I take a collaborative approach. I ask the players to tell me about the world or area they are playing in and we explore the things they bring up. That way I don't spend time creating things that don't interest them and will never impact them.

1. An initial hook or direction. The players should have a clear idea of what their long term goal is, or what kind of long term goals are available, even if they have to come up with "how" on their own.

For a game set in an old west town, for example, maybe they might explore the surrounding terrain. Or improve relationships with the local natives. Or conquer the local natives. Or get rich. Or become the best known gunslinger in the west. Or get elected sheriff and clean up the corruption. Or serve the local oil baron by corrupting the town.

Give people some ideas.

2. Interesting things to discover while pursuing their goals. Seed them with cool stuff.

The best I've seen for sandbox is similar to the way console and computer RPG's do it. Have a bunch of quests, some major, some minor, but allow the players to choose what they want to attempt.

It means more work for you initially, since you have to create story lines that won't all, or even mostly be used. But once the players choose you let them dictate the story. Put some hooks into the story for future possibilities, but let their actions dictate the direction.

After the initial few quest lines, I find sandbox to be the easiest to run, since you are (mostly) just reacting to the players decisions. It also get's rid of most of the problem of the PC's throwing wrenches into your plots. If the plot is dictated by the story the players are telling, they really can't ruin the story by their actions.

finding ways to not have them killed while they play was important for me. different approaches to how to get them 'punished' without death that make sense. this means they can explore and run into difficulties, even get in over their head if they make a mistake, and not have to die and start over (or quit). some examples are selling the character into slavery, or to someone that wants them for some other nefarious purpose (such as a doctor making strange experiments on unwilling subjects, etc.) these get them in trouble if they stray without stopping their game altogether, and lead to other escape-based adventures. otherwise, the gm above who posted that you would have to make stuff that gives them choices, many of which they wont take, is spot on. I would suggest having a number of adventures of various levels, and then let the characters gather rumors or info and then decide what they want to pursue. you let them do this and be ready with non-lethal consequences (good and bad) for their choices. And that is what I think is the essence of submersion for a person running a character (i.e. fun): personal choice and realistic consequences for them.

That's good general advice, too, for non-sandbox games.

I can't recommend slavery as I find that players hate the idea of capture as much as death, if not more. And, come to that, certain other plausible ways for them to fail (deaths of NPCs, for instance) might also be frustrating for the players. But most people have ways they're willing to fail, so a GM should ask them about those ways and focus on those.

An initial adventure certainly helps to kick-start things. Could be as simple as the archetypal 'tavern scene' being subject to a police raid. Will the PCs run or 'come quietly'? Whatever they do will land them in a situation they need to find a way out of.

Making sure they have well-developed characters helps, too, so they have goals and purposes that will motivate them.

The killer for most sandboxes is lack of motivation. If the players don't have motivation built into their character background, or built into the initial scenes, they will just drift around and get bored. One particular danger for sandboxes is lack of GM motivation. If the GM doesn't have a story to tell, and is dependent on the players telling a story, and the players are just drifting around, the whole thing can go down the pan quickly.

I think a series of short railroads can help. If the PCs are drifting, have a street gang assault them, or let them find someone's wallet, something to drag them along for a while and then leave the mini-adventure open-ended, so it can be resolved in different ways.

An over-arching theme can be useful too - the PCs are wanted for a crime they didn't commit, a member of the team went missing without trace, there are alien 'body-snatchers' hiding amongst the population, etc. Just be sure not to pin your theme to a single PC - Fred's sister has been kidnapped - cos you can guarantee that Fred's player will drop out of the game in the first three weeks!
This message was last edited by the player at 05:13, Fri 09 Mar 2018.

this ties into the idea of immersion for the characters via freedom of choice/realistic consequence, I believe. the more choices a character has that carry noticeable and believable consequences, the more they may invest in their character, and then in my experience, the more they are able to come up with their own motivations.

