Corym first. Karven and Yinvalur leapt to his aid. Bernhardt moved to administer first aid. Orlina and Yuliana retreated towards the entrance. Haymond and Sheyla tried to slay the zombies at their rear to clear a path for the others. Two of the zombies were spellcasters, and began slashing at Bernhardt, Karven, and Yinvalur with claws wreathed in life-draining energy.

Bernhardt stabilized Corym at -9 hit points (he would have been dead at -10). Sheyla bloodied the zombie at the party's rear, but was bloodied herself. Yinvalur managed to kill one of the spellcasters.

Orlina and Yuliana are just off screen, guarding the rear.

We were at the top of the next round, and already 30 minutes over our usual time, so we called the session there. Not a super satisfying place to end it, but I think everyone is looking forward to seeing if they can make it out alive. I let Corym's player know that they may want to prepare a new character for next week, just in case.

Next time, the party will either prevail, perish, or escape by the skin of their teeth from the zombies. They may press on deeper into the tomb or simply cut their losses and abandon their quest.