Session 19B: Beneath the Foundry

Tee missed all of this. Hearing the explosion she vaulted through the ventilation window and back onto the roof. Scampering twenty feet or so across the clay tiles and then looked through the ventilation windows about the first foundry. From this vantage point she could look down into the materials storehouse.

Although this week’s campaign journal is Session 19B, I’m actually going to continue chatting about the big, messy three-way confrontation between the PCs, the Shuul, and Shilukar that was wrapping up in the first part of Session 19.

This whole sequence worked really well at the table: The stakes were high. It was well-established that the PCs only had one shot at this. The battlefield was a complex, three-dimensional arena. The combination of action and stealth – which was being pursued in a mixture by both PCs and NPCs alike – gave the whole thing a very unusual texture and forced a lot of creative thinking (from both players and GM alike).

One of the essential elements that went into making this sequence work as well as it did is that I was blessed with a group of players willing (and able!) to seamlessly firewall player knowledge from character knowledge: When Tee split off from the group and got into trouble, the other PCs didn’t act as if they were gifted with clairvoyance and knew exactly what was going on. But they also didn’t fight so hard against the meta-knowledge that their characters turned into morons. Instead, they very smoothly took in what their characters knew and acted accordingly.

As the GM, I helped this process by specifying what information was flowing where: Something has just exploded in Room A, so characters X and Y hear the explosion; Y also sees the flash. And so forth. (These are really just crossovers, right? And they can be handled fairly seamlessly as brief “recap orientations” when you cut to the next group of PCs: “Okay, so you’ve just heard an explosion coming from the far side of the building. What are you doing?”)

By explicitly providing this information to the players, I’m removing the need for them to process it for themselves. They don’t have to think, “Okay, so the explosion just happened over there. Would I hear the yelling and then the explosion? Or just the explosion? Would I know where the explosion was? Or just a general direction?” All they need to do is focus on taking the information in as if they were their character and then making decisions based on the information they have.

THE GM’S FIREWALLS

The trick to doing this as a GM is to basically half-pretend that the other half of the party doesn’t exist when you cut between them. For example, let’s say that this wasn’t a situation with a split party: The PCs are exploring an area and, for whatever reason, an explosion goes off in the distance. What information would you give to the players in that situation? That’s the exact same information you should give to them even if the other half of the group were the ones causing the explosion.

This is kind of like a firewall in your own head, but rather than preventing meta-knowledge and character-knowledge from getting muddled up together, you’re preventing what Character A knows from getting muddled up with what Character B knows. You have to keep that clarity of perception clear in your own head so that you can present it clearly to the players, too.

Now, I say “half-pretend” because in actual practice the players DO know how you already described the scene of the rest of the group and you’ll use a sort of verbal shorthand to quickly review what they know without belaboring the details over and over again.

Which is one of the reasons why it’s great to have a group that can do this firewalling effectively. If I’d needed to take players into other rooms or pass notes or whatever, the pacing on this sequence would have suffered. Not only because of the logistical hassle of physically moving players around or writing out notes, but also because of the need to repeat information that otherwise would have only needed to be established once.

The GM also has to maintain firewalls between the NPCs. In fact, one of the quickest and easiest ways to make your NPCs feel like real people instead of puppets is for them to clearly demonstrate that their knowledge doesn’t map to the GM’s knowledge.

(An advanced technique you can use is to “cheat” this firewall in order to mimic NPCs with genius-level intellect that outstrips you own: In much the same way that it’s easy to solve a puzzle if you know the answer, so, too, can your NPC Sherlock Holmes make amazing “deductions” about the PCs because you already know the solution. But using this technique effectively is actually more difficult than it might seem, as it can easily lead to player frustration.)

This session also provides a great example of this kind of NPC firewalling, with both Shilukar and the Shuul being possessed of very different (and very incomplete) sets of facts.

You’ll also notice that, as the PCs figure this out, they’re able to take advantage of it in order to manipulate the NPCs.

OTHER FIREWALLS

Let’s back up for a moment: I said that passing notes and/or taking players into other rooms in order to have private conversations can have a negative impact on pacing. Does this mean I’m saying that you should never do this?

Not at all.

There is a cost to be paid for this stuff, but there are any number of circumstances in which the pay-off is worth it. The key principle, perhaps, is that you can’t put the genie back in the bottle: If you want to create an experience like surprise, paranoia, or mystery then it’s not enough to just ask someone to pretend that they don’t know a thing. They have to actually not know it.

Here are a few examples of where I’ve guarded information in order to prevent some or all of the players from gaining meta-knowledge.

