As a hobby exercise I've decided to convert my notes from an introductory AD&D adventure into a module and release it as a free PDF. I have no experience writing RPG adventures except for my own use, and only a vague notion of what makes a good module. I have lots of questions, I hope people here will be willing to help answer some of them.

Currently putting together the first draft. Work is busy at the moment so I haven't had any time this week but I reckon I should be able to finish it next week.

First problem: What's the best way to write Set Pieces?

Let's say the party enters a room and encounters a monster or NPC doing something. It's not the monster's lair, he's not posted there on duty and it's not a random encounter. He just happens to be in that place the first time the party goes there. There may be pretence for him to be there but in reality it's because that's where I want the party to encounter this monster for the first time. Is this bad design?

I was intrigued by this post on 1d30 blog which seems to say you shouldn't do that sort of thing. Instead you should have some kind of routine or schedule for determining what's going on in a room when the PCs enter it. I'm not sure his "better" room is better than the "bad" example, which seems a lot more interesting to me. How would you approach this? The post later says "Place encounters where you feel appropriate" which I'm not sure how to interpret in this context.

It's been said that a good dungeon should feel "alive" as if things are happening regardless of whether the PCs are there or not. That appeals to me, but it's ultimately a kind of an illusion; dungeons don't really exist when no-one's playing them. We don't plot the movement and activity of every individual creature in game time alongside the actions of the PCs, instead we have random encounter tables. At the end of the day every encounter only happens because the PCs are there to engage in it.

Using some method to decide if an encounter should or shouldn't happen now seems like potentially throwing away an opportunity to play the game, for no good reason. Or am I misunderstanding this?

Thanks

Originally I had a mini-scenario where the PCs encounter a group of itinerant priests on the outskirts of the dungeon. The priests were planning to delve further in, unaware of the awakened danger ahead. Saving the priests is a minor objective. The encounter happens "the first time the PCs enter this area" which is somewhat off the beaten track.

Thinking about what 1d30 wrote, I changed this to "when the PCs first reach the dungeon, the priests are three days travel away. Once they reach the outskirts they will rest for one day then, unless intercepted, enter the dungeon on the fourth day" plus there's a cool random table to determine what sort of gruesome fate they meet. It's more dynamic and "alive" now but is it better than just having them appear when the party needs them to?

Note the trigger for the timer is "when the PCs first reach the dungeon" so it still comes back to the problem(?) of things only happening because there are players around. I could set the timer running without the PCs doing anything, and if they stay in town for a week the priests all die with the PCs having no opportunity to save them, but where's the fun in that? I think there needs to be a happy middle-ground between blatant author fiat and joyless existentialism, I'm just not sure yet where to find it.

Quote from: Juisarian;836543As a hobby exercise I've decided to convert my notes from an introductory AD&D adventure into a module and release it as a free PDF. I have no experience writing RPG adventures except for my own use, and only a vague notion of what makes a good module. I have lots of questions, I hope people here will be willing to help answer some of them.

Currently putting together the first draft. Work is busy at the moment so I haven't had any time this week but I reckon I should be able to finish it next week.

Good luck and I hope it works out well.

Quote from: Juisarian;836543First problem: What's the best way to write Set Pieces?

Honestly? Read good scenarios, see what they did and use the same techniques.

Quote from: Juisarian;836543Let's say the party enters a room and encounters a monster or NPC doing something. It's not the monster's lair, he's not posted there on duty and it's not a random encounter. He just happens to be in that place the first time the party goes there. There may be pretence for him to be there but in reality it's because that's where I want the party to encounter this monster for the first time. Is this bad design?

Bad design? No, not at all. You are setting the scene with something that is going to happen.

If this is connected to another part of the scenario, then you might run into trouble if the NPC has already been encountered/killed/whatever, but for a first encounter with the NPC it is fine.

Quote from: Juisarian;836543I was intrigued by this post on 1d30 blog which seems to say you shouldn't do that sort of thing. Instead you should have some kind of routine or schedule for determining what's going on in a room when the PCs enter it. I'm not sure his "better" room is better than the "bad" example, which seems a lot more interesting to me. How would you approach this? The post later says "Place encounters where you feel appropriate" which I'm not sure how to interpret in this context.

