PDA

View Full Version : Advice for a first time sandbox gm



Elvenoutrider
2011-07-26, 11:59 AM
Playgrounders. This fall I will be running a sandbox 3.5 game involving a boatload of homebrew. The campaign setting is a post apocalypse setting involving modern technology made from magic. He apocalypse came about when the material plane denizen's advances in technology made belief in religion irrelevant, causing the gods to lose power until he'll finally won. The mortals managed to seal the plane from hell but it was too late to stop the fall of civilization.

So far I have prepared a few factions, a few friendly and unfriendly npcs, and a few encounters.

I am wondering if any o my fellow playgrounders who have ever run a sandbox have any advice for how to run one, or anything else I should create in advance. Any comments on the setting are also appreciated. Thanks in advance

Traab
2011-07-26, 12:18 PM
Create interchangeable storylines. Basically what I mean by that is, if they go with faction A, then they will warn how faction B is out to take over the world. If they went to faction B, then its faction A that wants to take over the world. That sort of thing.

Include several sets of ruins or other dungeon type setups in every direction. All designed the same way to start, with the only variance being things like, in a desert area you might face mummies, while in a jungle ruin you might face more wild animals or whatever. Then when they pick a dungeon to explore, you will have time to change the layout of the other ones before they finish and move on. That way it doesnt matter if they head north south east or west, you will have a dungeon prepared for them, without having to design 12 unique dungeons right off the bat just in case.

Each faction should have similar quest triggers. The players have no way to be sure you havent created unique quest lines for all possibilities, so it doesnt matter that both factions A and B will send you to the same places for the same things, just with slight flavor variations in what they say. This saves you a great deal of effort by not forcing you to create unique storylines for everything all at once.

hydroplatypus
2011-07-26, 01:47 PM
In addition to interchangeable storyline quests, also use interchangeable locations. Basically design 2 or 3 locations such as a village, a town, and a city that have no actual location in the world. (this is in addition to the other fixed locations you have). This allows you to use these whenever the PCs get really far away from what you planned. They will never know that you didn't plan for it to be there. Additionally have 1 or 2 quests like this, and MANY NPCs like this. I call this the Schrodinger DM style.:smallbiggrin:

KineticDiplomat
2011-07-26, 02:00 PM
Factions, motivations, goals, relationships, and reasons why they might need adventurers. These are the bones of a sandbox. In straight plot, you carefully and lovingly sculpt a handful of key factions, characters etc to make them each as awesome/memorable/specifically challening and sufficiently epic as you can make them.

Sandbox does not work that way. Instead, what you really need is a network link diagram for your sandbox. the actual diagram is optional but the basic concept is not: These are the powers "in play". Here's who they like, here's who they don't, here's their resources, and heres what they want.

From there, I would reccommend that you modify the TACTICAL/OPERATIONAL plots, schemes, and effects of the factions being directly effected by the PCs mildly per session based on their influence.

Roughly every major event or season, advance the state of the world 1 "turn" if you will. What did A do to B, what did B do to C, etc. Whats the new layout? How does this create new relationships and new actions.

Traab
2011-07-26, 02:03 PM
The best part about doing things this way, is that you dont drive yourself insane trying to cover every possibility at once with something unique. It gives you time to set things up ahead of the group. And it also works as a cover for anything you might have missed. For example, you give them a small town just so they can resupply, sell off any treasure, buy new items, whatever. There is nothing else to do there, its just so they can open up some bag space after that alst dungeon crawl they did. Somehow they get it into their thick heads that there must be SOMETHING to do here. So instead of cursing quietly and trying to ad lib an event, you can just pull out the folder titled "Random Town Event #001" that would work in any location from a group of mud huts to camelot itself. While they are busy solving random town event #001, you can be thinking up random town event #002 for the next time they try something like this.

Yora
2011-07-26, 02:05 PM
The players should decide on a rough overall goal before making characters. "you can do anything you want" does not work well to start a campaign. "We want to be pirates" or "we want to bring down an evil kingdom" proovide some general direction the game will take.

HyperionWolf
2011-07-26, 03:14 PM
Also, you could make sure the World is everchanging.

If they are Pirates, make sure that once they steal everything from a squad of merchant ships, the next ones will bring the brand-new "S.S Übercannon" along.

In a Sandbox game, you need to make sure they keep entertained, keep things new. :smalltongue:

Ekul
2011-07-26, 03:41 PM
Make your own random encounter table, and have an entire section of it related to NPCs that could kick off some sort of event. Whenever I GM, I'm no good at predicting what people will do, so I normally sandbox it. I let the players decide the setting, and their goal, and then I create it for them.

