DISCUSSING
In the Shadow of the Spire – Session 30A: The Breaking of the Dawn

The red-sashed knight approached with his sword drawn. Tor and Kalerecent stood calmly, careful to give no cause for alarm. As the knight drew nearer they raised their hands and displayed their rings. The knight relaxed slightly, but kept his blade on guard.

“What happened here?” Tor asked.

“Sir Kabel attempted to assassinate the Novarch.”

If I could only give one tip to GMs about pacing in RPGs, it would be to do a cliffhanger every single time the opportunity presents itself. It’s virtually impossible to have too many of them.

We’ve previously discussed cliffhangers at the end of sessions, but here we have a cliffhanger happening in the middle of a session. This is made possible by the fact that the players have split the party: If they were all together, I wouldn’t be able to cut away from Tor immediately after delivering the shocking news that Sir Kabel has attempted to assassinate the Novarch.

This is one of the primary reasons why, in The Art of Pacing, I described splitting the party as pacing on easy mode: There are just so many extra tools you have at your disposal as soon as the PCs are no longer all together in the same scene.

The trick, of course, is getting the PCs to split up in the first place, particularly when “don’t split the party” has become such a maxim in RPG fandom.

The key to this is that the PCs need to have multiple desires which cannot be resolved sequentially (i.e., they either have to both be done right now, or they can’t be done or become much more difficult to do). This tends to rather difficult to pull off with a linear adventure, but often happens all the time and with little or no effort with non-linear scenarios that you’re actively playing.

In this case, of course, the PCs want to both help Kalerecent take Rasnir’s body to the Godskeep AND keep the Banewarrens securely guarded. They can’t be in two places at the same time, and so splitting the party becomes inevitable.

CAMPAIGN COLLISION

What happens over the next session and a half is one of my favorite moments form the entire campaign. And the fact that it kicks off with this scene — of two knights of the Order of the Dawn bearing the body of their dead comrade home at the very moment that the Order is breaking in a bloody conflict — is, if I may say so, about as perfect as one could hope for.

Which is why it’s so interesting that I didn’t plan for any of this happen.

Let’s peel back the curtain here and take a closer look at how this played out.

First, as I’ve previously discussed a bit, the schisming of the Imperial Church was intended to play out as a background event. It was intended to add some depth and flavor to the campaign world in a way that was, at best, tangentially related to what the PCs were doing.

But Dominic unexpectedly presented himself to Rehobath as the Chosen of Vehthyl, which allowed Rehobath to move up his timetable and declare himself Novarch several weeks earlier than I’d expected. And then Tor ended up getting squired in the Order of the Dawn, placing two of the PCs at basically ground zero.

The schism was now very much onstage.

Second, I had keyed the Breaking of the Dawn — in which Sir Kabel gathered loyalists within the Order at the tournament field north of Ptolus to arrange the arrest of the “False Novarch,” only to be betrayed by Sir Gemmell — to my campaign status document as a timed event: It was going to take place at a specific date and time.

Third, Tor — completely oblivious to this — made plans to take Iltumar the would-be hero to the tournament field and do some practice swordplay with him in an effort to give his aspirations a path that didn’t lead straight to the chaos cults. By sheer coincidence, Tor scheduled this training excursion with Iltumar at the exact same time Sir Kabel was going to be at the tournament field.

This prompted me to prep the events of the Breaking of the Dawn in much more detail — basically as a mini-scenario, since it now seemed quite likely that Tor would be directly involved. But then the evolving situation with the Banewarrens caused Tor to cancel his plans with Iltumar!

Regardless, the Breaking of the Dawn was still keyed temporally.

The fourth element here, of course, is Kalerecent. Rather than being keyed to a specific time, Kalerecent was keyed in a status quo: Whenever the PCs arrived at the Banewarrens, he would be waiting with Rasnir’s corpse. (A sufficiently long delay in the PCs reaching the Banewarrens, or if they had come to the Banewarrens and then left again before actually meeting Kalerecent, might have changed that. But that’s purely hypothetical since it didn’t play out that way.)

So in my prep notes, these two things — Kalerecent wanting to take Rasnir’s body back to the Godskeep after being assured that the Banewarrens were secure and the Breaking of the Dawn — were completely unrelated to each other. It was entirely coincidental that things played out this way. And, in fact, it’s quite easy to imagine a scenario in which:

• none of the PCs chose to accompany Kalerecent;
• Tor stayed in the Banewarrens (“as a fellow member of the Order, I’ll take up your oath, Kalerecent, until you can return”) while some other group of PCs accompanied Kalerecent;
• the PCs screwed up and the Pactlords killed Kalerecent when they returned to the Banewarrens;

or any number of other possibilities.