I have seen scenarios where the GM rolls to see who is in each room when the PCs enter the room. This is very much a "random encounter" model and can make for an interesting scenario, but is too random for my tastes.

What I would do is to roll to see what the denizens are doing, if it is not important, but to have set scenes for important NPCs/Events.

Quote from: Juisarian;836543It's been said that a good dungeon should feel "alive" as if things are happening regardless of whether the PCs are there or not. That appeals to me, but it's ultimately a kind of an illusion; dungeons don't really exist when no-one's playing them. We don't plot the movement and activity of every individual creature in game time alongside the actions of the PCs, instead we have random encounter tables. At the end of the day every encounter only happens because the PCs are there to engage in it.

Using some method to decide if an encounter should or shouldn't happen now seems like potentially throwing away an opportunity to play the game, for no good reason. Or am I misunderstanding this?

No, you are not misunderstanding it, that pretty much reflects my views.

If you roll everything randomly, then you might not encounter an important NPC doing an important thing.

If you have the important NPC doing the important thing, then you can use this to drive part of the scenario.

Quote from: Juisarian;836544Originally I had a mini-scenario where the PCs encounter a group of itinerant priests on the outskirts of the dungeon. The priests were planning to delve further in, unaware of the awakened danger ahead. Saving the priests is a minor objective. The encounter happens "the first time the PCs enter this area" which is somewhat off the beaten track.

Sounds good to me.

Quote from: Juisarian;836544Thinking about what 1d30 wrote, I changed this to "when the PCs first reach the dungeon, the priests are three days travel away. Once they reach the outskirts they will rest for one day then, unless intercepted, enter the dungeon on the fourth day" plus there's a cool random table to determine what sort of gruesome fate they meet. It's more dynamic and "alive" now but is it better than just having them appear when the party needs them to?

So, rather than having a nice, simple event, you now have a conditional event that relies on the priests spending a day resting and entering the dungeon on the fourth day. How does this improve the scenario?

I can see the benefits of having a random table saying when the priests arrive/what state they are in, but not much.

Quote from: Juisarian;836544Note the trigger for the timer is "when the PCs first reach the dungeon" so it still comes back to the problem(?) of things only happening because there are players around. I could set the timer running without the PCs doing anything, and if they stay in town for a week the priests all die with the PCs having no opportunity to save them, but where's the fun in that? I think there needs to be a happy middle-ground between blatant author fiat and joyless existentialism, I'm just not sure yet where to find it.

Stick to your guns and have the priests appear when the PCs appear. Make it a bit random, if you like, but I cannot see the point of rolling to see what happened to the priests because the PCs were late, unless that was an integral part of the scenario.

The balance between as you say existentialism and author fiat is a very difficult one to navigate because every gamer has a different level of preference for this.
Personally I don't demand it be randomized, but I want an element of plausibility where what's going on doesn't seem to be a huge string of unbelievable coincidences. Suspending disbelief is a big part of what the game is about, IMHO, so I don't want it too contrived - the old cliche where you find someone who happens to be dying at an exactly appropriate time (and here's the treasure map) stretches disbelief for me and probably earns you an eyeroll. Same if you happen to stumble into the evil priests' lair at exactly the right moment to foil the ritual that destroys the world.
Other people of course just want to get on with the game and have an exciting story or whatever and aren't bothered at all. I guess a middle of the road approach will make the most players have. You can also arrange background events so that what seems to be an unlikely coincidence on the face of it actually does have some sort of in-world rationale.

Unrelatedly, the randomization of encounters element probably has another purpose as well, that being to keep the GM as surprised as the players, again how much a given GM will appreciate this is a YMMV thing.

I use two methods when designing scenarios. Sometimes I'll have a set piece arranged and it doesn't matter when the PCs arrive, it goes off as planned. Other scenarios there is a specific time for the event (with a small fudgability factor) and the PCs may miss it, arrive early or catch it in the middle.

One time at a convention I was running a CoC adventure and the PC detectives happened to learn about a planned sacrificial ritual taking place in a cave long before it was set to occur, maybe 6 hours before. When they arrived at the cave I had several cultists in mid-prep just sitting around eating lunch and listening to a baseball game on the radio. Needless to say the cultists were surprised and easily wiped out with no harm to the PCs.

