Sandbox world map issues

[fireinthedust]

I'm trying to build a Sandbox setting, aimed at 6th level as epic (ie: the classic E6 campaign).

The problem is that, while making up individual areas isn't a problem, fitting them together into a larger world is. I don't know where to place different communities so that they work. I make a map, then put the forest in one spot, and I know I need a river for several of them, but then when I draw the river it just looks contrived (possibly because I'm contriving it as I look at it).


For those of you who design "local areas", with several communities and locations scattered about, how do you determine where things go?


The problem gets worse as I try to think in terms of World Map-making. I have the general ideas of where I'm going to put the different regions down, and they work, and are looking good.
The problem is where to put an inner sea, or whether I need yet another mountain range (as, obviously, I just put another mountain range on the other side of that country!). Fiddly little details I havn't worked out yet; or else taking the many ideas I do have worked out (like Mountain Dungeon A and Swamp Temple B, to be in Naiton X) and figuring out where in said nation both of them go.


Maybe I'm making the small local area too big compared to the larger world-map size. Maybe I should just have the big map, and say "yeah, it's all in there" without placing it just yet?


Still, in short: two different scales of maps, and no idea where to put things in either one.

How do you pick where things go? Isn't map making arbitrary?

 

[S'mon]

Well, for map-drawing the best approach is to look at real-world geographic maps, see how rivers flow, how mountain ranges look, that kind of thing.

For locating the sandbox, well I suggest having a big, very lightly sketched world-map, say 100 miles to the cm, then a small sandbox, say 5 miles to the cm, and have a general idea where it's located - "on the edge of that mountain range" or "on the frontier of that nation" but don't detail the intermediate scale yet. That gives you tons of room for later expansion on the edges of your initial sandbox area, or to create new sandboxes in the same world.

 

[S'mon]

My most recently created campaign world, a Dark Ages setting set centuries after the fall of the (quasi-Roman) Sharan Empire, I started with a general idea that the geography resembled the real world, then placed various sandboxes for various campaigns and possible campaigns - "This is where the northern Pyrenees/southern France would be", "this is Aquitaine", "this is Navarre", without ever needing to draw a world-map. I used some of Rob Conley's "Points of Light" sandboxes. I think this approach worked extremely well.

 

[MatthewJHanson]

Maybe I should just have the big map, and say "yeah, it's all in there" without placing it just yet?

I think starting small and ambiguous is a good way to go. Your players don't actually need to know where something is until their characters plan to go there.

Also, unless your your players are cartographers, they probably won't notice if your maps look contrived. They are more interested in exploring the cool dungeons than figuring whether your rivers run the same direction.

And if they do notice something, work with it.
Player 1: "Why is there a swamp on top of a mountain?"
DM: "You'll have to explore it to find out."
Payer 2: "Oh, maybe a powerful black dragon magically created the swamp to make the mountain feel more like home."
DM: "Maybe." [starts writing down notes]

 

[Olgar Shiverstone]

One technique -- take a real world map, note the layout of the terrain, change the cities and towns to villages and hamlets, then invert/reverse/mirror the map so it isn't a 1-to-1 analog of the real world. That way you let the real world work for you on geography and economics, and can focus on the fantasy.

My preferred approach, though, is not to worry about the larger world. I prefer to build my sandbox small at first -- a village, a dungeon, some surrounding wilderness, and some rumors of adventrue elsewhere -- then only expand as the players choose to follow specific leads. The world grows organically as the players explore it, and you don't have to have many more details prepared than you are ready to handle.

 

[LostSoul]

Here's a series of blog posts on creating a fantasy sandbox: Bat in the Attic: How to make a Fantasy Sandbox

That's far too involved for me, but you may find it works. When I generated my current setting, I used an existing map (the Nentir Vale), rolled on some tables I wrote to flesh it out, and tried to make sense of things. I also decided that there aren't any nations; they've all fallen, just to make it easier on myself.

 

[Ariosto]

A geophysical atlas should include a basic discussion of landforms, climate, seasons, vegetation patterns, and maybe more. A historical atlas will be all about showing patterns of migration, settlement, trade and so on. Works such as The World Almanac and Book of Facts and The Viking Desk Encyclopedia are other possible aids.

Such works need not be up to date. Not that there have not been advances in scientific investigation in these fields, but such masterpieces as the maps for The Lord of the Rings predate the development of plate tectonic theory.

If a physical library is not convenient, you could try Google, Wikipedia, etc..
----------------------------------------------------------------------------------
As to what the Old Masters did: Dave Arneson's fairly compact Blackmoor region was, IIRC, based on "an old Dutch map" reversed (as if put on a light table and traced on the other side of the paper). He may have changed the scale, and altered some of the terrain. Gygax based his original "world of Greyhawk" (not exactly the published one) on a map of North America.
----------------------------------------------------------------------------------
It might be fun to start with a map of Ontario. You may already be familiar enough with the Canadian Shield region, the Hudson Bay Lowlands, and the Great Lakes - Saint Lawrence Valley to have a sense of what "feels" right. The south might not be so populous, with more forest and less agriculture.

Maybe maps from the 17th, 18th or 19th century would give a good feel. Just change the names. Toronto, for instance, could be "Metropolis"...!

 

[Sunseeker]

I always start from a central point, and imagine how the civilization would expand from there. Take a rive-mouth for example, the logical movement is upstream. Perhaps upstream they find the river goes into a great forest which is near impassable. So they build a new town on the edge of the forest, having good materials(the forest) and water, the town grows.

Since they can't move upstream(or can but it's too difficult for mass movement), they move outwards, around the forest, until perhaps, they run into mountains. They locate themselves near a small stream, and this town revolves around mining.

Once you've established about 3 major towns, you fill in the area in between and around them with farmland.

Remember that the key to classic civilizations were defense and resources. Defense ensured resources were controlled, resources ensured weath could be produced. Under this guideline, towns should be established at or near strategic locations. River mouths, river connections, passes in mountains, large harbors.

While it's true that taking into account modern geological theories, the fact is that fantasy maps are more useful when the geography suits your needs as much as towns do.

Generally when I make maps, it's map first, civilization second. I usually work on the idea that a "nation" takes up most of a geographical area, such as a large plain between two mountain ranges. Then, I design additional nation-zones to attach to the first one.