On Eating your own Dogfood: generating adventure settings using AFG MOSTROTRON and MONDOTRON

I’m a firm beliver in eating your own dogfood. As I’m putting together “Castles on the Hills, Dungeons beneath” and I was feeling the white page syndrome, I decided to prototype MONDOTRON, an incomplete campaign Points Of Interest generator to kickstart writing on CotHDB: toss dice in, shake, get dungeons, settlements, weird and secret stuff out. Creatures would come out of MOSTROTRON, the AFG creature generator (you can find something similar but with a simpler structure here).

Two approaches seemed obvious: one more simple, teleological and user-oriented and one more simulationist. After jotting down a number of deciding factors to be used for the simulationist approach I decided to ignore them (for now) and go forward with the simple “table with of possible results”. MONDOTRON is going to be a stopgap solution: I need a usable setting generator soon rather than a perfect one later.

The simplest approach is to, for each hex or node in your map (maybe generated with Hexographer), do the following:

  1. come up with a name, possibly based on the geographical features.
  2. use a table to determine if monsters, treasures and traps environmental hazards are present.
  3. place monters, treasure and traps environmental hazards appropriate to the location. If the room location is empty, add an interesting and visible landmark or a special thing (ruins and standing stones are always a decent option), if the treasure is unguarded it might be buried, monsters might be in a dungeon and so on.
While this guarantees results which are exactly the same as dungeons with bad weather, such as a tribe of goblins and their hoard in a 30′x30′ room hollow in the woods, the result is not an open-sky dungeon: the main difference between dungeons and wilderness is the quality of the interaction with the environment. In a dungeon stone walls stop you from evading the ogre sentinels while in the wilderness only the most extreme environments, such as lava, crevices and vertical cliffs stop you in the same way, and only if no way to fly, abseil or climb are available. In the wilderness the lay of the land is much more interesting, interactive and varied than dungeon stone walls, and forgetting this will lead to frustrated players.

So, yes, this approach works. Can be improved on but it’s solid.

On the other side bringing a pickaxe in a dungeon might allow patient adventurers to loudly dig themselves and treasure out of trouble. It’s been done and it works.

A more sophisticated approach is to have a table with expected results, like this one:

  1. Dungeon: underground complex. You might be familiar with the concept.
  2. Special place: hot springs, standing stones, an extremely deep crevice. Players’ interest should be piqued.
  3. Hidden place: an hideout, a crypt, a buried treasure, you get it. The location must not only be remote but also purposely hidden.
  4. Outpost: small, single settlement, like a farm, an hunters’ lodge or a fort
  5. Village: a bigger settlement. Towns and cities should be extremely rare as they need extensive food supplies, which means miles of fields around them and security from monsters.
  6. Nothing special.

So for each hex you roll on the above table and then on the relative subtables. If you want to have a more mundane setting roll a dice bigger than a d6. You can also throw in modifiers depending on the location and its surrounding. This is the solution used for MONDOTRON. Sure it has a number of shortcomings: distribution of content can be far from perfect (but of course improvable), and a random approach leaves you hostage to luck. Also it doesn’t usually produce extremely interesting results because of its structure: the goodness of the result is on a first instance due to the single entries and not by their interaction.

It has of course a lot of plus sides: first, it’s really easy to write and to use. Second, leaning on any simple generator takes away hard choices but allows the author to take over when the creative cogs start cranking.

Also, it rides intuitive continuity toward the weird fantasy and pulpy end of the spectrum, possibly forcing a more heavy-handed approach to justifications to incongruence resolution. Intuitive continuity tells us that such incongruences are in some way justified, and can be roughly classified in one of the following classes:

  • No interaction. the first, simplest case is that the two entities do not interact meaningfully. Mostly non-changing.
  • Unbalanced interaction. This class of interaction has uneven tensions. The situation is therefore dynamic, change happening or about to happen. This part of the scenario is already evolving and will reach its natural conclusion unless meddled with, conclusions that will have consequences. Players should feel empowered not only to cause change through action, but also through deliberate inaction.
  • Balanced interaction. The third case is that the interaction is somehow beneficial for or tolerated by both sides. The tension must be balanced by another tension: this tension can be either internal (like a deal between a city and a warlord) or external (such as parents stopping brothers from beating the heck out of each other). The tensions will mostly counter each other and bring a non-evolving status quo, but might as well result in a new tension toward another element in the scenario. The resulting tension might end up destabilizing the whole: the example is a city paying some kind of tribute to a warlord and, as the lord makes new enemies, the city will suffer from their hostility too, in turn causing problems with the warlord and going back in due time to an unbalanced interaction.

All three lead to adventure scenarios but timing and opportunities are much different. Players can decide to intervene in all three scenarios, by either:

  • for the first class, forcing interaction between the parties.
  • for the second and third class, removing interaction or its causes, for example by blockading traffic between the city and the Warlord.
  • for all cases, acting in favour or against one of the parties, or as a new separate party, thus either stabilizing or destabilizing the situation.

Now, a number of factions, people and circumstances might lead adventurers to do some of the above. I’m pretty sure you can came up with these on your own.