Of course, there's no reason you can't change things up using broad strokes. For example: if it's day time the inhabitants of a dungeon are outside farming, fishing, gathering berries and wood, etc. At dusk they head back inside and in the early evening cook dinner. The PCs decide when they want to approach the dungeon, you decide what phase in the cycle events are.

When I write up campaigns I write up a timeline of events for when certain things happen, again with broad strokes and some fudgability. This helps keep the land "alive" and active despite what the PCs do. Let's say the PCs intend to assault an evil wizard's tower. I'll write in my notes something like: April 10, delegation from Evil Warlord leaves castle for wizard tower. April 25 delegation arrives at tower. May 3 delegation leaves with wizard's answer. May 19 delegation returns to castle. Maybe the PCs wipe out the wizard before the delegation even arrives. Maybe it's there when the PCs attack (in which case you'll have to decide if they participate in the defense or not.) Maybe the delegation has already left and will be returning home with a message from a soon-to-be-dead wizard. Maybe the PCs encounter them on the road.

Other events are static and it doesn't matter when PCs arrive. A family of bears are always in or near their cave. Joe the fisherman is always on the river. Not to say that Joe is in his boat 24 hours a day, but it doesn't matter what time of day the PCs first encounter him.

Quote from: Doughdee222;836578I use two methods when designing scenarios. Sometimes I'll have a set piece arranged and it doesn't matter when the PCs arrive, it goes off as planned. Other scenarios there is a specific time for the event (with a small fudgability factor) and the PCs may miss it, arrive early or catch it in the middle.

One time at a convention I was running a CoC adventure and the PC detectives happened to learn about a planned sacrificial ritual taking place in a cave long before it was set to occur, maybe 6 hours before. When they arrived at the cave I had several cultists in mid-prep just sitting around eating lunch and listening to a baseball game on the radio. Needless to say the cultists were surprised and easily wiped out with no harm to the PCs.

Of course, there's no reason you can't change things up using broad strokes. For example: if it's day time the inhabitants of a dungeon are outside farming, fishing, gathering berries and wood, etc. At dusk they head back inside and in the early evening cook dinner. The PCs decide when they want to approach the dungeon, you decide what phase in the cycle events are.

When I write up campaigns I write up a timeline of events for when certain things happen, again with broad strokes and some fudgability. This helps keep the land "alive" and active despite what the PCs do. Let's say the PCs intend to assault an evil wizard's tower. I'll write in my notes something like: April 10, delegation from Evil Warlord leaves castle for wizard tower. April 25 delegation arrives at tower. May 3 delegation leaves with wizard's answer. May 19 delegation returns to castle. Maybe the PCs wipe out the wizard before the delegation even arrives. Maybe it's there when the PCs attack (in which case you'll have to decide if they participate in the defense or not.) Maybe the delegation has already left and will be returning home with a message from a soon-to-be-dead wizard. Maybe the PCs encounter them on the road.

This:). It's not random, you can guess the pattern, but what the PCs choose about the timing of their actions actually matters.

QuoteOther events are static and it doesn't matter when PCs arrive. A family of bears are always in or near their cave. Joe the fisherman is always on the river. Not to say that Joe is in his boat 24 hours a day, but it doesn't matter what time of day the PCs first encounter him.

"Go to the river, meet Joe."
"Even if we went there at night in order to ahem, borrow his boat for a law-sensitive operation?"
"It's what it says in the scenario, you get to meet him".

And that only serves to remind me why I avoid scenarios;).

Thanks for the feedback so far, lots of good ideas. I think people are saying it's best to use a variety of approaches, depending on what works for the encounter and how it relates to the adventure as a whole.

Small things work better if left to play themselves out but key stuff should be placed where PCs will find it (unless they go off on a complete tangent, but that's part of the game). Encounters that benefit from running on a timeline should do so. To use my foolhardy priests example, if the NPCs just wait around in limbo for the PCs to find them it becomes a virtual auto-win, but if they have their own course of actions the risk of missing out on the reward for saving them seems more meaningful. They key will be to ensure the PCs have enough motivation and knowledge to take up the task.

Quote from: soltakss;836547Honestly? Read good scenarios, see what they did and use the same techniques.

Any you would recommend? I'm keen to figure out how to present once-off set pieces to the GM in a way that makes sense and helps them run the game.

