DISCUSSING
In the Shadow of the Spire – Session 28B: On the Eve of the Banewarrens

The mansion on Nibeck Street that Jevicca had identified as the origin point for the appearance of the surge of Tavan Zith’s wild magic was very close to Pythoness House. So close, in fact, that they feared there might be a connection. Could the cultists be responsible for the breaching of the Banewarrens?

“If we check it out and there’s nothing there,” Ranthir pointed out, “then we’ve lost nothing. But if there is…”

During the last session, we talked about how I structured the second act of the campaign using two tracks — the chaos cultists and the Banewarrens. Beginning in this session, we can almost immediately see the effects of this structure in actual play.

First, the two tracks confuse the players’ understanding of the situation. Until they learn enough to disambiguate the tracks, this will obfuscate the truth of what’s happening. This makes the campaigns’ enigma(s) fiendishly Byzantine for the players, and therefore even more satisfying for them when they do unravel what’s going on (in large part by figuring out how to disambiguate the tracks).

But until they do, their own actions will often cause interactions — directly and indirectly — between the tracks. The initial effects that we see in this session are fairly minor: Their suspicion that the chaos cults might be involved with the Banewarrens causes them to double back to Pythoness House, where they have a cool roleplaying encounter with the ghost Taunell.

Paradoxically, however, the complexity of these player-forged connections between the tracks will often grow in complexity at the same time that the players are disambiguating the tracks and, therefore, simplifying their understanding of a situation becoming ever more convoluted.

And even when this doesn’t happen, the consequences of the players’ choices will nevertheless be significant. (For example, their verification that Pythoness House is, in fact, vacant in this session — something they would otherwise not have been prompted to do — will actually end up having a profound impact on how later events in the campaign play out.)

They needed to question Tavan Zith, and the only way they could think to do that was by going to Castle Shard. They also needed to know if Lord Zavere was the one responsible for opening the Banewarrens. And, if so, why.

As they rode, Dominic looked at the others. “So… do we have any idea how we’re going to do this without getting killed?”

Agnarr shrugged. “Sure. We ask him. If he didn’t do it, we don’t get killed.”

Of course, these two major tracks are not the only threads in the campaign. This is, after all, Act II. The stuff that the PCs did in Act I of the campaign continues to unspool, and that includes:

• Their relationship with Lord Zavere and Lady Rill at Castle Shard.
• Their deep suspicion of Rehobath and, by extension, the Imperial Church.

And these threads are also interacting with the major tracks and with each other.

For the players, this colors their understanding of Rehobath’s agenda and creates paranoid suspicion of what Zavere might really be up to. There are layers upon layers upon layers! (And the players are unwittingly in the act of adding even more layers themselves.)

But on my side of the DM screen, everything remains neatly sorted into discrete boxes that are easy to prep and easy to run.

Campaign Journal: Session 28C – Running the Campaign: On the Efficacy of Burning Oil
In the Shadow of the Spire: Index

DISCUSSING
In the Shadow of the Spire – Session 28A: The Maw Beckons

They left. Once they were safely in the carriage and driving away from the Cathedral they talked things over.

“I don’t trust him,” Tor said.

Dominic nodded. “You can put crimson robes on a pig, it’s still not a novarch.”

But they would practically be getting paid twice for the same job. There was no reason to pass that up.

Scenario hooks are the methods by which PCs become aware that an adventure exists, are enticed to engage the adventure, and/or are forced to engage the adventure.

If you’re prepping a plot, then you’ll usually only have a single scenario hook which will also tell the PCs what they’re supposed to do (in order to set the predetermined plot in motion). If, on the other hand, you don’t prep plots and, instead, prep situations, you’ll find that you have A LOT more flexibility in the scenario hooks you set up.

One particularly powerful technique is, in fact, to have multiple scenario hooks pointing at the same scenario. You may do this for purely practical reasons (fulfilling the Three Clue Rule, for example), but it can also be deployed to great effect.