• In the Ego Hunter one shot for Eclipse Phase, each PC is playing a forked version of the same character. Each fork has access to a unique subset of information and also a unique goal. I prepped custom handouts and took each player aside for private sessions. (The scenario is based around paranoia, secret agendas, and also the discovery of character and identity through incomplete information.)
• Similarly, in the Wilderness of Mirrors structure I designed for the Infinity RPG, each PC has a secret agenda. As the name suggests, the goal is to create paranoia and uncertainty in a universe filled with warring factions.
• In the Tomb of Horrors, when PCs choose to move through magical portals (that I know are one-way and, therefore, they cannot return through) my preferred method of resolution is to begin strict timekeeping and keep records of when the other characters pass through the portal. I can then jump to the other side of the portal and begin resolving actions as the PCs arrive one-by-one on the same schedule. (This heightens the extreme paranoia which is at the heart of the scenario.)
• In the Ptolus campaign, Tee’s decision to keep the Dreaming Arts and the other secrets of her elven clan secret from the rest of the PCs was the player’s choice. (Which is, at least 19 times out of 20, a good rule of thumb to follow: If a player requests a private meeting, honor the request. There’s some reason why they feel strongly about keeping this information secret, and you should generally default to respecting that.)

Session 19A: Routing the Shuul

Tor had caught sight of Tee out of the corner of her eye. Tee, looking down the street, saw Tor give an almost imperceptible nod of acknowledgement. Tee gestured frantically down towards the front door, trying to make Tor understand that Shilukar was heading their way.

Let’s briefly review what brought us to this point.

First, back in Session 17, Shilukar managed to escape when the PCs raided his lair.

This was not the expected outcome: Once the PCs tracked down his lair, I’d assumed that Shilukar would be either captured or dead by the end of their raid.

Second, the PCs chose not to immediately take the Idol of Ravvan from Shilukar’s lair, considering it to be too dangerous for them to handle.

This was not the expected outcome: I’d assumed the PCs would loot the Idol. When they left it in situ, I was left scrambling and had to look at the totality of the scenario I had designed to figure out what would happen next.

The short version is that there was a third party that had been keeping Shilukar under observation and they took advantage of the unguarded lair to sneak in and grab the Idol. (I’d created this third party to satisfy the Three Clue Rule by providing one possible path the PCs could follow to Shilukar’s Lair. The PCs never followed the trail that would have taken them to this clue, but that didn’t mean the third party observers weren’t still there.)

The PCs wouldn’t directly cross paths with this third party until much, much later in the campaign, but it would ultimately result in the creation of a completely new scenario in Act II. If you’re reading this several years in the future, you can track where the Idol ends up, look at the campaign-transforming consequences that happen as a result, and kind of have your mind blown when you look back and realize it all boils down to this one decision.

Third, the PCs very cleverly think to hire Shim to find Shilukar.

Before this happened, I actually assumed that the PCs had basically completely failed here: Shilukar was in the wind and there was no clear path (or, at least, no prepared path) by which they could find him before the deadline he had given Lord Zavere expired. I assumed the next time they saw Shilukar, he would be a well-protected “guest” of Castle Shard.

(Another interesting decision the PCs make here at the dawn of Act II: To not tell Shim that they had lost their memories and to NOT hire him to investigate their period of lost time. I had really assumed that one or both of these things would happen, which would have potentially revealed a lot of new information about what they had been doing.)

Fourth, because the PCs had hidden their identities from Shilukar during Session 17, I had (for various reasons) concluded that Shilukar would blame the Shuul for the assault on his lair. As a result, I noted in the campaign status document for Session 18:

09/08/790: Shilukar is reported to have broken into a Shuul facility.

Shilukar also wanted his Idol back, and if he thought the Shuul had it I figured he would go looking for it.

As a result, however, when the PCs asked Shim to track down Shilukar, I looked at the totality of my notes, made some skill checks for Shim, and concluded that he must have somehow discovered the upcoming heist.

“I don’t know what you’re doing in Agnarr’s room, but Shilukar is planning to attack the Foundry in the Guildsman’s District in less than 30 minutes. If you want him before dawn, this will be your only chance.”

All of which ultimately brings us to the PCs staging a raid on the Foundry in order to capture Shilukar, who is simultaneously staging a raid on the Foundry.

Why the Foundry? Well, I hadn’t specified which Shuul facility Shilukar would have been attacking overnight. The Foundry, however, is the only Shuul facility which is thoroughly detailed in Monte Cook’s Ptolus, including – most importantly! – a map. When you’re faced with unexpectedly improvising an entire scenario in the middle of a session, it’s not bad practice to default to whatever tools you’ve already got at your fingertips.