I think the above poster is correct in that the gm has to have conflict written into the way things work in the world. this  gives characters something to choose, even if its just to stay out of trouble (which may or may not be easy). I prefer not to saddle the characters with too much forced encounters, although sometimes a raid on the tavern by the local militia is a good kick-starter for fun :)

on rpol when I ran my game for five or so years I learned the hard way that even more than with table top gaming, the gm needs to leave every scheduled posting with at least on decision for the group to make. it doesn't have to be anything more important than which inn to sleep at, or where to stable the horses, or what to eat for dinner, where to hide to pick pocket people in the bazaar, etc.  it takes a little bit of extra effort when you post as a gm with this mind, but it helps tremendously in keeping a flow to a sandbox game.

I think sandbox games can be some of the best, but only if the gm is presenting a stable and consistent piece of the world for the characters to play in. building conflict into every part of your gaming world is then paramount. and again, the conflict is anything from who to go to war with, to which alley to hide in from the militia when they raid the tavern.

Would it help to clarify the difference between sandbox and open world? There might be some overlap.
I figure sandbox puts more emphasis player driven, whereas open world is more about the size of where players can go.

Could try to emphasize to incoming players that it is a sandbox game, and that their character should be built accordingly (motivations to do their own things), and they may need to play accordingly (more self driven). Because some people might just see “sandbox” and not think that much about it.
You might find some players are more leaders and others more followers. Could try to get them grouped up accordingly (at least one leader per group).

This is sort of advice for any game, but it would help to keep the size of the game and cast reasonable.
I’ve fallen into the trap of “I’ve got so many players, awesome!” and then realizing I couldn’t handle them all.
Could bring on co-gm’s or helpful players to mitigate the load.

Though it can be tough, you might try to keep all the players in the same general area (like a town or city). Groups are more likely to affect each other and cross paths. If a player ends up alone because of leavers, they can more easily join another group.
Getting plotlines to all entwine to some degree could be nice too.
It might be more of a contained sandbox.

Something I have wanted to try is, tell the players “you make your own hooks/motivations, but they should tie into the plotlines the GM has set up.” That way if players are following the hook of one player, but that player leaves, it can salvaged more easily. Probably helps keep everything together too.

Co-GM's are a bit of a sticky wicket. When using a Co-GM you should do so from the beginning. I've seen it done well, and I've seen it done terribly. Both need to either be already telling the same story, or one needs to be only dealing with one aspect of the game.

I'm currently a Co-GM in a game for a friend of mine, and my sole areas of responsibility is editing the posts. with a small amount of help regarding rules, though I don't make the rulings. I simply edit his and the player's posts for readability and grammar. (My friend has speech issues and asked for that help.)

I've seen games with Co-GM's that were able to work well, and other's that died from GM infighting/confusion. The ones that worked well were those that followed the rules above. I believe that the key to a success in that situation is the same as in most. Communication.

Good point. The more emphasis, the better, though it doesn't always work. I've had games where I've driven home the concept of self-motivation, and the players have still drifted. I think it's largely pot-luck on getting the right bunch of players together. There may be nothing wrong with any of them individually, but they just don't spark off each other very well.


Definitely. I've been in a game where there was a GM and two co-GMs, the GM left and the two co-GMs continued running the game, but it only worked because everyone was on the same page to start with, and stayed on the same page by communicating their ideas at every opportunity.

I'ven been thinking about running a Night City Cyberpunk game and it feels like open world, rather than sandbox, would be a better approach.

• I'd like to have a background campaign that intrudes into their world from time to time, but otherwise provide the characters with flexibility. That's my defintion of open world (and how it differs from sandbox) but happy to get feedback. I'd like to ensure that players understood what I was expecting from them up-front and would hate to use the wrong branding.
• Cyberpunk is a very gig-based game out of the box with a Fixer acting as the adventure noticeboard so that seems to work beautifully.
• I was also intent on it being low-level, almost gang-in-the-hood level, rather than taking on big corps - at least to get started. Just treating the corps as the big boogeyman and working collaboratively with the players to flesh out the neighbourhood in some detail with their background and hooks. Feels like session 0 could last a while :)

Does anyone have any specific thoughts and feedback on making open-world work for this genre and this level of game?