One of my favorite techniques, actually, is to have two different patrons offer to hire the PCs for the same job; or, more accurately, for jobs involving the same scenario. This setup creates the context for framing tough dilemmas. (“Do we chase after the assassin to claim the bounty or do we save the Jewel of Erthasard from the river of lava?”) In fact, you can do this from the moment the job offer comes in: If Patron A asks them to murder the CEO of Abletek and Patron B asks them to work as the CEO’s security detail during an upcoming business conference, you’re immediately forcing the players to really think about the scenario they’re being hooked into: What do they want to have happen to the CEO? They can’t just sit back and passively do whatever they’re told to do. They’re going to have make a decision.

And, once they’re thinking about the situation and making choices for themselves, they may end up deciding they want something completely different from either patron.

Another technique I enjoy using as surprising scenario hooks: It’s easy to have a hook tell the PCs exactly what’s happening. “There are goblins in the Old Tower and they’ve been raiding the local farms.” But it can often be more effective to not do that: Maybe the villagers think there are goblins at the Old Tower, but it’s actually an infestation of imps. Or the goblins in the tower are actually just orphans, and they’re not the ones responsible for the recent raids.

A surprising scenario hook, as the name suggests, sets things up for the players to be surprised later in the scenario. And there are, of course, all kinds of ways for you to use this surprise, whether for dramatic or strategic effect.

In this session, I’m combining both of these techniques while hooking the PCs into the Banewarrens. Not only are they being simultaneously approached by two patrons with different objectives related to the Banewarrens, but the true nature of those objectives are not immediately apparent to the players.

In this case, this also means that the PCs can initially believe that there’s no conflict between the two commissions. The surprising reversal will come when they discover the truth and realize their twin masters cannot, in fact, be satisfied simultaneously.

We have another name for that: Conflict.

Delicious, delightful conflict.

The other subtlety here is Tavan Zith. In the original Banewarrens book, Zith does not actually function as a scenario hook. (There’s no way for the PCs to backtrack from Zith to the Banewarrens.) The encounter with Zith, however, functions as a justification: The PCs interacting with Zith is used to justify the Inverted Pyramid (and, in my version, the Church) deciding to hire the PCs for this job.

I had also, knowing these hooks for the Banewarrens were coming, made a point of laying groundwork with both Jevicca Nor and the Imperial Church earlier in the campaign. I wasn’t sure exactly how this earlier involvement with these factions would play out, but really any involvement would either (a) help justify the PCs getting approached for this gig and/or (b) create tension that could be similarly paid off in the Banewarrens. In practice, this turned out even better than I could have ever anticipated:

“We live in a time of prophecy,” Rehobath said. “And you seem to have a habit of finding yourselves in the middle of it.”

“What do you mean?”

“The extraordinary events in Oldtown today — in which I have been told you were involved — are the beginning of what will be a new chapter in history. Tavan Zith has returned to this world, and if the prophecies are true that means that the Banewarrens have been opened.”

I actually did take the further step of making Tavan Zith an actual scenario hook: The PCs could have either backtracked his path by canvassing Oldtown (they didn’t do this) or interrogated him (they tried this, but failed their rolls). This is what I refer to as a curiosity hook (i.e., no one tells the PCs to go check out where Zith came from; but Zith’s presence and extraordinary actions make them aware of the scenario, and they can pursue it and/or get involved with it if their curiosity so inclines them).

You may be thinking: “A die roll for the scenario hook? But what if they failed the check?!”

Well… they did fail. But I had three more hooks lined up, so that’s okay.

Of far more concern would be if the players ended up simply not being interested in the Banewarrens at all. What should we do about that?

First, I’ve spent some time priming the pump here by layering in a bunch of foreshadowing about the Banewarrens. For example, the “Drill of the Banewarrens” in Session 16A. By the time we go to Act II, the players were already intrigued by the Banewarrens, which made them eager to jump at clear hooks pointing them in that direction.

Second, another advantage of using multiple hooks is that it gives the players multiple reasons to be interested in the scenario: Do you want to make allies with the Inverted Pyramid? Pursue your relationship with the Imperial Church? Get close to one or the other so that you can screw them over later? Pursue the powerful magical treasures within? Figure out how to put an end to the threat posed by Tavan Zith?