So at this point I:

• Grab my Shilukar stat block.
• Go to page 186 of Ptolus and grab the map of the Foundry.
• Go to page 133 of Ptolus and grab the stat blocks for the Typical Shuul Agent.

I look at the map and:

• Since it’s the middle of the night, conclude that the Foundry is mostly shut down for the night.
• Figure out what sort of guard the Shuul would place on the facility: Guards go here. A pair of patrols around the outside of the building.
• Ask myself, “How would Shilukar broach these defenses?”

You saw the answers to these questions in Session 18C (which is when all this improvisation was going on).

At that point, I can turn things over to the players.

They sneak up and put the building under observation. They see the two patrols circling the building. Tee’s player asks me if there’s a blind spot in which she could sneak into the building. I look at the map: I’d decided the patrols were equidistant around the building to maximize their field of view, but because of how the warehouse sticks out on one side of the building… yeah, there’s an unusually long period during which neither patrol would have their eyes on that corner.

From this point forward, I’m just playing the scenario: I know what Shilukar will do. How do the Shuul respond to that? As the PCs take actions, how do all of my NPCs react to that? Having put the pieces into play, now I’m free to just live in the moment and see what happens.

FINISHING TOUCHES

Because we had to wrap for the night before the raid on the Foundry was finished, however, I was able to spend some time between sessions polishing things up. This was particularly relevant when it came to the lower level of the Foundry, which I was able to spend a few minutes fleshing out with a proper key.

In addition to being able to give some meaningful thought to what various projects the Shuul might be working on down there, this also gave me the opportunity to seed the Foundry basement with a number of clues pointing to other nodes in the campaign.

And as we’ve been discussing unforeseen outcomes and long-ranging consequences, consider the letter they find from Maeda to Brother Savane down there.

That letter doesn’t exist if the PCs don’t make some very clever choices to bring them some place I never thought they’d be. And it also probably doesn’t exist if the timing of the sessions had been a little different, the raid hadn’t been split up across two sessions, and I hadn’t gotten the extra prep time to write it up.

And it literally changes everything in the campaign.

This is Session 19. You can literally go to Session 119 and discover that the entire session is dependent on this letter existing and the decision Dominic makes as a result of it.

Session 18C: The Smoke of the Foundry

Sun Tzu said:

If large numbers of trees move, they are approaching. If there are many [visible] obstacles in the heavy grass, it is to make us suspicious. If the birds take flight, there is an ambush. If the animals are afraid, [enemy] forces are mounting a sudden attack.

Back in Session 13 I discussed the value of having a toolkit of basic tactical techniques in your back pocket as a Game Master. The headline for that essay was “Simple Simulatonist Tactics,” and the reason I did that is because I believe that the Game Master’s tactical acumen is not necessarily limited to the diegetic tactics of their NPCs.

In other words, the tactical techniques you use as a GM don’t always map to what would purely constitute tactics in the real world. You can think of tactics as being the techniques by which you achieve your desired effect upon the battlefield: In the real world / simulationist play, the desired effect is usually victory. (Often the method used is to manipulate your opponents’ thoughts and their decisions – see the Battles of Trebia and Cannae, for example – but ultimately you goal is victory.)

In dramatist and gamist play, however, your tactical choices as the GM may be made to achieve ends other than victory. Instead, you use tactics to achieve either dramatic effects or to create interesting challenges.

For example, the combat system in D&D 4th Edition is heavily designed to create My Perfect Encounters™. These encounters are balanced on a razor’s edge in order to create a gamist tactical challenge, and in order for the game to work properly the GM needs to make tactical choices appropriate to that paradigm.

When I was playing the D&D Gamma World version of this ruleset, for example, I once forgot what system I was using and had the bad guys perform a tactical retreat from a combat they were losing and seek reinforcements: This completely unbalanced the precarious encounter balance of 4th Edition and resulted in a near-TPK. (You can read a fully playtest report here.)

Even when a system isn’t mechanically tying your hands like this, though, you can still make decisions like this: It would be more interesting, from a gamist perspective, to fight the Big Bad Guy in his stronghold, so he won’t come charging out from his sanctum to save his minions. It’s dramatically more interesting to fight a sequence of elementally-themed bad guys, even if it would make more sense for the bad guys to form mixed-force tactical groups. And so forth.

HEAR THE REINFORCEMENTS

Here’s a simple dramatist tactical technique your can add to your toolkit: Letting the PCs hear the reinforcements coming.

(To be clear: This is certainly something that can arise naturally out of simulationist play – as the result of opposed Listen and Stealth checks, for example – but if the GM is specifically choosing to let the PCs hear the reinforcements coming order to create effect, that’s a dramatist decision.)

