The Player-Built Dungeon

This is related to the Pre-Mapped Dungeon, so if you want you can read my thoughts on that first.

A few weeks ago I gave my good buddy a bunch of tools I made to help in DMing and asked him to use them to run me as a player.  I wanted to see how they might work for someone else.  Another big reason I wanted to do it was to see what it would be like to experience a Dwarven Outpost.  Would it feel different to run through a dungeon that I had made myself?

It was.  It was cool.  My hireling died in the first room to a panda-headed crab-thing and I desperately searched the place for what I knew I would recognize as the treasure corridor.  Once I found it, I was very cautious about the traps I knew to be there.  And once I'd found the weird treasure he'd put in there, I beat a hasty retreat knowing I had most likely found the best the place had to offer.

Because of the last few posts I got to thinking what if players could have that same experience-- not just of a type of dungeon they can become familiar with by encountering them several times, which the outpost kit was meant to facilitate-- but of a dungeon they knew well because they made it themselves?

Now, I don't think it would work to say "design a dungeon and we'll run through it" or even to take the more modern, indie route of "let's design a dungeon together that will be fun to run through."  I think the DM has this role because the fun of exploration requires not knowing what is behind the next door (and also having a single creator probably gives a place a more consistent tone and logic). 

But maybe what we could say is "draw up the manor house of your ancestors" or "make a map of the urban sewers you grew up in."  Then the DM can take that map and apply decay, add monsters, and traps left by the waves of inhabitants that have been there since the character left.  And if the party visits that location the player who drew the map would get a little extra spotlight that session: "The secret entrance should be just past the stables, but the stables appear to be gone . . ."

Now, my experience of uncertain familiarity with the dungeon worked in part for me because 1) it had been months since I made the outpost kit and it wasn't fresh in my mind and 2) the tetramorph aspect of it let it be shuffled around a bit.

So what might work here is to have players make something well in advance of them experiencing it.  I'm not sure about the tetramorphs bit.  It would make "familiar with, but not sure about" work better but it feels like a lot to pawn off on a player.  I suppose if you had enough players with magic-user characters you could ask each of them to decide on a room a typical Mages Guild would have.

Hmm, or maybe I could just design a set of tetramorphs stencils for each class, say typical sewers for thieves, typical church catacombs for clerics, and then give each out to players who choose those classes.  They could even have them in hand as they explore "By the shape of this room I think were are in the central junction."

You would think that high mortality games could cause a problem.  That having a player draw a map of a guild hall and then having that player's character die immediately would make the map a waste of effort.  But the goal here is familiarity for the player, not necessarily rigid narrative logic for the characters.  Heck, you could just say that player's new character had heard tales of such a place.

So, maybe I should have really titled this post The Familiar-to-the-Player Dungeon.

Mini-Hex Stencils

I thought, "Hey, what if I turned my hexagonal geomorphs into stencils?"  So I did.  Most of the work was already done.  I just shrank them, edited a bit, and cut them out to test.  They worked okay.  I figured something cut by hand would be a little off and they were.  But still a fun idea to play around with.  Here is what I ended up with:

I cut the inner parts first, then the hexes:

I used whiteout to label the hexes and rolled to place them back in the template:

I don't know if it was just my shoddy job or if there is something different about shifting hexes, but the 3D definitely seems off in my test map.

Dwarven Outpost Kit

I was kinda wanting to enter something in this year's One-Page Dungeon contest, but the visual dungeon I had in mind has been requiring me to learn more to even be able to make it.  When I saw Roger's cool outdoor location he entered, though, it inspired me and I really wanted to submit something.  I made a push, but after working many hours yesterday and today I just didn't make it in time.  It wasn't a stocked dungeon anyway, and I still get to share it with you.  So onward.  Let's recap a little:

• To justify using a stencil it needs to do something extra for us or it would be faster and easier to just draw a dungeon by hand.
• I think a few things that stencils might help with are hard to draw shapes like perfect ovals and triangles and such, repetitive structures, and my latest discovery-- adding depth with a pseudo-isometric view.
• Keep in mind stamps and linocuts like here.  I think they would function similarly but be easier to use, while being harder to make.

I tried to combine all these ideas into a cool tool.  It still takes some time to use; it isn't for use at the table, really.  But I'm hoping it could make for some interesting location based explorations missions.  Here is what I came up with:

And here are some pics of me eatin' my own dogfood:

I used a discarded report cover that had a pretty tough but flexible black plastic back.