If I just used one hook, the reaction to that hook might be negative: “We’d like to help, but we don’t have time,” or, “We don’t trust the Church, so we’re not going to tangled up with them.” But with multiple hooks in play, it actually becomes exponentially more likely that the players will see a reason why they want to get involved. (And, again, not just the ones you package up for them. They’re very likely to come up with their own reasons.)

Third, even if turns out that the players aren’t interested in this scenario, the fact that I’ve already set things up so that there are multiple factions involved with interests that directly compete with each other will make it super easy for me to figure out what would happen next even in the absence of the PCs. In this case, the Banewarrens would drop into the campaign’s background events. From that position, they would continue to affect the campaign world, and likely things that the players ARE interested in. It’s extremely plausible that this would generate additional hooks in the future, which may or may not pull the PCs in after all. (Albeit into a scenario that may have already been radically transformed by their earlier decision not to get involved.)

NEXT:
Campaign Journal: Session 28B – Running the Campaign: Multi-Threaded Campaigns
In the Shadow of the Spire: Index

DISCUSSING
In the Shadow of the Spire – Session 27D: The Maw Opens

Tee read: “The Saint of Chaos shall return and the Banewarrens shall ope their maw. And the name of doom shall be Tavan Zith.”

“What does that mean?” Elestra asked.

“I don’t know,” Tee said. “Let’s ask him.”

She pulled Zith out of her bag of holding. Tor bound him securely. Tee blindfolded him. And Dominic healed him.

As soon as Tavan Zith awoke, however, they all felt a sickening, bursting feeling erupting in their chests. Agnarr instinctively smashed the pommel of his sword into the dark elf’s nose, breaking it and sending him plunging back into unconsciousness.

In a previous Running the Campaign column, at the beginning of Act II, I discussed the fact that I designed the second act to be triggered using two external events — events that originated from outside the domain of the PCs experience and, therefore, could not be anticipated or prevented. (Or, at least, were extremely unlikely to be anticipated or prevented.)

The first of these events was the letter from Shim that arrived in Session 18, informing the PCs that (a) they had hired him during their period of lost memories to find a magical artifact and (b) he’d found it. The second, of course, happens in this session, when Tavan Zith, the Saint of Chaos, appears on the street.

The first trigger is designed to hook the PCs into the Night of Dissolution campaign, which revolves around the cults of chaos and was designed for 4^th to 9^th level characters by Monte Cook.

Similarly, the second trigger leads to the Banewarrens, another campaign created by Monte Cook, this time designed for 6^th to 10^th level characters.

As I’ve described previously, it was my desire to run the Banewarrens that was the primary impetus for the entire campaign. But when I read Night of Dissolution, I was fascinated by it. Which campaign should I run? Could we wrap Banewarrens and then run another Ptolus campaign featuring the Night of Dissolution?

Then I realized that I could just run both of them at the same time!

And although I significantly expanded both of them, these two campaigns remain the primary spine(s) of Act II.

Taking published adventures like this, combining them, and adapting them to the PCs is something I discuss in more detail in The Campaign Stitch. Often when I’m doing work like this, I will be looking for opportunities to create crossovers between the adventures — to tie them together and make them a single, unified whole.

For example, there are a number of factions in Ptolus interested in the Banewarrens and how they can be exploited. It would be perfectly natural for the chaos cults — another powerful faction active in Ptolus — to also become involved in the intrigues around the Banewarrens.

But I actually made a specific decision to NOT do that.

Instead, I used a different technique: The Second Track.

I knew that both the Banewarrens and the Night of Dissolution would be big, complicated conspiracies that the PCs would have to work to unravel. If I fully crossed the streams and truly merged the conspiracies, there was a real risk of the whole thing collapsing under its own Byzantine complexity. It would be hopelessly confusing.

But I knew I didn’t actually have to do that in order to get the same effect! When the PCs first started interacting with the two conspiracies, the players wouldn’t have the information necessary to distinguish them. So, from their perspective, the conspiracies WOULD be merged together, and they’d be utterly overwhelmed.

This meant that:

• I, as the DM, didn’t need to deal with the complexity. (Because I could clearly distinguish between the two conspiracies and wouldn’t’ get confused.)
• Once the players figured out how to distinguish the conspiracies, they would ALSO no longer be confused. The complexity would fall away and the disparate mysteries would cleanly resolve themselves.

