Metroidvania

[Razanir]

As we all know, there's a sliding scale of linearity v openness. And extremes are bad in both directions. Make a campaign too linear and the players complain about railroading. Make a world too open, and the players (in my experience) travel with the wind and never have any sort of a goal. How fair would this be as a compromise:

Metroidvania. The players get free reign over a smaller portion of the world surrounded by insurmountable waist-high fences. The players get a sense of control over the world, but I don't have to worry about them deciding to travel halfway across the world on a whim and making me need to improvise a new part of the world

[Kyberwulf]

Doesn't work, as soon as players know about the fences. The first thing they want to do is get over them somehow.

[Mando Knight]

Metroidvania doesn't have insurmountable waist-high fences. It has walls that need a particular level of strength to be knocked down, gaps that require grappling hooks or long jumps to clear, and other terrain and barricades that are impassible when you first encounter them, but trivial once you return with the proper empowering item... or if the players are determined enough.

[Mewtarthio]

A bit odd that you'd try to constrain your players by putting them in the genre that invented sequence breaking...

[Razanir]

Perhaps Metroidvania isn't the best name...

I meant the fences for keeping them within the region. So in what I'm considering doing, they have free reign over a floating island, but because of the evil king, there's a trade embargo and airships off are hard to come by.

[Grinner]

Take note: All Metroidvania games generally take place in a ship, underground, or in a building. The tactics involved in reigning in players won't work as well in an open air environment. Moreover, the ways in which a player can interact with the environment are far more limited in a videogame than a tabletop RPG.

When gravity prevents them from reaching the upper ledge, they can build a ladder.

Political situations like you suggested should be slightly more effective, but that won't prevent them from trying to build a raft or sneak through the evil kingdom.

[Rhynn]

Doesn't work, as soon as players know about the fences. The first thing they want to do is get over them somehow.

Yup. It also pretty much requires railroading. "Oh, you can't cross these mountains." "Why not?" "BECAUSE!" Any reasonable obstacle you can come up with, the PCs can find a way to overcome, and you're left with railroading.

Fortunately, the premise is false:

Make a world too open, and the players (in my experience) travel with the wind and never have any sort of a goal. How fair would this be as a compromise:

That's not a problem of openness, that's a problem of GMing.

Sandboxing works fine. You create a world, map it, and put things in it (including adventure locations). You can plop more things in it at any point. You start wide and shallow, and add depth as time allows and as PC interest indicates. (Basically, prep the next place they're planning to go ahead of time, between sessions.)

You can't really get more open than that. It's a completely silly notion that openness or sandboxing somehow requires you not to put in adventure locations - how would that even work? It's not a game if there's nothing to do.

Fortunately, there's some great resources out there:
How to create a sandbox
How to run a sandbox hexcrawl

There's also entire games built around it, like Adventurer Conqueror King, that are completely wonderful.

Metroidvania doesn't have insurmountable waist-high fences. It has walls that need a particular level of strength to be knocked down, gaps that require grappling hooks or long jumps to clear, and other terrain and barricades that are impassible when you first encounter them, but trivial once you return with the proper empowering item... or if the players are determined enough.

Now, this makes a lot more sense.

I think the analogy fits. The Metroid games only have so big of an area to explore in total. So does a sandbox/open campaign: you cannot map and detail the entire world. You should start small, and then expand as the PCs' abilities, interests, and travel dictates.

As the PCs' powers of travel and ability to deal with adversity grows, they can go further and explore new and more challenging places.

For instance, in my Dark Sun sandbox, the players start in the city of Tyr. Exploring the nearby deserts and the Tablelands is a no-brainer. There's paths, caravan routes, known oasises, and forts, leading to other cities, and along them can be found places for adventure. But this is all comparatively "safe" (which isn't saying much, on Athas).

To the west are the Ringing Mountains, and behind them lies the terrible, alien jungle of the Forest Ridge; both are incredibly difficult to cross, and starting PCs would probably die (unless they got really lucky). That's not an insurmountable, that's a "eventually!" Once they get past them, they'll find the Hinterlands, a wide open area with many more locations; and beyond them, the Crimson Savannah, where the presence of the Kreen Empire presents a different sort of obstacle.

