Wednesday, October 31, 2012

Blood Island and Hexcrawling—A different approach.

Blood Island is huge, like a couple hundred miles by a couple hundred miles. It's a hexcrawl, in that the parties go from hex to hex and discover what they can. Each hex has either:

1 Nothing. Describe the terrain well.

2 Random monster encounter.

3 Something special (old mine, cave system, ruin, monster plus den to explore, former settlement of their people, former settlement of the enemy, place of mystery or wonder, place with a strange magical effect, special monster, intelligent monster, camp or settlement of an intelligent species/race never before seen to make an ally or enemy of, distinctive natural landmark, magical natural landmark, portal that takes you to somewhere else on the island, portal that takes you to another plane, natural hazard like quicksand or mudpit, a non-natural trap of some sort, entrance to the underdark, etc.)

4 Something special, like above, but which gives clues to unlocking the greater mystery of the island.

I've yet to figure out the odds of each thing on the d100 roll, but they order above is roughly from common to rare.

As a DM, if I want to put a special thing which gives clues somewhere on the island, like say on hex 276, and that clue is somehow important to unlocking the greater mystery, but the party never goes there, and therefore never discovers it, what a waste of time that would be to create the whole complex/dungeon/whatever. It would also annoy the players to have to figure out what hex they haven’t gone to, in order to find the thing that unlocks the riddle so they can conquer the whatever.

At the same time, you don't want to make wherever they go be the next thing you want them to find, based on what you've just created. That's a railroad, in that no matter what the players decide to do, they will end up with the encounter you prepared for that session. That defeats the purpose of the whole sandbox approach to hexmap exploration.

Here's what I've decided to try out: we already randomize monsters, so why not randomize all the other encounters? If I create a half dozen or so of each type of encounter, and number each one, if I roll for what is in the next hex the players encounter and it comes up “abandoned mine,” I go to my collection of half a dozen abandoned mines which I've already mapped out and stocked with monsters and treasure and clues and giant rats and copper pieces and all that cool stuff, and randomly choose one to be the one the players encounter.  After that, I never use that encounter again.

If I take this approach, there is no railroad, and no DM time and resources are wasted. Wilderness encounters become a series of random one page encounter areas, basically.

Obviously I reserve the right to re-roll if the encounter doesn’t make sense for either the geographical area, or if it doesn't fit with what the players have already discovered nearby. But I think this approach can work.

It will require a lot of upfront design time on my part, but after each session all I need to do is redesign a few new encounters of whatever encounter types the players explored that session. That way I don't re-use things, the players aren't bored, and it keeps it fresh.

It seems like a good way to make a wilderness hexcrawl interesting for the players, as opposed to a lot of random monster encounter areas, avoids a railroad, and it maximizes DM efficiency and creativity by making sure that whatever I make up is eventually used.  The other benefit is that if the party randomly teleports or otherwise finds a way to go to a some random part of the island (never underestimate how good players can fuck up all preconceived notions of what that adventuring session was going to be about), you as DM don't have to have filled in all 6000 hexes in advance to be ready for what they do and where they end up.  Also note that this method doesn't prevent a DM from making certain hexes have definite fixed encounters. If I want hex 541 to have the remains of the tower of the arch-mage Felix Unger, it can be there as a permanent static encounter, without interfering with the random hex system above.

What do you think?


  1. Sounds plausible. Some might see it as taunting the quantum ogre ("I built this thing, you're going to encounter it!") but I'm not concerned with that.

    It might suffer a little in that if the hexes are populated as they are explored you might lose the ability to establish links between them.

    I think player agency is really important. Running a hex crawl, I'd look for information that lets me advise the players, given them hints as to which way they want to go. If the hexes are basically random then you could end up with a situation where they might as well act as random, much like we did in Dwimmermount.

    Especially with your "reveals clues to the mystery" locations, you probably want to identify hooks that join them. You don't necessarily need to place the locations ahead of time, but being able to give hints as to where to go next is really useful.

  2. It's a good point. I think this approach has to go hand in hand with established static locations to explore in certain hexes, or else the clues in the 4th type of encounter would be less useful. each clue would need to explain both something about the island's mystery, as well as give a clue as to where to go next.

  3. "Here's what I've decided to try out: we already randomize monsters, so why not randomize all the other encounters? If I create a half dozen or so of each type of encounter, and number each one, if I roll for what is in the next hex the players encounter and it comes up “abandoned mine,” I go to my collection of half a dozen abandoned mines which I've already mapped out and stocked with monsters and treasure and clues and giant rats and copper pieces and all that cool stuff, and randomly choose one to be the one the players encounter. After that, I never use that encounter again."

    Wouldn't that fix what's on that particular hex once it's the first time it's explored? Every party that enters that hex afterwards would find what the first one found. You won't railroad the first ones to enter an hex, but all that will come behind. Maybe I'm missing something...

  4. At the very least, the type 4 encounter would give the DM a reason and/or a location to place another type 4 encounter in a static place. Thus requiring me to make more type 4 encounters.

  5. Exactly Gabriel. If the first party came across ruins in a certain hex, and a few months later another party came across the same hex, the ruins would have to be there. Not so for a monster encounter, there could be a different monster there this time, or nothing. But not another ruin. and if the hex was empty last time of a ruin, the next time a new group goes there, there wouldn't be a ruin there this time. It wouldn't make sense. It's not an everchanging island, it's just a place where the players and to a certain extent the dm discover what is where at the same time.