Quote from: Bloody Stupid Johnson;836552I want an element of plausibility where what's going on doesn't seem to be a huge string of unbelievable coincidences. Suspending disbelief is a big part of what the game is about, IMHO, so I don't want it too contrived  ...  You can also arrange background events so that what seems to be an unlikely coincidence on the face of it actually does have some sort of in-world rationale.

Sounds good, I will work on this

Quote from: Doughdee222;836578One time at a convention I was running a CoC adventure and the PC detectives happened to learn about a planned sacrificial ritual taking place in a cave long before it was set to occur, maybe 6 hours before. When they arrived at the cave I had several cultists in mid-prep just sitting around eating lunch and listening to a baseball game on the radio. Needless to say the cultists were surprised and easily wiped out with no harm to the PCs.

Hopefully I can come up with a way to write it so a half-decent GM can run with ideas like that.

Quote from: Juisarian;836761Any you would recommend? I'm keen to figure out how to present once-off set pieces to the GM in a way that makes sense and helps them run the game.

Scenarios are very dependent on the system used.

What system were you going to write for?

If it is BRP/Legend/RuneQuest/D100ish, then I would probably make some recommendations, but probably not for anything else.

Quote from: soltakss;836781Scenarios are very dependent on the system used.

What system were you going to write for?

If it is BRP/Legend/RuneQuest/D100ish, then I would probably make some recommendations, but probably not for anything else.

I'm using D&D style rules but anything that works is good. For the purposes of understanding good adventure writing I don't think system, setting or genre makes a big difference.

Second problem: Does my town need a map?

At the moment the town outside the dungeon is just a list of useful NPCs, their personalities and functions. It can be easily replaced with the GM's preferred fantasy village. I made it up for the game I ran so I figured I would include it as potentially useful content.

There's no indication which direction the main street runs, where the well is, who lives next door to the blacksmith or things like that. As a consumer of RPG modules, would you feel let down by this lack of detail?

I wouldn't say its strictly needed or that I'd be disappointed. It might be useful, if it impinges on the adventure or as a way of determining what random NPCs they're liable to run into. I haven't used a module in a while so I'm not sure how much of an omission it is, competition wise.

If you plan to publish an AD&D adventure, consider your system choice and how you will promote the final product. AKA, are you going to make it OSRIC, Basic Fantasy, LL, S&W or go generalized and call it "for Old School Fantasy Roleplaying" or whatever. You probably want to ask these questions on Dragonsfoot since many OSR publishers and OSR customers are there.

My basic feedback for you:

1) Include lots of options and choices for players and DMs.
2) Create a living environment, not a static one.
3) Playtest the hell out of it before publishing.

Quote from: Juisarian;836543We don't plot the movement and activity of every individual creature in game time alongside the actions of the PCs

Actually, I do that all the time. It's a very effective for the smaller areas that make up 95% of scenario locations. I generally only step up to using random encounters to model activity in a complex when the complex becomes too large to manage.

More generally, separating mobile elements (i.e. creatures) from static elements, as the 1d30 blog suggests, makes a lot of sense because it makes it much easier to manage and use the dynamic elements of your adventure.

If the PCs get into a firefight in the dining room, for example, it doesn't make a lot of sense for the security standing in the next room to just ignore that. While it's certainly possible to look at the descriptions for every room near the dining room and then cross out anyone who you decide will respond to the firefight, it's A LOT easier to have an independent roster that you can quickly reference and use with a single glance.

The other thing to note is that the 1d30 blog is also talking about running a mega-dungeon. Now, as I note above, using random encounters to simulate activity in a large complex like a mega-dungeon makes a lot of sense. But it still makes sense to separate active creatures from static elements. For one thing, as I talk about in Re-Running the Megadungeon, a key aspect of running a megadungeon is that it changes and evolves and responds over time. In an effective mega-dungeon, the monsters don't just sit passively in their lairs waiting for the PCs to come back and slaughter them.

It's much easier to move the creatures around or repopulate dungeon complexes if you can (at least mostly) leave the static key alone and just focus on independent, easy to reference and easy to use rosters.