I think a pretty standard and easy way to introduce a party to an open world is you start them on tracks with them meeting up with the fixer to be given their first job as a team. This accomplishes a few things:
Establishing the fixer as their source of jobs.
Introduces them to some of the world.
Sets them up for your style of gameplay.

Your goal with this first job is to make it relatively easy - not saying it's not a challenge but the threat level and reward aren't high. The reason being the players (ideally) won't be super paranoid/on high alert. This gives them room to breathe, explore, learn and most importantly meet people. Ideally the job should bring them in contact with several people and places, some of which can be of use in the future. So an investigation job is a good way to do that. It inherently involves asking questions, visiting different locations. You need to flesh out their 'home turf' with this quest.

This investigation is a good way to spring-board into a second job while in the meant time you can be scattering the idea of other things they could be doing.

So as an example:

First job: One of the fixer's people disappeared mysteriously in the neighbourhood. Here's some info on what they were doing. The goal is to find out what happened to them and recover something they were carrying. You might start by talking to x person over at y place.

The job will pretty much be a daisy chain, person x can point them to a couple of other places they can go to continue the investigation, the next places points them to another couple of places and so on. Ideally you'll sprinkle enough info into each person or place that say you've got 7 or 8 people/places of interests and once they've gone to maybe half of them they can start piecing together what happened and where.

Maybe this leads them to an alley where this person died which puts them in conflict with a small time gang who 'owns' the alleys here. A nice intro to some conflict resolution and gives a small time enemy who can be a recurring threat.

This gang might have found the body and looted it but weren't the ones who killed him. At this point this should accomplish the fixer's goals: they found out what happened to his agent, and they recovered the item. The job's not done yet though which can easily lead into the second job, accomplish what this other agent couldn't maybe this item needs to be delivered somewhere dangerous or who knows.

But either way the players now should have a decent idea of the neighbourhood and might have a few little interesting bits of info about the people in the neighbourhood they can use in the future, or that you can use as story hooks.

Remember that the goal of open games is not to just sit back and say it's up to the players what to do. Well it is in a sense, but you need to scatter lots of little breadcrumbs and then see which ones they pick up.

Honestly, I'm not sure what the difference is between open world and sandbox.

I don't think there is any real difference, if I had to create differences the way I would split it up is open world follows a fairly traditional format of here's a quest for you to do but if you decide to go over there and try to do something then the DM isn't going to say 'but the module doesn't include that'. While sandbox is more scattering a bunch of info and potential plot hooks around and letting the players discover them and decide on which ones to act upon.

Yes, this got me as well. A quick look through some searches suggests that lots of people can't agree either but the consensus suggests that a game can be either of them or both.

I grabbed one post that seemed as coherent as any other explanation.



Feels like:

• Sandbox allows for far more creativity in what a player wants to do. Players may have access to more unlimted resources and they make their own fun, creating their own NPCs, their own locations, their own stories,and so on. Theoretically nothing is off limits.
• Whereas open world allows you to go anywhere (within whatever borders have been defined) but with the world defined to some extent by a GM?

Story that is player driven vs GM driven is what that post is saying.

But since we don't agree and I doubt anyone can cite early sources of "games without plot" as an authority on the usage they are going to remain interchangeable.

Technically EVERY game is a "sandbox" by their definition of your GM can think on his feet well enough and the players are obstinate enough.

In reply to Centauri (msg # 6):

Could you expand on this please?

At the moment I'm slowly working on what will be my first game on RPOL, and I'm worried, just a little that the questions I'm going to be asking as part of RTJ is a little too much for sighted folks.

Nah, it's probably not really what you're after. I let my players do what they will and go where they want, but I don't plan out a "living world" as seems to be the sandbox ideal.