That’s really the beauty of prepping scenarios that can be actively played: You never know how all of your disparate toys will come together to create something of astonishing and unexpected beauty.

Campaign Journal: Session 30B – Running the Campaign: Spray Your Bullets
In the Shadow of the Spire: Index

SESSION 30A: THE BREAKING OF THE DAWN

October 12th, 2008
The 16th Day of Kadal in the 790th Year of the Seyrunian Dynasty

Agnarr beat on the wall where the umber hulk had disappeared.

“Maybe it’s gone away to die,” Elestra suggested.

Agnarr took small comfort from that small hope, but pursuit was impossible and they didn’t dare to go any deeper into these unknown caverns in their current state. Nursing their wounds, they retreated back towards the Banewarrens.

When they returned to where Kalerecent was waiting for them, they found him with his sword drawn. Seeing that it was them, he sheathed his sword and hailed them. He had heard the sounds of combat and been worried for their fortunes.

They described their encounter with the lamia, the arrival of the reinforcements wearing bone rings, and their eventual triumph. Kalerecent nodded. “I’m not sure how many goblins there were, but I think you’ve disposed of those I fought before.”

“Except for the green-skinned crone who went through the door,” Ranthir said.

“Yes. Except for her.” Kalerecent’s face was grim.

But with the other creatures slain, Kalerecent felt the complex was secure enough that he was willing to return to the surface with Rasnir’s body… as long as some sort of watch was kept on the door. Tor had sworn himself to help Kalerecent in bearing Rasnir back to the Godskeep, but the others agreed to stay behind and keep a watch over the Banewarrens.

Kalerecent carried the body himself. Tor walked at his side. The others came as far as the mansion before bidding them goodbye.

Once Kalerecent was out of earshot, they returned to the closet where they had left the orc woman tied up. Yanking it open, they found her still bound and gagged inside.

They removed her gag and showed her the weapons and rings of her comrades. “We killed them,” Tee said. “We can kill you. Now tell us what you know.”

But the orc woman was reticent. She sneered. “You have only made it more certain that I will never speak to you.”

“Aren’t you speaking to us now?” Agnarr frowned.

Dominic leaned in close. “Can’t we speak to the dead?”

But the orc woman said nothing.

“I don’t think it’s working,” Elestra said.

“No, seriously, I’m asking: Can’t we speak to the dead?”

Tee gagged her again and shut the door.

THE GATES ARE SHUT

Kalerecent only had to carry Rasnir’s body as far as the Emperor’s Road. Once there, Tor was able to hail a carriage. (They had to pay the carriage driver a little extra to bear Rasnir’s body.)

They rode in silence down into the Temple District. As they were approaching the Street of a Million Gods, however, the carriage began to slow. They were still a few blocks away from the Cathedral.

Tor leaned out the window and looked up at the carriage driver. “What’s going on? Why are you stopping?”

“Some sort of disturbance, sir,” the carriage driver said. “We should be able to ride through, though.”

Tor could see it now: Clotted groups of people were streaming down the street in the opposite direction.

Kalerecent frowned. “I don’t like this.” He opened the door and stepped out of the carriage. To the driver he said, “Wait for us here. And guard the body of my friend with your life.” To Tor he said, “Are you coming?”

Tor nodded and climbed out of the carriage, as well. He tossed a few coins to the driver to “compensate for his pains” and followed Kalerecent down the street.

As they approached Discourse Street, the furrow of Kalerecent’s brow deepened. “The gates of the Godskeep have been shut.”

“Is that unusual?” Tor asked.

“I’ve been a knight of the Order for seven years. I’ve never seen them shut before.”

Something was wrong. Kalerecent led the way as they circled around towards the south gate of the keep. As they passed between two of the Cathedral’s outbuildings and emerged into the courtyard between the Cathedral and the Godskeep, they were met with a grisly sight: The south gate, too, had been shut and dozens of bloodied bodies were strewn across the grassy sward. Members of the Order wearing red sashes were moving between the bodies. Some where being healed… most of them were not.

