Saturday 31 March 2012

The Contents of Hexes

A few days ago I posted a selection of grabs from google maps, indicating how large a 1-mile hex actually is, and the kind of things that you might find in one. Then Chris responded, to the effect that (and I'm putting words into his mouth here) 1-mile hexes are all very well, but they are perhaps too granular - it's a heck of a lot of work to map out a setting at the 1-mile level. He posits an "emergent exploring rule" for 6-mile hexes, wherein you simply discover more things in a 6-mile hex the more time you spend exploring it and the more guidance/information you have about it.

The commenters immediately provide solutions, which are pretty good. Myrystyr's one seems like the best:

A generalised X-in-dY check per Z-time-period, mayhap? Default 2/1d6 per day - 1/ for stumbling-about blind luck, 3/ or 4/ for guides, clues, or character types with better search capability - and up to 3 or 4 times per day for potentially feature-rich locales... And, roll two different-coloured d6 at the same time, 1 the feature check and 1 the encounter check.

I like the elegant simplicity of this, and I think you can add some other modifiers: the chance would increase or decrease depending on terrain (lower for a forest, higher for plains), demographics (lower for empty wilderness, higher for settled), and so on.

Chris also suggests some other rules of thumb:

  • Castles, cities and the like should all be in plain sight unless intentionally hidden away (like Gondolin or Derinkuyu). Heck, roads point you directly to most of them.
  • Infamous lairs, ruins and dungeons should, of course, retain their "Here be dragons" hex map icons and easy-to-find status. The yokels can point out exactly in which direction the castle we don't go near lies.
  • More obscure lairs, lost ruins, buried tombs and especially treasure map loot should require a bit of active hunting out by adventuring parties.

The main, immediately identifiable problem is that you now need some way of governing how many locales can be discovered in this way. The amount of "interesting stuff" in a hex has to be finite. But it is also variable. In my 1-mile hex examples, some have only one thing that you could make into a "locale", but some have several. This is a function, mostly, of how settled the area is, but it also introduces other factors. The more mountainous a region is, for instance, the more stuff you will get per hex on average, because each hex represents more land. (Hexes are "as the crow flies", whereas the land underneath is folded by hills and valleys, making it more packed with actual physical geography.)

What I would suggest is that, by default for a 6-mile hex, any settlement of more than 250 people, any castles or towers, and major geographical features like lakes, rives and mountains, are basically discoverable without the players doing much more than travel through the hex a few times.

You would then have a default of 3 'discoverable' things per 6-mile hex, modifiable by geography, level of settlement, history, and so on:

  • One of these would be something that the players can find if they make an effort to explore. All it would require would be for them to say "we look around the area".
  • Two of them would be 'secret', obscure, or lost things (a ruin, buried tomb, a secretive monster's lair, etc.) These would be near impossible to find by exploration without some level of guidance or inside knowledge. 

This would allow the players to discover stuff just by looking, but would also reward efforts they made to interact with locals to try to get information, and to risk random encounters by searching.

"Why 3 things per 6-mile hex? Isn't that too much?" I hear you cry.

The answer is: no. To illustrate, I'll use two of the 1-mile examples I put up a few days ago, zoomed out to 6-miles (ish; they're a bit smaller and probably more like 8 kilometres):

This is Canterbury. As you can see, there is not only the city itself, but two other large villages or small towns (Bridge and Sturry) plus a shitload of other hamlets, and plenty of woods as well as farmlands. In a densely settled area like this (probably much more densely settled than in medieval times, admittedly) you would need at the very least 3 adventuring locales not even mentioning Canterbury itself.

This is Tiree -a good chunk of the island. Note that there are 5 large bays to be found here, along with several smaller ones and a lot of little sheltered coves in the centre-right. There are four scattered farming villages, an airport, and 3 lochs. There are also some tiny islets here and there in the sea, too. Again, at least 3 adventuring locales needed.