One of the advantages is that your entire campaign can't get screwed if the wrong combination of events happen. I have a player who REFUSES flat out to be railroaded (He's also usually our DM, and he's very good at forcing us to do what he says without railroading, because if we don't complete the plot, we can't go to the towns to get/sell/upgrade items anymore because they'll get destroyed. There's no real point to derailing his campaigns and we don't ever have chaotic stupid players.) I have adapted to him, because every other player will do what they're told, and I don't try plots anymore, just let it unfold. Especially if the players are creative in their motivations.

Not everybody is creative in their motivations, though. Avoid putting these people in positions of power in your sandbox. They will sit around, do nothing, and complain that nobody else is doing anything.

Xyk
2011-07-26, 08:52 PM
Have a looooong list of names at your disposal. There are so many people in a sandbox world, and your players will often ask for names.

John Campbell
2011-07-27, 01:21 AM
The common failure mode for sandbox games is the players sitting around staring at each other going, "What do you want to do?" "I dunno, what do you want to do?"

The way to avoid this is to make it a living world. Don't just draw a map and put a bunch of encounters on it and then wonder why your players aren't going to any of those places and having the encounters. Have things going on that the PCs aren't driving, things that they can affect if they choose to engage with them, and that will affect them even if they don't. Make them react to things. They don't have to be big things; just relevant enough to them that they'll notice them. You don't have to have a preconceived plot for how they'll react; just poke them, and whatever way they jump, run with it.

And figure out how those things will play out if the PCs don't get involved - because oftentimes they won't. And even when they do, it may not be in the way you'd expect. Changing a series of events that assumes that the PCs aren't doing anything to reflect their actions is a lot easier than changing one that assumes that they were doing something completely different than they actually did.

And have lots of these things, not necessarily major, not necessarily related, but lots. You can never tell what's going to catch the PCs' interest, and which hooks they're going to blithely ignore.

Be good at improvisation. I don't know how to help with that if you haven't already got it. But take notes. When you make something up on the fly because your players did something you didn't expect and weren't prepared for, write it down.

And the names. I need to start doing that; that's what always trips me up. I've been known to come up with hobbies, history, and motivations on the fly for the busboy at a Denny's that the PCs exchanged about a dozen words with and never saw again, but ask me for an NPC's name and I choke.

Altair_the_Vexed
2011-07-27, 08:26 AM
Something I like to do once I've thought up my schemes and plans of the various factions in the setting, is run the NPCs plans in the background while the PCs deal with whatever they feel like getting their teeth into first.

This means that - for example - while the PCs are exploring Desertia, the Inquisition of the Holy Reich is persecuting the Monks of Enlightenment, the King's increasing feebleness means that the Grand Visier is now running the kingdom and raising taxes to line his pockets and bribe the barons to recognise his nephew as heir, and the Guild of Smugglers is moving into Capital City to exploit the fact that the Watch are now underfunded.

Or if the PCs are busy dealing with the Inquisition, then the Tomb-lords of Desertia are rising up and gathering an army of undead, and the Grand Visier, Smugglers and so on continue as perviously.

This means you need to have an idea of what each plot will do if it is left unchecked.
Effectively, this gives the PC party something to deal with after they've persued their preferred goal - and that something is now more powerful and well-established, because it got left alone for a while.

Of course, I'm currently utterly burnt out and waiting to get my gumption back, so I may not be giving the best advice.

EDIT: Ha! Ninja'd by Mr Campbell himself.

LibraryOgre
2011-07-27, 09:57 AM
My trick for a sandbox game is to know what would happen if the PCs aren't there. Have multiple factions with multiple plans going, and know what's going to happen if the PCs decide "To **** with it, we're going to Undermountain."

(HATE THEM SO MUCH!!!!!)

If the PCs don't engage with the plot, let it continue apace. Let the world change around them, either providing them opportunities for adventure or changing the world in which they adventure. As they engage with the plot, figure out what's likely to happen to other people's plans. If they wind up wiping out a sizable number of the drow-turned-lythari, how does that change the plans of the other drow in Cormanthor? How does the lack of that pressure change what's going on in the hobgoblin domain of Ssessrendale? How does the increase in power of the hobgoblins change what happens to the Returned elven kingdom, or the Dale's council, or Zhentil Keep's plans in its war with Hillsafar?

Hoddypeak
2011-07-27, 10:44 AM
To take this in a different direction from other people's advice and give you a different perspective let's look at what the 1e DMG has to say.


The milieu for initial adventures should be kept to a size commensurate with the needs of campaign participants - your available time as compared with the demands of the players. This will typically result in your giving them a brief background, placing them in a settlement, and stating that they should prepare themselves to find and explore the dungeon/ruin they know is nearby.

To paraphrase, he goes on to suggest start small, give the players only as much background as is needed for the first session, and build out from there, adding on as need demands. This is my preference for sandbox games. This does require staying a few steps ahead of where your players are, and good improv abilities. Then, as you have ideas and write them down, you can introduce them as appropriate.