Kalerecent stepped out into the open and Tor followed his lead. One of the red-sashed knights inspecting the corpses spotted them and stood up. “Halt!”

He approached with his sword drawn. Tor and Kalerecent stood calmly, careful to give no cause for alarm. As the knight drew nearer they raised their hands and displayed their rings. The knight relaxed slightly, but kept his blade on guard.

“What happened here?” Tor asked.

“Sir Kabel attempted to assassinate the Novarch.”

Running the Campaign: Cut on the Cliffhanger  – Campaign Journal: Session 30B
In the Shadow of the Spire: Index

Go to Campaign Status Documents

If you’re running a sandbox campaign with ‘crawl-type scenarios (dungeoncrawls, hexcrawls, urbancrawls, etc.), you’ll most likely want to restock those scenarios.

For example, let’s say that there’s a ruined keep west of the village in Hex F6. As the campaign begins, this dungeon is filled with rausling bandits who have been staging raids on the merchant caravans and other travelers passing through the area. At some point, the PCs decide that they’re tired of being harassed by rauslings, so they track the bandits back to the keep and wipe out them out.

The keep now stands empty. That’s a really easy form of status quo, and it could persist for any length of time. But do you want the keep to stand empty forever?

Quite possibly not! In fact, it can be very interesting to have someone else move into a dungeon the PCs have already cleared out once. Going into a dungeoncrawl when you already know the layout of the dungeon is a distinct experience and a very different strategic challenge. Plus, the changes made by the new inhabitants can provide cool surprises!

So what I’ll do when running a hexcrawl sandbox is simply list all of the locations that are currently empty. After each session, I’ll go down the list and make a restocking check for each location on the list (usually 1 in 6 or 1 in 8). If the check succeeds, I’ll figure out who’s moved in (more on that below), create a new version of the adventure notes, and remove the location form the restocking list.

Updating the scenario notes is actually quite similar to doing a dungeon status update, with most of the work being done by simply swapping in a new adversary roster. Rather than doing a “diff file” for the updated room key, though, I’ll take the time to briefly modify my original adventure notes, print up a new copy, and file it in the appropriate spot. (There’s no way to be certain when the PCs will actually re-engage with the dungeon, after all.)

Tip: Even in non-sandbox campaigns, it can be fun to restock and revisit old dungeons. For example, in my Ptolus campaign the PCs cleared out the Temple of the Ebon Hand. Later they hit another cultist stronghold, but some of the cultists escaped. These cultist refugees ended up taking refuge in the abandoned Temple of the Ebon Hand, where the PCs eventually tracked them down.

RESTOCKING MEGADUNGEONS

The same methodology can also be applied to megadungeons. Basically you want to split the megadungeon into sectors and then make restocking checks for each applicable sector.

There are a couple different approaches I’ve used for defining sectors. In my Castle Blackmoor campaign, I followed Dave Arneson’s lead and split each dungeon level roughly into quarters. In other megadungeons, I’ve typically taken a more bespoke approach of looking for the “natural” break points in the dungeon. (For example, if you’ve got two chokepoints in the dungeon, all the rooms between those chokepoints are a natural “sector” that’s likely to be occupied by a particular faction.) The most important thing, I think, is for the sectors to make sense to you. As long as that’s true, it will be much easier for you to keep track of them and also to interpret the “meaning” of your restocking checks.

There are two types of restocking checks you can use in the megadungeon: empty sectors and disturbed sectors.

Empty sectors work more or less like cleared hexes, checking each empty sector once per prep period. (Although I’ve found that 1 in 4 or 1 in 6 seem to have better results as intervals.)

Disturbed sectors, on the other hand, involve tracking each sector the PCs enter during an expedition into the megadungeon, and then checking each one during the next prep period. After checking the disturbed sectors from the previous session, you clear that list.

Of course, both techniques can also be used in tandem (checking both empty and disturbed sectors during each prep period).

Tip: These techniques can also be used with medium-sized lairs and other dungeons that don’t quite aspire to “mega.”

OTHER RESTOCKING CHECKLISTS

You can use similar techniques with other scenario types, too.

For example, if you’re running a heist scenario and the PCs get made during onsite surveillance. How long does it take for the target to figure out that something is wrong and increase their security? Obviously you could just make that decision, but you could also drop this onto your campaign status document and make daily “restocking” checks to see when the reinforcements get installed.

(This can become quite interesting in a more complicated campaign where the heist is only one of several things that the PCs are juggling at the same time.)