But 3 is a default, and can be reduced. For example:

The mountains of Assynt. It's empty and devoid of settlement, though there is still plenty to interact with - 6 named lochs, plus lots of smaller ones, and a folded, rugged landscape concealing God Knows What. 3 adventuring locales might be inappropriate in such a landscape, because you don't want to be silly - not every lake has to have a Water in the Waters; not every mountain has to sit atop an ancient dwarven citadel. But perhaps it's also unrealistic to just say "it's an empty hex".

Of course, you wouldn't want your players to figure your system out. An alternative would be to randomise the number of adventuring locales in a 6-mile hex: d3-1 for sparse, d3 for average, d3+1 for settled, etc. Modified also by type of geography.


  1. Another consideration, perhaps, is what qualifies as a locale / encounter. Finding an unusual rock outcropping or crossing paths with a herd of deer may not matter, but if you're tracking minor stuff like that the number of encounters / locales should be higher. Arguably much higher.

    1. Yes. I would subsume stuff like weird rock outcroppings or crossing paths with deer in random encounter tables, so you could get any number of those.

      Locales are anything fixed and static that contains either treasure or monsters. Could be as simple as a cave with a bear in it, or as complex as an entire megadungeon.

  2. I think 3 features per 60-mile hex is realistic, but I have 0-1 major & 0-5 minor features in 25(!)-mile hex and I already heard players say to each other "There are so many things around here, we can just walk around out base settlement till 10th level". :)

    1. You mean per 6-mile rather than 60, right?

      I don't mind them wandering around their starting settlement for the entire game if that's what they want to do and provided there's enough stuff to explore. Saves me work mapping out other areas. ;)

  3. This comment has been removed by the author.

    1. I like overcomplicating things. ;)

      Also, it doesn't quite result in what I want. If you use % in lair, you basically just get monsters in lairs - or else castles and villages if the inhabitants are hostile. What about other kinds of locale - ones not inhabited by anything?

      Also also, what about unique monsters or NPCs and their lairs/homes/whatever? Random encounter tables with % in lair won't generate those.

      It is simple, but it doesn't quite do it for me.

  4. Nice pics! The amount of adventure locations you should suggest is about the same as whats in the rules (1-3 per 5 mile hex).

    As for finding them, why not simply use % in lair as it was intended? It works just fine. Roll for random encounters at least once per hex or as often as desired. Castles and villages etc. are simply kinds of lairs if the inhabitants are hostile, and if not then the party should be able to find them basically at will as you suggest - provided they have not gotten lost. The DM will have to already know how many castles and villages are in the hex, but that's the sort of basic detail that should be known at the start anyway. No point in overcomplicating things.

    1. I replied above to the comment you deleted. I was too quick for you. ;)

  5. You could still do the three things but what they are would change depending on the area.
    So the remote, unpopulated area, would be:
    1. Lone dragon's hunting ground.
    2. Reclusive hermit, 250 years old. Can be asked questions.
    3. Small lake, only water source for entire hex. Double wandering monster chance here.

    Compared to Canterbury:
    1. Town of Shady Grove. Cursed, PCs have to ...
    2. Knight challenging all passers by.
    3. St Cuthbert's Cathederal. Home to relic. Very rich. Current bishop very active in giving out quests and tasks to adventuring groups to increase the glory of the religion.

    The more populated area might not necessarily have more interesting things happening there. People would just be farming etc. The interesting monsters would have been killed a long time ago.

    I think the three rule might have more to do with how much you want to prepare and how much choice your players can handle at one time.


    1. Yep, although I think people overemphasis prep. Your examples are totally fine and perfectly useable in a game - the only extra things you would need are stat lines for the dragon and the knight. It doesn't take long.

  6. While I'm enjoying these posts, and find them really insightful they also strengthen the realization that I don't like hexmaps for this sort of thing. There simply isn't an ideal scale that maps to what's "on the ground".

    I mean, if each hex holds X things, or Y things if you get granular, or one thing if you get abstract, why are we using hexes at that point? While they may be good for tracking travelling speed that's not so simple either, is it. If a hex contains a road, but is actually bigger than that, how are the characters moving "though the hex", how do you factor in speeds for exploration etc.

    I dunno. I guess I don't have a better solution, but hexes don't really do it for me at this point.

    1. I think I might write a post about that.