That’s the beauty of the Second Track.

With that being said, however, I didn’t want these two halves of the campaign to be completely siloed from each other. That would feel pretty artificial. So I looked for some subtle crossovers (which I knew would also seed the players’ initial confusion between the two threads).

I came up with two.

First, as we saw in Session 20, the PCs discovered the Prophecy of the Saint of Chaos in Pythoness House (a site associated with the chaos cults):

The Saint of Chaos shall return and the Banewarrens shall ope their maw. And the name of doom shall be Tavan Zith.

Tavan Zith wasn’t a big focus point for the chaos cults, but given his unique curse, the idea of him being an avatar of chaos made perfect sense. (And if the PCs did end up tipping Tavan Zith’s arrival to Wuntad or the other chaos cultists, they could easily interpret it as a sign that their time had come and the Night of Dissolution was foreordained.)

Second, I identified the Pactlords of the Quaan as a faction who could potentially intersect with both the chaos cults and the Banewarrens. (We haven’t met them yet, but they’ll be showing up shortly.) The short version is that they were big enough that I could have one wing of the Pactlords tangentially involved with the chaos cults and a completely different set of Pactlords focused on the Banewarrens. Just enough crossover that the PCs would find references to the Pactlords in both places and assume a connection, but distinct enough that they wouldn’t cause the two threads to collide with each other.

This division between Chaos Cults and Banewarrens, I should note, is quite explicit in my own notes: There’s a binder of chaos cult-related adventures and a completely separate binder of Banewarrens-related adventures. A really clear example of how you can have absolute clarity in your own perception of the campaign, while nonetheless miring the players in delightful enigma.

NEXT:
Campaign Journal: Session 28A – Running the Campaign: One Job, Multiple Patrons
In the Shadow of the Spire: Index

DISCUSSING
In the Shadow of the Spire – Session 27C: The Saint of Chaos

Now that the half-orc was more of a curiosity than a threat, the crowd that had been scattering in a rapid retreat instead began to draw closer. But just as it seemed as if they had successfully calmed the situation, another man suddenly grabbed at his eyes. Bolts of blue lightning shot out of them, striking several people in the crowd. The thick stench of ozone filled the air. At least a half dozen people collapsed.

Panic erupted once again. In the midst of it, Tee was suddenly struck by the sight of a dark-cloaked man striding boldly down the street and seemingly oblivious to the chaos around him…

Crowds.

In an urban campaign, you find them cropping up all the time:

• on the street
• in a busy market
• at the local tavern
• storming the necromancer’s castle with pitchforks

Handling dozens or hundreds (or thousands) of NPCs individually would, obviously, be a hilariously bad idea. So you generally want to figure out some way of handling the entire crowd as a single entity.

Often, of course, a crowd is just part of the set dressing: You’re in a shopping mall and it’s filled with people. That’s pretty straightforward. At most you’ll want to think about what effect the crowd might have on the actions of the PCs and significant NPCs in the scene. (For example, the PCs might need to make a DC 10 Dexterity (Acrobatics) check to race through a thick crowd; on a failure, the crowd is treated as difficult terrain for them.)

But sometimes the Green Goblin comes swooping in on his glider and starts throwing pumpkin bombs around. Now combat has broken out and the crowd is panicked.

What often seems to happen is that the crowd is described as running and shouting (while having little or no effect on how things play out), and then they completely clear out as quickly as possible to simplify things even more.

But where’s the fun in that?

When I was preparing the riot scene in Session 4 of the campaign, I prepped a full set of rules for handling crowds and mobs in D&D 3^rd Edition. After some refinements from playtesting them, I posted them here on the Alexandrian way back in 2007.

Those rules are useful (with advanced options that help when the crowd is the focal point of the scene), and you could use them as a basic structure for fashioning similar rules in other games:

• What is the effect of moving through a crowd? (A moving crowd?)
• What happens when a crowd panics?
• How can the PCs manipulate crowds?
• What happens when the crowd turns into a mob? (i.e., a crowd that can take focused violent action, whether directed or random)

If you want to keep it simpler, though, I have a few quick rules of thumb for handling crowds.