RESTOCKING PROCEDURES

The actual process of restocking can be considerably easier if you’re using a game system that comes packaged with procedural content generators. For example, you just have to flip through the 1974 D&D manuals to find detailed dungeon stocking tables and treasure generators. On the other hand, these tools are largely or completely absent from 5^th Edition D&D. (And most other RPGs.)

When these tools are absent, I’ll often make the effort to create them. There are also a few universal techniques you can use to fake it until you make it. For example, the AD&D 1^st Edition Monster Manual conveniently has almost exactly one hundred pages of monsters. I’ve done a whole bunch of ad hoc stocking over the years by just rolling percentile dice, flipping to the appropriate page, and then randomly picking a monster on that page.

Also think about how you can use tools you’ve already created. For example, if you’re running a hexcrawl or megadungeon you’ve probably created a random encounter table. You can use this same table to determine creature type(s) for your restocking. (This can be limiting, though, as it means you won’t be injecting anything new into your campaign. On the other hand, that may actually be desirable in many cases.)

Next: Event Fallout

Go to Campaign Status Documents

As I mentioned in my original article describing the use of campaign status documents, every campaign is unique and that means that the campaign status document for every campaign will be a little different. This series will be taking a closer look at some of the very specific tools I’ve developed and used for campaign status documents over the years, including examples drawn from actual play, and it will certainly make more sense if you’re familiar with campaign status documents.

I’m going to refer to these as campaign status modules. And, as I say, there’s nothing sacrosanct about them. The whole point is to take this stuff and adapt it, molding it to whatever the immediate needs of your current campaign are.

Also, importantly, I’ve never had a campaign status document that featured ALL of the modules we’ll be talking about. You really want to identify the stuff that you need and use that. (And only that.) Including stuff that you don’t need actually creates a negative value, filling your status document with a bunch of cruft that makes it harder to maintain the vital material and use it during play.

TRACKING DUNGEONS

I’d like to start by taking a peek at how I track dungeon status. This is a form of scenario update, and I did take a look at quite a few of the techniques I use in the original article.

• adversary rosters
• updated room keys
• scenario timelines
• return to the status quo

What I’d like to do here is look at some specific examples of this in practice.

The first thing is that, as with everything else on the campaign status document, you can and should keep it simple: Just because you have all of these tools available for tracking dungeon status, it doesn’t follow that you need to use them all. For example, here’s what the dungeon status for the Kolat Towers in my Dragon Heist campaign looked like in the Session 13 status document:

5.4 ZHENTARIM – KOLAT TOWERS

• Mimic killed.

• K12: Two destroyed staffs.

This was all that was needed given the changes that the PCs had affected at that point.

As more details are needed, the tools I most commonly invoke are the adversary roster and updated room key.

Prepping a dungeon around an adversary roster not only makes it much, much easier to run the dungeon as a dynamic environment, it also makes updating the dungeon incredibly easy: The things most likely to change about a dungeon, after all, are its denizens. During play you can easily update the roster as necessary (e.g., crossing off casualties). Then, between sessions, you can simply update the roster, put the current version in your status document, and ignore the original.

The flipside of this coin are the physical fixtures of the dungeon. The updated room key is a simple “diff file” that reminds you of any changes that have been made because of the PCs’ actions. (For example, if they rip a door off its hinges, then I should try to remember that next time they pass through that room.) I generally find that I don’t need more then one or two sentences to jog my memory for this kind of stuff.

Here’s an example update sheet for a dungeon from Monte Cook’s Banewarrens campaign:

CURRENT ENTOURAGE (10/20/790)

BW08 – LOCATION STATUS

• AREA 12: Great white shark corpse on edge of Conflagration (partially eaten by slaadi).

• AREA 12 – CAVE: Bison carcass.

• AREA 21: Xorn refuses to make alliance with the Golden One.

(The xorn here is listed in the key rather than the adversary roster because it won’t leave Area 21 under most circumstances.)

Design Note: In these notes, “BW08” is an alphanumeric code I use to refer to this specific adventure. (In the Dragon Heist example, “5.4” fulfills the same function.) These codes help me organize my notes and, as you can see here, make it easy to cross-reference the scenario (either in my campaign status document or another scenario).