To the south are the Dead Lands, an obsidian plain full of undead and the ruins of ancient cities. Exploring here is obviously deadly and dangerous, and only powerful PCs will survive it, except by luck.

To the east is the Sea of Silt, which presents huge physical obstacles to adventure: the PCs need to be stinking rich to get their own silt skimmers, or have access to magical transportation or flying mounts (or maybe some friendly or charmed giants to carry them). Beyond lies one of the dark wonders of Athas: the Valley of Ash and Fire, and Ur Draxa, the lost city of the Dragon.

To the north, things are actually easier, at least to start, but there are great stretches of deadly wasteland between areas of interest, ultimately making travel arduous and requiring the PCs to get tougher to go further.

Meanwhile, in my Savage Frontier sandbox, to the north there is increasingly wild and dangerous territory culminating at the Spine of the World Mountains, which would be a great challenge to cross; and beyond them lie lost wonders in the Great Glacier. To the west lies the Sea of Swords, and in and beyond it more wonders on islands. To the east is the Anauroch, an unforgiving desert that only the best-prepared could cross. Only to the south is the way clear, but that way lies settled lands, not a frontier of adventure - there's little motivation for PCs to go there, although "it's there!" is always enough.

So, a sandbox can easily and naturally be divided into regions that aren't all easily "open" to the PCs. Physical obstacles like mountain ranges, oceans, and deserts can do it very well, as can political obstacles like hostile nations.


If I sound like I'm advocating sandboxes as the best form of campaign, I am - for fantasy RPGs, anyway. For cyberpunk, SF, and modern games, not so much, although still yes in some circumstances and styles of game...

[SiuiS]

As we all know, there's a sliding scale of linearity v openness. And extremes are bad in both directions. Make a campaign too linear and the players complain about railroading. Make a world too open, and the players (in my experience) travel with the wind and never have any sort of a goal. How fair would this be as a compromise:

Metroidvania. The players get free reign over a smaller portion of the world surrounded by insurmountable waist-high fences. The players get a sense of control over the world, but I don't have to worry about them deciding to travel halfway across the world on a whim and making me need to improvise a new part of the world

Metroidvania doesn't have insurmountable waist-high fences. It has walls that need a particular level of strength to be knocked down, gaps that require grappling hooks or long jumps to clear, and other terrain and barricades that are impassible when you first encounter them, but trivial once you return with the proper empowering item... or if the players are determined enough.

There is a lengthy post series on a DMing blog that covers this, actually; Schroedinger, Chekhov, Samus. I think it falls into the category of things a lot of folks have been doing anywhay, but when you see it spelled out clearly you go "... Huh."

For instance, in my Dark Sun sandbox, the players start in the city of Tyr. Exploring the nearby deserts and the Tablelands is a no-brainer. There's paths, caravan routes, known oasises, and forts, leading to other cities, and along them can be found places for adventure. But this is all comparatively "safe" (which isn't saying much, on Athas).

To the west are the Ringing Mountains, and behind them lies the terrible, alien jungle of the Forest Ridge; both are incredibly difficult to cross, and starting PCs would probably die (unless they got really lucky). That's not an insurmountable, that's a "eventually!" Once they get past them, they'll find the Hinterlands, a wide open area with many more locations; and beyond them, the Crimson Savannah, where the presence of the Kreen Empire presents a different sort of obstacle.

To the south are the Dead Lands, an obsidian plain full of undead and the ruins of ancient cities. Exploring here is obviously deadly and dangerous, and only powerful PCs will survive it, except by luck.

To the east is the Sea of Silt, which presents huge physical obstacles to adventure: the PCs need to be stinking rich to get their own silt skimmers, or have access to magical transportation or flying mounts (or maybe some friendly or charmed giants to carry them). Beyond lies one of the dark wonders of Athas: the Valley of Ash and Fire, and Ur Draxa, the lost city of the Dragon.