The common failure mode for sandbox games is the players sitting around staring at each other going, "What do you want to do?" "I dunno, what do you want to do?"
This is especially true at the very start of the game, when they don't really have any goal of their own at all. I don't think starting them on the rails for about the first 5 minutes, "go find and explore the Caves of Chaos," is a bad idea, then just let things flow from there.

I likewise agree with those above me who say keep an idea of what factions are around and what their goals are. If you know the goals of the different factions, that's really all you need, as you can determine how they'll respond to offers of aid or attempts to interfere based on what your players do. Pre-planning storylines is, in my opinion, folly.

Frozen_Feet
2011-07-27, 11:38 AM
As I've been running a sandbox campaign for the last year, here are some things that worked for me (and maybe some things that didn't).

A) Have a map. A large map. It doesn't need to be detailed, by having a rough idea where your players will go if they decide to hike to the other end of the continent is handy. You can add details as the game goes on. Put some key locations on it (like capitals of a kingdom) to draw attention of your players.

B) Show the map to your players. It will give them ideas of what to do.

C) Make a daily random encounter table or several for different environments. The most common result should be "boredom". This might sound boring, but it allows for glossing over some of the travel time. The encounters shouldn't be just monsters, but things like unmapped islands, small settlements, ruins, travelling tribes etc. You can mark down (or make your players mark down) the location of such events, filling up the map and allowing the players to return some time if they want to.

Addendum: realize that this means random encounters are not tangential to the plot; they are the plot. If the player run afoul of something overwhelming, don't bail them out! Let them or make them realize that everything they find or meet can be significant.

D) Keep track of travel time, rations and weather. If your players want to go somewhere, make them work for it. Never gloss over these just because your players need to be somewhere; in fact, scrap any notion of your players needing to be anywhere! Don't make them be everywhere just when something interesting is going to happen. If they're going to be early or late, let them; allow it to flow naturally from their own choices and decisions.

E) Improvize and take notes of what you do during play, as opposed to preplanning. The wider the arena you let your players wander in, the harder it becomes to anticipate what they're going to do or where they're going to do. So don't bother trying to pre-plan a story - let the story write itself during the course of the game.

F) That said, if you have some grand plot for the setting in mind... let it resolve itself as would be logical if your players keep ignoring it. Flat out tell them that they're not going to be heroes or protagonists of the big story if they themselves don't work towards it.

A computer game by Spiderweb Software called Exile 3: The Ruined World is a good example of how this works: if the player keeps dawdling and not saving the world, all the cities will fall to ruin one by one and be infested with monsters. Plot important NPCs will not die, but everyone else will. Sidequests and shops become inaccessible.

Take that idea and run with it. This way, your big plot can provide framework for the PCs even if they miss all the clues and go astray.

Elvenoutrider
2011-07-28, 09:07 PM
Thanks for all the advice, I dont suppose any of you have any recommendations for random encounter tables that have worked well for you

Immonen
2011-07-28, 10:29 PM
Why yes, yes I do.

http://kroncke.webs.com/forelvenoutrider.htm

This took me about a week to type up, and it has served me well. Not very campaign specific, either, but you may have to edit to your own taste.

Raum
2011-07-29, 11:24 AM
Playgrounders. This fall I will be running a sandbox 3.5 game involving a boatload of homebrew. First, decide what you mean by a sandbox. My definition includes: Player character choices matter. They change or guide the game in a material manner.
Player character choice is unfettered within the context or scope of the game's genre, world, and situation.So a choice between two (or more) prebuilt stories isn't a sandbox, it's a choice of scripts. And the illusion of choice where the bad guys are always the other team is still a script.


I am wondering if any o my fellow playgrounders who have ever run a sandbox have any advice for how to run one, or anything else I should create in advance. Any comments on the setting are also appreciated. Thanks in advanceWhen it comes to NPCs, include three items in their creation: Goals. What do the NPCs want? Power? Vengeance? Or just food on the table?
Resources. This covers the NPC's abilities and limitations. If he's rich he'll use money to try and reach his goals. If he has the Sword of Demonreach, that's what he'll use.
Some idea of how the PCs are likely to learn of, run into, or otherwise interact with the NPC. Just notes really...and likely to change as you flesh things out. It could be as simple as "bartender at the Lucky Dog tavern" or as complex as following up on a series of bribes made to get a town council's vote. Either way, don't get too detailed - just a few notes.

I wouldn't even stat NPCs unless I think the PCs are getting close to running into them. Up to that point, the three pieces of information above are enough.

Some general advice: Get the players involved. Pissing off their characters is probably the easiest way to do so...but certainly not the only way.
Have some minions (from one of those NPCs with an inimical goal) going after something the PCs have or are simply near ready to go if players are at a loss.
Leave lots of clues. Three times as many as you think you'll need is a basic rule.
Avoid planning things player character choices can reasonably change or avoid.