First, give the crowd some basic characteristics so that it “exists” in the scene. I recommend:

• Making the crowd difficult terrain (or whatever the local equivalent is in your current RPG). As mentioned above, let the PCs make an Acrobatics check as part of movement to ignore this (by deftly weaving through the crowd).
• Having the crowd offer cover to anyone in it.

Design Note: These two factors have a nice balancing effect — the cover encourages a character to move into a crowd; the difficult terrain imposes a cost for doing so.

Second, put the crowd on your initiative list. In D&D, I like putting them at initiative count 10 (so that PCs might go before or after the crowd, depending on their initiative check). This is, if nothing else, a great way to make sure you don’t forget to include the crowd in the scene.

Whenever the crowd’s initiative count comes up, the crowd does something. This might be:

• Just a colorful description (which will help make sure that the crowd is a consistent part of the scene and doesn’t get forgotten about or glossed over).
• A bystander in the crowd being placed in jeopardy.
• A random character needs to make a saving throw or take damage.
• Make a saving throw or get knocked down.
• The crowd moves.

And so forth.

Make sure to have the crowd affect (or potentially affect) both NPCs and PCs.

Third, create a short list of crowd actions. These work like legendary actions in D&D 5^th Edition: The crowd has actions or reactions they take after another character’s action, and they can take X of them per round. (Let’s say three, by default.) The things they can do will be similar:

• Knocking people down
• Interfering with attacks (the jostle the archer’s arm, inflicting disadvantage on the attack roll)
• Making an attack against a character
• Moving

Et cetera.

Keep in mind that the crowd is not a bad guy, so these actions are a choice you’re making as the GM to model the crowd’s behavior. This also means that some crowd actions might actually be detrimental to the crowd. For example, a crowd reaction might be “1d6 bystanders get caught in the crossfire.”

NEXT:
Campaign Journal: Session 27D – Running the Campaign: Trigger & Stitch
In the Shadow of the Spire: Index

DISCUSSING
In the Shadow of the Spire – Session 27B: Sights of Venom

Ranthir used his more powerful spell of clairvoyance to peer into the room… and there, standing in the midst of wrecked furniture and miscellaneous debris, he saw two massive, insectoid creatures.

At the sight, he blanched.

As he watched, one of the creatures reached out with its sharp talon and literally drilled the still-drafting curtain into the wall, pinning it in place.

In our last installment of Running the Campaign, we talked about what happens when the PCs miss clues. That actually continues into this section of the session: The project site (i.e., the apartment building controlled by cultists) had been prepped using status quo design. That meant that everything inside the building was basically held in a state of plausible stasis up until the point that the PCs interacted with it.

Once Ranthir cast his clairvoyance spells, therefore, and peeked inside, that status quo was disrupted and events started playing out. One of those events was the argument between members of the Ebon Hand and the Brotherhood of Venom. My anticipation had been that some very important information would get dropped during this conversation (i.e., clues), but because Ranthir was (a) using a spell which only granted sight, not sound; and (b) he couldn’t read lips worth a damn, most of that information was forever lost.

(Well, until the PCs gained it in a different way. Three Clue Rule and all that.)

As you’ll see in future sessions, the decision here to briefly engage the project site (setting events in motion) and then almost immediately withdrawing (“Let’s get out of here.”) had a significant impact on how subsequent events would play out.

But there was also something else the PCs did here that I didn’t expect:

The apartment building being used by the cultists was one of several similar buildings lining Crossing Street. Since Ranthir would only be able to target two specific locations with his spells, they decided to scout out the other buildings to get a better sense of what the layout might be like inside the cult’s building.

This tactic emerged because I have a giant, 8-foot-long map of Ptolus hanging on my wall during sessions, which meant that the players could see exactly what these buildings looked like:

But such a moment could easily arise in any number of ways. The key point here is that the PCs unexpectedly went into a building I had not anticipated them going into.

Now what?

This, of course, is exactly why so many video games nail the doors shut on all the buildings in town.

