I’m in the process of trying to come up with a random hex stocking method that works for me. My point of departure is the traditional dungeon stocking method:
- 1-2 monster (3 in 6 chance of treasure)
- 3 trap (2 in 6 chance of treasure)
- 4 special
- 5-6 empty (1 in 6 chance of treasure)
In the context of wilderness stocking, “monster” would be interpreted as a lair or dungeon, “trap” as a hazard of some sort (or perhaps an abandoned ruin that is uninhabited but still dangerous), and “special” as everything else (including, probably, settlements). Each of those categories could have a subtable or set of subtables to determine the type of lair, etc. I want to keep the system as simple as possible, but I think I need more than this for inspiration, because I don’t find myself actually finishing a stocking process. That’s a sign to me that I need more help from the tables.
One thing that is blocking me is how settlements interact with the stocking. I could just place the settlements, and then stock the areas between them, but I kind of want the generator to do that work for me. It seems that there are really only three or four meaningful settlement sizes for my purposes here. Stronghold, town, village, and isolated settlement (outpost, traveler’s inn, farmstead, etc). Maybe half of the special results would result in some kind of occupied settlement. Ruins would be covered in the monster, trap, and empty (when with treasure) results on the main table.
I don’t care much about things like logical food supplies (I can come up with after the fact explanations), but I do sort of like the idea of graduated civilization and wilderness. Here is another place where three or four categories seem appropriate: civilization (town, fortress, etc), threatened ground (the border between civilization and wilderness), and wilderness. There is a mathematical choice to be made regarding how this works: should the stocking roll be independent or related to results in adjacent hexes? If the process is independent, then we can infer the level of civilization (and thus danger) from the resulting distribution, which will end up being regular.
If it is dependent, then the process would be more like an organic outgrowth from some seed hex (probably the starting town), which would have some chance of going down in civilization level and some chance of going up. The chance of civilization level decreasing as you expand outwards would probably be greater than the chance of civilization level increasing, resulting in a setting that is dominated by wilderness (and thus adventure opportunities). Victor Raymond uses a system like this to generate terrain type in his Wilderness Architect series of articles in Fight On! (issues #2 and #3). He places settlements by determining random direction and distance from the starting settlement.
So, to expand the the “4 special” hex result:
- Trick (magic statue, etc)
- Stronghold (50% chance includes another settlement)
The meaning of this table (based on expected values): 1 in 12 hexes will contain a settlement, and 1 in 48 hexes will contain a stronghold. Following the DCC recommendation of 100 miles square, I am considering approximately 16 x 16 six mile hexes, which is 256 hexes (and also compatible with my ideas on hex zooming). Overall, such a wilderness would have (approximately) 86 lairs, 86 empty hexes, 43 specials (21 of which would be settlements) and 43 hazards. How does that distribution look? One thing that does not seem quite right is that an outpost is just as likely as a stronghold using this scheme, but on the other hand this will lead to around 5 strongholds on the map, which seems to be about right (especially if they are of varying levels of power and influence). Also, the “monster” result would include things like bandit forts and the towers of evil magicians.
Any ideas welcome.