The final piece of the puzzle, scenario timelines, are tied to the concept of status quo design: The dungeon exists in a literal or effective state of status quo (i.e., how it is described in your initial adventure notes) until it is perturbed by the PCs. For example:

• 10/05/790: Tee attacks Temple of Deep Chaos.
• 10/06/790: Tee unleashes nightmare on Arveth, leaving her fatigued next day.
• 10/07/790: Arveth hits Tee with Dais of Vengeance.
• 10/08/790: Tee unleashes nightmare on Arveth, leaving her fatigued next day.
• 10/09/790: Arveth switches sleeping pattern so that she won’t be asleep at night.
• 10/10/790 (4 AM): Arveth hits Tee with Dais of Vengeance (forced to watch her friends’ eyes ripped out).
• 10/10/790 (11 PM): Rissien and Santiel are kidnapped from Narred and taken to Temple of Deep Chaos.
• 10/13/790: Santiel is blinded.
• 10/14/790: Santiel’s eyes delivered to Tee.
• 10/15/790: Santiel is killed.
• 10/16/790: Rissien suspended in Kaleidoscope Temple.
• 10/21/790: Rissien is killed. (Possibly rescued by Dark Leaf.)

I use strikeout text to indicate events that have already happened. In some cases I’ll simply delete these entries, but I’ve too often found that it can be essential to easily reference this past continuity during play. In fact, for many types of actions, it’s far more efficient to simply list what happened (and then describing things accordingly) rather than trying to account for every individual change in the key.

It follows, of course, that the items which have not been struck out are stuff that hasn’t happened yet. They may, in fact, never happen. (If, for example, Tee catches up with Arveth before she can kidnap Rissien and Santiel.) Such events are generally based on the intentions and plans of the NPCs, and prepping them can be smart if (a) they’re sufficiently complex or convoluted that it will be valuable to puzzle them out between sessions, (b) juggling all the off-screen actions of the NPCs would be too difficult to handle during play, and/or (c) they would involve some form of additional prep (new stat blocks, physical props, etc.) that can’t be improvised during the session.

Once I start rolling out timelines, there are two key questions I ask:

• When is it likely that the PCs will re-engage with this dungeon? I won’t prep timelines much beyond that point because the likelihood of wasted prep becomes high.
• If the PCs don’t further interact with this dungeon, what will the new status quo be?

The latter question can be easy to overlook, but is a really essential component of efficient, smart prep. Some situations will just continue to spiral out of control (spinning their chaos out into the rest of the campaign), but a lot of scenarios will instead settle down into a status quo (e.g., the mafiosos bring in new muscle to guard their drug operation and then… that’s it, they’ve taken the precautions they think they need to take). You can simply prep up to that new status quo, file it in the appropriate section of your campaign notes, and then stop thinking about it until it becomes relevant again.

Next: Restocking Checklists

Go to Table of Contents

The Alexandrian Remix of Descent Into Avernus utilizes a number of additional books from Wizards of the Coast, the DM’s Guild, and other third-party content. (This wealth of support material is one of the great things about using published D&D campaigns.)

By popular demand, here’s an authoritative list of all the resources referenced/required by the Avernus Remix (not including Descent Into Avernus itself).

AVERNIAN HEXCRAWL

The hex key for the Avernian Hexcrawl includes locations referencing these books:

Warlords of Avernus
Forges of Avernus
Bitter Rivals
Dance of Deathless Frost
Vysianter’s Guide to the Red Wastes of Avernus
Tyrants of the Purple City
Encounters in Avernus
Infernal Encounters
Ruined Prospects
Diabolical Dive
Infernal Insurgency
Abyssal Incursion
Temple of the Broken Prince

Note that several of these were used for a single location and could be swapped out if you don’t want to go whole hog.

AVERNIAN RANDOM ENCOUNTERS

The Avernian random encounter tables pull demons and devils from a wide array of awesome monster manuals:

Mordenkainen’s Tome of Foes
Volo’s Guide to Monsters
Encounters in Avernus
Infernal Encounters
Emirikol’s Guide to Devils
Scientific Secrets of Avernus
Book of Fiends 5E
Avernus Rising: Fang & Claw

REVIEWS

If you’d like more information about the books listed above, most of them were detailed as part of the reviews series done as part of the Avernus Remix:

Capsule Reviews: DMs Guild
Capsule Reviews: Rhodarin Avernus
Capsule Reviews: Adventurers League (Season 9)

Links above are affiliate links and help support the Alexandrian! Hope this helps!