In reply to Sightless314 (msg # 18):

Here are some examples of alternate fail abilities.
Injuries: One or more of the PC's stats are lowered for a lingering injury until a quest of some sort is taken to heal the injury.

Capture: The PC's wake up in chains set to become sacrifices/slaves/ pawns for a BBEG. This leads to them having to plot the escape and figure out how to stay alive with reduced resources. (Just don't make it too difficult, or easy.)

Robbed: They survive, but find they've been stripped of most if not all equipment. (Hard to gauge since they might need some to survive. Maybe they lost any held items, backpacks etc... but nothing that was hidden or hard to remove (armor). Note: Don't take a wizards spellbook or a clerics holy symbol, as without them they are practically worthless.

Jailed: Only really works in a lawful area. (City/town/kingdom.) I use it when players become murderhobos or get out of line towards NPCs. They wake up in prison, or in a jailer's wagon on the way to the prison. Once there they have to either figure out how to escape (and here I do make it hard) or figure out some way to fix whatever they did. This could be by completing some very hard task, usually under supervision of a powerful NPC, or through a Geas. Or it might be by serving time or being banished from the area. Being banished could be very difficult since you could have a quest objective in the area they are banished from. And if discovered they will be attacked on sight. One option I have used is to have them conscripted into a war that's brewing. They serve and survive the war they earn their freedom. This sets up all kinds of possible scenarios. Do they fight hard and honorably? Do they run the first chance they get? Do they desert to the other faction? Espionage, betrayal, incompetent leadership, all are possible.

There are others but this gives you a start. I encourage you to discuss these options with your players, as some groups don't want certain options on the table. Better to discuss options ahead of time.

There are also ways to fail which do not directly do harm to the party.

For example the BBEG is far too powerful and arrogant to dirty his hands killing or even capturing these pathetic things that tried to stop him. What better than to leave them battered and beaten on the floor so close to success while he just packs up and goes to another hideout. They will wake up amongst the leftovers of his master plan, oh the humiliation and seething hatred they will feel is just too delicious.

Somewhat hand in hand there is 'sort of succeed sort of fail' the party manages to not get captured/killed/etc, but they took too long and the BBEG completed whatever he was doing. Now he can enact the final ritual/move on to the next phase of his plan/rule over the desolation he's created.

Ozymandias anyone?

Kind of to piggy back on Advisor.  The idea of failing forward.  If they succeed there are more threats to overcome.  If they fail the BBEG encounter.  "You have fought valiantly.  I think killing you would be a waste.  Come, bring them to the chamber, we shall convert them to death knights."

Now you have the players in a dungeon and can stage a kind of prison break.  Give them a prison to play around in for a bit.

Is it just me or did anyone else immediately flash to the Buffy the Vampire musical episode "Where do we go from here" song?

Specifically, the line the battle's done and we kinda won so we sound our victory cheer

In reply to Advisor (msg # 21):


I did this from time to time in table top games, and its certainly something I want to do here. Another favorite is, there are several items you need to collect in a dungeon, and the party gets some, and the antagonist gets some. The party can then decide if they want to concentrate on protecting what they have, or going after the rest, or trying to do both.

I actually used this as part of my intro into a Temple of Elemental Evil campaign since the characters are supposed to enter Hommlet with little money or equipment. They were captured by bandits/slavers (who will show up in a future part of the campaign so they get a bit of revenge/retribution)

I left their minor stuff (bedrolls, mess kits, etc.) alone but for items like weapons, armor, rations we rolled a d4 for each item:
1 = Item was stolen
2 = Item is there and damaged beyond use/repair
3 = Item is there and damaged but still usable
4 = Item is there and in good condition

For items with a quantity (gold, rations, torches, etc.), a roll was made for the quantity remaining. For example, if they had 20 gold, we rolled a d20 and that was how much gold they had left (maybe it was hidden in their sock)...

It worked pretty well and, while they were disappointed about losing their stuff, they all thought it was a fair method...
This message was last edited by the player at 13:58, Sat 20 May 2023.