This is a great technique because it gives the players a space in which to make decisions that aren’t possible when the reinforcements just burst into the middle of the encounter: How can they prepare for what’s coming? Is there a way that they can delay the arrival of the reinforcements? Do they take the opportunity to withdraw before the reinforcements arrive? Do they pull out some heavy guns to clear their current opponents? And so forth.

One of the Shuul who had been in the front hall, satisfied that Tor and Dominic were pacified, turned and headed back through the hall and into the materials storehouse. Two more turned and headed into the second foundry, directly below Tee’s feet. In the sudden silence pervading the Foundry, Tee could hear their footsteps echoing ominously below her…

You don’t really get the Platonic ideal of, “Oh shit! There are more goblins coming!” in this multi-faceted conflict between the PCs, Shilukar, and the Shuul, but there are numerous examples of how auditory cues of what’s happening elsewhere can affect the immediate battlefield.

In this case, the basic technique of hearing the reinforcements coming is being complicated through the more advanced techniques of crossovers. In practice, this single battle is being treated as several different encounters, with things happening in one battle crossing over (often through those auditory cues) into the other encounters. (The proximity of those encounters also means that characters – particularly the NPCs – are often rushing between one encounter and another.)

Session 18B: Missed Opportunities

“The Crimson Coil?” Elestra said. “I think I’ve heard something about them. Random acts of violence. Vandalism. That kind of thing. I got the impression they hadn’t been around for years, though.”

“That’s right,” Tee said. “I was still living here. The cult members wore blood-red robes and hoods. They’d spontaneously appear in huge gatherings to wreak random chaos. Then, about two or three years ago, the Knights of the Pale tracked them to their stronghold – I think it was called Pythoness House. Reportedly the whole cult was wiped out.

Last year I wrote Rulings in Practice: Gather Information, which discussed how to handle information-gathering tasks like canvassing and research.

The 3^rd Edition of D&D, however, features a very interesting and fairly clear-cut distinction between “information you gather” (in the form of a Gather Information check) and “information you just know” (in the form of a Knowledge (local) check). Given that the information in question was often the exact same information, I found it necessary to really figure out exactly what the relationship was between these skills and how they should be handled in play. It’s a distinction that I think remains valuable in subsequent editions of D&D, including Pathfinder (which has the exact same distinction) and even 5^th Edition (which lacks the specific skills but nevertheless will run into similar issues of adjudication).

The short version:

Knowledge (local) gets you the same information, but at +5 DC.

Knowing something off the top of your head is simply more difficult than going to look it up. From a balance standpoint, the +5 DC is also the cost you pay for not needing to spend 1d4+1 hours (in the case of 3^rd Edition) looking for the information.

This same basic principle can be applied widely, even in systems that don’t mechanically distinguish between different forms of knowledge acquisition: In some cases, depending on the exact mechanics involved, you might test to see if they know something off the top of their head and then, if they spend the time necessary to actively look for that information, you can apply a bonus to their check.

(If you wanted to focus even more attention on this aspect of game play, you could even vary the size of that bonus based on how they actively look: Better libraries give better research bonuses, for example.)

In practice, there are some forms of information that very specifically require canvassing and, therefore, Knowledge (local) simply isn’t applicable for. (Did anyone see Bob Anonymous in the silver light district on Friday night? You need to actually go down there and ask around.) In practice, though, I recommend generally being pretty liberal in what you allow people to now.

ACCLIMATION

The other issue raised by Knowledge (local) was exactly what “local” consists of and also what happens with the skill when you leave one “local” and go to a different “local.”

As a loose guideline, I would apply a -4 penalty to Knowledge (local) checks on a scale from whatever your “local” was: Different city in the same area. Different region in the same kingdom. Different kingdom. Different continent (or similar divide). Different plane.

(So the penalty would be -8 if you were from the boondocks and went to the capital.)

This penalty could be quickly abrogated by a character who became acclimated to a new locale. At various times I would use a period of a fortnight or a month, but at the end of each period the size of the penalty would be reduced by -1. (That reduction was against the base -4, not the total penalty.)

The logic was that part of your Knowledge (local) skill is, in fact, the knack for picking up local knowledge.

On one occasion a player wanted to speed this process up, so I allowed them to make Gather Information checks once per day (and with the normal time required) vs. DC 15. Each success reduced the number of days required by the margin of success on the check.

This never really came up in the Ptolus campaign, of course, because it’s all set in one city. But in other campaigns I found it was a nice balance between making the Knowledge (local) skill useful for globetrotting characters and also allowing characters to enter strange places f without instantly being experts on their minutia.

Session 18A: Key to the Past