  6. You might also try prepopulating at least a few. I don't know how many hexes you expect to explore each session, but if you were to have perhaps a two-hex path in each direction identified then you should be able to cover most player options.

    This gives you some ability to predict what the players can encounter in each direction, which lets them have a hope of meaningful decisions. For instance, a village in this hex and ogres next door gives you an immediate "we have an ogre problem" (or if you find the ogres first, loot taken from the village they just raided) situation.

  7. There's the other thing, I've yet to figure out the scale of this beast, on a hex by hex basis. 100 miles x 100? 500 x 500? That would determine what each half inch hex represents and the plausibility of there being ruins missed the first time. At the same time, as a player I would feel ripped off if every time I went to a hex I had a chance to discover a new ruin. I never get a sense of completeness, of having finished anything, of having cleared the damn hex. It would be as if I did nothing last session, and that the island doesn't need to be more than 3 hexes big to fit in an ass-ton of stuff, since I could keep going back to the same hex over and over again. I know realistically there could be dozens, if not hundreds, or ruins and encounter areas per hex. But I would feel overwhelmed and ripped off as a player if I could never clear a hex. New random monsters in the same hex are one thing, but ruin after mysterious ruin would get to be annoying.

  8. With the predesigned encounters I could definitely pre-populate a few for the players so as to give them a chance to make meaningful decisions as to where to go next. I think you're right overall Keith, that this approach would require a mix of the two approaches, random and pre-populated. 

  9. I wonder if this similar post might help you with your questions of scale.

  10. You know, in addition to each hex being a random thing, you could have a series of interconnected hex regions which all interrelate, say a main hex and several surrounding it, as a random roll. That way when the party encounter something, they can make meaningful decisions as to what hex to go to next based on what they discover or encounter in the first hex.

  11. thats a good link man, thanks. similar to what i am doing.

  12. You know, in addition to each hex being a random thing, you could have a series of interconnected hex regions which all interrelate, say a main hex and several surrounding it, as a random roll.

    This sounds like a good idea. I was thinking of how you might synthesize this at the time (ogre+village hexes), but it's not unreasonable to have 'complexes' like that. They don't even have to be adjacent, just identify that, for example, certain hexes are a certain distance from others.

  13. I think you're getting somewhere with the area of hexes thing.

    Maybe it would be good to start thinking of Blood Island as a series of areas, each larger than, say, 5 to 10% of the available hexes. So, 10 to 20 areas total. In any one area, make there be a max and minimum number of things over types 1-4. For example, with monsters you could think about it ecologically, depending on the type of terrain, whether you have one or more aggressive predators, etc.

    As for the ruins and special categories, the hex areas hold the remains of specific types of civilizations, as well as clues to the "Big Picture" stuff about the nature of the island, and what have you.

    Finally, using areas rather than individual hexes as the way you thinking about hex-stocking, you can make the island make sense in a more concrete way. Things are in particular places because they're *supposed* to be there. I think having a better sense of the nature of things in a particular setting sets the table for enhanced player agency, because it allows them to "get" what the setting is about.

    Now, what you do in the individual hexes would depend on the area you're talking about. Terrain would need to make sense from hex to hex (and area to area). You could completely randomize monsters or put them in colonies based on area, that kind of thing.

    In any case, I think I'd get overwhelmed by that many hexes if I don't understand how they connect with each other in a larger sense. I'm guessing that players might also.

  14. Maybe each region has a random encounter list and some of the entries point you to another region's encounter list. You can probably start with the starting region and any nearby surrounding regions and then flesh out other regions as the PCs close in on them.

  15. This also seems to be similar to what Arneson might have done, according to D. H. Boggs:

  16. I did something nearly identical a few years back for my daughters. I created a map, and what I had at that time labeled "zones". This was all in response to the West Marches campaign that was so popular a while ago.

    The zones were defined generally by topography, but I had a few in larger areas (plains come to mind immediately) that were split into zones for no immediately defined reasons. (Later I had a randomizer that I came up w/ that was "themed"...with things like "black lightening storms from space plague this area", "ancient city state", etc.)

    And then I had created / borrowed / liberated all sorts of random charts that allowed me to describe the zone on a large scale, and then populate it on a detailed level. (3 mi hexes)

    Anyway, it worked really well, and allowed a lot of really interesting stories to be generated as they adventured. It REALLY kept me interested because I never knew what was going to pop up, and many of the resultant combinations were totally wacky and fun to play.

    I heartily encourage your endeavor and would love to see what you come up with.

  17. I put together a big post on the question of scale yesterday, but blogger ate it.

    Anyway, I just set up a random hexmap and ran into the issue of scale. I wanted to compare to real-world places so that I would have some idea of the range of terrain types I should include.

    Rule of thumb that I found is: figure out how far across each hex is (I picked 10 miles). Square that number, then multiply by .93 - that determines how many square miles each hex covers. Multiply that value by the number of hexes in your map, and you have actual area for the map.

    Wikipedia pages for Countries and states include their area in square miles or KMs.

    That meant that my 40x40 hex map, with 10-mile hexes, covered an area of about 150,000 square miles, or the size of Germany. Not bad for a "little" map.

  18. incidentally, I discovered that hexes 40 miles on a side (about 35 miles center to center the short way) is not that far off 1000 square miles (by Jeremy's formula, 1116 square, more or less). This is close enough to me to be considered 1000 square miles, and is a lovely size for demesne-level play (I used to use 'domain', but that gets confusing in D&D 3.x).

    It also happens to be roughly 36 six-mile hexes, so you get some nice congruency there.


Note: Only a member of this blog may post a comment.