The other thing about set pieces in the sense you seem to be using the term (a specific character in a specific location doing a specific thing) is that they're a form of very specific contingency planning. I'm not a fan of contigency planning. You can some times get away with it in very specific circumstances, but they tend to get messed up if the PCs do something unexpected (like getting into a firefight in the dining room, or entering through a window instead of a door, or finding unexpected use for that C-4 they looted a couple of sessions ago).

And, more generally, I find that the "set pieces" that evolve naturally and organically out of the actions taken by my players are virtually always more memorable than the stuff that I pour sweat, blood, and tears trying to force into the session.

For example, one of the most memorable "set piece"-like moments in my D&D campaign was when a PC fighter leapt out of a window and the giant, worm-like monster pursuing them smashed through the wall behind them as they both plunged to the ground below. Incredibly cool moment. Completely impossible to force into play.

And what tends to happen when you focus your energy and attention on making sure that your prepared contingencies happen is that you miss the opportunities to make giant worms crash through walls in a cascade of shattered mortar.

Now, the fact that you're publishing the adventure requires this to be couched in a caveat: With a published scenario you need to also communicate your creative vision for the scenario to the GM reading the scenario. (Which you obviously don't need to do when prepping for your home table.)

I recommend sidebars: Any time you want to tell the GM reading your scenario about a really cool way you think the various components of the scenario could be used, put it in a sidebar. This lets the main text of the scenario focus on all the tools the GM needs to run the scenario, while still allowing you to communicate cool ideas to them.

Quote from: Justin Alexander;837322If the PCs get into a firefight in the dining room, for example, it doesn't make a lot of sense for the security standing in the next room to just ignore that. While it's certainly possible to look at the descriptions for every room near the dining room and then cross out anyone who you decide will respond to the firefight, it's A LOT easier to have an independent roster that you can quickly reference and use with a single glance.

"Order of battle" stuff as it's sometimes called seems pretty easy to document, e.g. "the guards in room 16 will responds to any disturbance [here] in 2 rounds"

Would you go as far as putting the room descriptions and NPCs/monsters in separate sections of the document? I can see the advantage of this but it also breaks up the relevance of the content and likely requires page flipping and cross referencing while trying to run the adventure.

Quote from: Justin AlexanderIt's much easier to move the creatures around or repopulate dungeon complexes if you can (at least mostly) leave the static key alone and just focus on independent, easy to reference and easy to use rosters.

Some areas in my dungeon are intended to be repopulatable. I've put some brief notes in the descriptions as to how this will work. If I had a lot of these rooms I'd probably want to use a "population table" or bullet point list to make this easy to understand. Wandering monsters are a big element of my dungeon so I hope it will be fairly dynamic even if a lot of rooms get cleared out.

Quote from: Justin AlexanderNow, the fact that you're publishing the adventure requires this to be couched in a caveat: With a published scenario you need to also communicate your creative vision for the scenario to the GM reading the scenario. (Which you obviously don't need to do when prepping for your home table.)

That's the issue I'm facing. It's easy to run with stuff that comes up in a homebrewed adventure because I already have an innate understanding or how it all fits together and what should and shouldn't happen.

Quote from: Justin AlexanderI recommend sidebars: Any time you want to tell the GM reading your scenario about a really cool way you think the various components of the scenario could be used, put it in a sidebar. This lets the main text of the scenario focus on all the tools the GM needs to run the scenario, while still allowing you to communicate cool ideas to them.

Sounds like a reasonable approach. By the way, I remember reading through your site a while back. It's got a lot of good content, thanks for taking the time to put it out there.

Quote from: Justin Alexander;837322For example, one of the most memorable "set piece"-like moments in my D&D campaign was when a PC fighter leapt out of a window and the giant, worm-like monster pursuing them smashed through the wall behind them as they both plunged to the ground below. Incredibly cool moment. Completely impossible to force into play.

Stuff like that is why I play RPGs. As an aside, when I first ran this dungeon, the players set up an ambush near the entrance, bypassed room 1 completely, killed the guys in room 2, then concluded they'd killed the BBEG and headed back to town. They spent a week there being feted as heroes and getting involved in all sorts of stupid shit. When the monsters attacked again the villagers ran the PCs out town and basically forced them to go finish the job if they ever wanted to show their faces in civilisation again. It wasn't something I could have ever planned but it all made sense and flowed naturally in the context of the game.