To the north, things are actually easier, at least to start, but there are great stretches of deadly wasteland between areas of interest, ultimately making travel arduous and requiring the PCs to get tougher to go further.

Meanwhile, in my Savage Frontier sandbox, to the north there is increasingly wild and dangerous territory culminating at the Spine of the World Mountains, which would be a great challenge to cross; and beyond them lie lost wonders in the Great Glacier. To the west lies the Sea of Swords, and in and beyond it more wonders on islands. To the east is the Anauroch, an unforgiving desert that only the best-prepared could cross. Only to the south is the way clear, but that way lies settled lands, not a frontier of adventure - there's little motivation for PCs to go there, although "it's there!" is always enough.

So, a sandbox can easily and naturally be divided into regions that aren't all easily "open" to the PCs. Physical obstacles like mountain ranges, oceans, and deserts can do it very well, as can political obstacles like hostile nations.


If I sound like I'm advocating sandboxes as the best form of campaign, I am - for fantasy RPGs, anyway. For cyberpunk, SF, and modern games, not so much, although still yes in some circumstances and styles of game...

Gorramew! Now I gotta set up an ACKS L5R game...

[Coidzor]

Doesn't work, as soon as players know about the fences. The first thing they want to do is get over them somehow.

This. You want to hook the players so they don't want to go gallivanting off somewhere that there's no plot without some kind of provocation.

[Rhynn]

Gorramew! Now I gotta set up an ACKS L5R game...

I was just thinking about ACKS for L5R or Sengoku today, as a consequence playing Total War: Shogun 2 and the Gekokujo mod for Mount & Blade: Warband and thinking about how I'd adapt 13 Assassins as an open but limited-scope campaign*... It's not really surprising, though - it is a D&D retroclone, and for all its implied setting, that means it can be adapted for most fantasy uses.

* Well, a cross of 13 Assassins and Forty-Seven Ronin. "Honor demands you kill this daimyo. Here's your map and what you know. Begin." The idea is that the daimyo goes about his routines, travels his clan's provinces from castle to castle, and the PCs would get to decide whether to ambush him on the road, try to assault or infiltrate a castle, try to become his trusted men to get close to him, start a conspiracy, or even try to get some of his vassals to turn against him... sort of a "directed sandbox" I guess: a limited area, a firm goal, but total freedom to travel within that area and pursue the goal.

This. You want to hook the players so they don't want to go gallivanting off somewhere that there's no plot without some kind of provocation.

Yup. Even in a sandbox, it's essential to provide... not direction so much as obvious options. Hooks. The traditional method is to create several rumors for eahc adventure location, and have the PCs run into one for several of them at the start of the first session (they can uncover more by asking around). Old D&D modules almost all included a list of rumors about town.

[Knaight]

The players get free reign over a smaller portion of the world surrounded by insurmountable waist-high fences. The players get a sense of control over the world, but I don't have to worry about them deciding to travel halfway across the world on a whim and making me need to improvise a new part of the world

I'd recommend against keeping a small setting by virtue of the "you can't leave" set of methods (though there are places where this works), and instead recommend trying something in the "you have a reason to stay" set of methods. Maybe the characters have existing connections, property, and generally a place in society that is worth keeping that is also in a particular place. Maybe there is something about a particular place that makes it a good place to achieve their goals. There are a lot of options, and using them can help keep a sandbox manageable, and potentially help make a campaign better.

[Rhynn]

I'd recommend against keeping a small setting by virtue of the "you can't leave" set of methods (though there are places where this works), and instead recommend trying something in the "you have a reason to stay" set of methods. Maybe the characters have existing connections, property, and generally a place in society that is worth keeping that is also in a particular place. Maybe there is something about a particular place that makes it a good place to achieve their goals. There are a lot of options, and using them can help keep a sandbox manageable, and potentially help make a campaign better.

Sandbox or not, this is always good advice, IMO. Tying the PCs to the setting in concrete ways makes the GM's job easier, and it involves the players more in both their PCs, the NPCs, and the setting in general. Better game all around, I think.