IN THIS CASE…

In this case, the players’ proposed reason for going into the building conveniently gave me the solution: They hypothesized that the neighboring apartment buildings, although slightly different in size, would have similar floorplans to the project site. I had floorplans for the project site, so it was relatively easy for me to just use those floorplans as the basis for some quick improvisation.

This exact scenario probably won’t crop up that often for you, but the general principle can be more broadly applied: Grab a floorplan you already have prepped — from the current session or perhaps from a previous session — and use it.

Just like these apartment buildings, the similarity of the buildings can be quite diegetic: The world is filled with structures built to a common floorplan.

MAKE IT UP

Obviously the easiest thing for me to say is, “Just make it up.”

Easy to say and great if it works. But improvisation takes practice and, honestly, no matter how much practice you get, there’ll still be times when you come up dry. That’s what the rest of this article is for.

But before we dive into that stuff, a quick word about making it up: Don’t feel like the whole building needs to spring full-blown from your brow like Athena doing Doric cosplay. You can build it up over time, describing only what the PCs need to know at any given moment. As play proceeds, a sketchy understanding of the building will start filling in with details.

A few thoughts on this:

• The first thing you’re likely to need is the exterior of the building. What’s the first thing that pops into you head when you think of the building? Describe that.
• If it’s a tactical situation, a key thing here will be entrances (do more than one) and windows.
• The second thing is to think about why the PCs are interested in the building: They’ll have probably already told you. (They’re looking for the CEO’s office. Or they’re trying to get to the roof. Or they want to hack the mainframe.) Roughly speaking, where is that stuff? First floor? Basement? Top floor?
• Once they pick an entrance, describe the lobby or front room or kitchen or whatever it is they see when they go through that door.
• Once again, think about where the exits are and start getting a sketchy feeling for where they might lead (with some thought for how they might connect to the PCs’ goals).

And then proceed along those lines.

But it also doesn’t have to be that complicated!

It’s very often true that you don’t actually need a floorplan at all.

For example, if the PCs have come here to meet with the CEO, you don’t need to know the whole building. In fact, you can probably just cut straight to a scene in the CEO’s office.

On the other hand, if you do need a floorplan and you need it right now (it’s a tactical situation, you’re playing with a VTT, etc.), then you can…

GOOGLE IT

Just hit up a search engine and type in whatever building type you’re looking for plus “blueprints” or “floorplans.”

This tends to work most reliably with modern buildings, but adding “fantasy” or “science fiction” to the search can often pull up what you need. (More reliably with the former than the latter.) These days if you add “RPG,” too, you’re likely to get a full-blown battlemap more often than not.

BUILD YOUR STOCK

Instead of scrambling with image searches at the table, you can get ahead of the game by building up a supply of stock floorplans for common locations.

• 4 or 5 different houses
• 2 or 3 warehouses
• 3 or 4 offices buildings
• A shopping mall

That sort of thing.

You can make a big push to assemble this in a marathon prep session, but it’s also something you can slowly build up over time: When you prep an adventure with a house, for example, tuck the floorplan for that house into the accordion folder or computer directory where you’re keeping your generic stock of floorplans. Over time you’ll just sort of accrete what you need.

Either way, you’ll slowly develop a sense of exactly what type of floorplans you’re likely to need, and that knowledge can often transfer from one setting to another. (As can many of the floorplans, in fact. Particularly if they don’t need to be seen by the players.)

RANDOM FLOORPLANS

Another option is to use a random generator to create the floorplan you need on-the-fly.

If you poke around a bit, you can find a number of online generators, like this Random Inn Generator from Inkwell Ideas. Collect these links in your digital notes and you can get something like this with the press of a button:

Personally, I prefer a more generic generator that I can use with just dice-and-paper. You can find the tool I use in my article on Streetcrawling Tools. It provides enough of a scaffold that I can iterate the rest, but is generic enough that I generally only need the one tool. That way I don’t feel overwhelmed hunting for precisely the right tool if my stock of generic floorplans doesn’t have exactly what I’m looking for.

It’s sort of the multitool of, “Oh crap, my players just went into a random building!”

NEXT:
Campaign Journal: Session 27C – Running the Campaign: Playing to the Crowd
In the Shadow of the Spire: Index