After the surgery.  Yeah, the fish ponds gave me some trouble.  The good thing is these should last a while so you only have to cut them out once.

Here's me trying them out.  I didn't finish the map, but you can see I added a hallway linking the Barracks to the Smithy.  I tried randomizing, but I think the best way to use these is just arrange one of each tetramorph to taste.  You could put more space between them and intersperse other rooms if you want.  Anyway, let me know if any of you try it out.

The Procedural Dungeon

Take the idea of index card geomorphs, geomorphs that are stencils, dungeons built by certain cultures, and I think you've got a recipe for quick, DIY dungeons with a sense of logic and history.

First, think of a culture that's left structures peppered around your game world.  Let's pick Dwarves.  Second, decide on some features their strongholds/outposts almost always have.  Let's say:

• cave fish pond
• barracks
• smithy
• ore storage
• smelting room with chimney
• throne room
• hidden gem storage
• secret emergency exit/bolt route

Now, decide on the most common design for each of these features-- and they can be tetronimo shaped as long as each fits on a single index card-- and cut them into stencils. I realize that step might be trickier than it sounds but am confident the gross features can be caught even if you have to hand draw in finer details.

Then you should be set for the next time players go off map or an encounter roll calls for creatures to be in lair.  You pull out your stack of stencils for Dwarven Outpost (you can keep different types bundled with rubberbands) roll dice, or shuffle and draw the cards, then trace them on your graph paper.  It might take a few minutes but you'll have a consistent dungeon with a map for your campaign folder.

If it works as I imagine players could learn things about these dungeons that would add a sense of verisimilitude to the imagined world:  "Wait, this looks to be a Dwarven outpost, they almost always have a secret gem room."  Or "These Dwarven outposts tend to have smelting rooms with chimneys, so we might find a small but definite exit to the surface there."

If the stencils work as I hope, the next design challenge would be to make sure all of your recurring dungeons have features that players would find interesting, like the examples above.  maybe cultists have libraries, outposts of the old magic-rich empire always have a brass head mounted somewhere, and tombs of the old empire tend to have map rooms with a diorama display of the surrounding countryside (and the location of more tombs).

(I plan to try to actually produce some of these but I'm currently house sitting for friends so it may be a while.)

Update: I had a hard time titling this, Procedural isn't right.  I think I probably should have called it Template dungeon (but that sounded kind of boring) . Oh, well. That makes me wonder what a true DIY procdeural dungeon would look like, too.

Modular Danger Room

I love pictograms and symbols and trying to convey meaning in its simplest form.  And I always loved the idea of the danger room for heroes, because, well it's a lot like exploring a dungeon-- you don't know what to expect.  Also, you can play it solo.

I think Zak posted a random danger room generator in the last few months.  I know I made one for DC Heroes years ago, but I'd have to dig it out because I don't remember anything about that attempt.  Anyway, what if you took the hexagonal geomorphs I've been posting, throw in a plethora of cool hazard symbols:

and make a modular danger room:

This time when a trigger fires, heroes have a specific hazard to negotiate.  Lasers, poison gas, acid, magnetic fields.  I know it should probably be a bit more specific than the damaging energy type for the danger room hazard, but maybe you could devise symbols to represent turrets, walls, robots and the like.  Instead of treasure heroes might need to get flags or such.

Hexagonal Geomorph VI

Looks like some family visitors are going to postpone my screenprinting trials.  Frustrating, I'm afraid if I don't get this done in the next two weeks or so I might lay it by for a long time again (last time was two years).

Oh well, the good news is that all the planning and making these by hand made for churning digital versions out quite quickly.  I haven't even printed/cut/tested all these out yet.  Here are two more:

Which gets us to 12.  We have a funny shaped die to handle that.

One thing you might notice is that I am sticking to hexes that have to 3-6 sides used as viable exits.  This is for two reasons 1) aesthetically I didn't like the look of stranded island of dungeon, the little pockets that an unreachable one-side-accessible path would make, and 2) I was never able to figure out mathematically the increase in probability of ending the catacomb that each dead-end added to the tiles made, so I'm just going by intuiton here until I playtest these more to see how big a catacomb can get before you hit all deadends.

Another thought, I am torn between having a small set of pretty re-usable tiles and interesting, individualistic tiles.  For example tile 11 above could have a little room to the right instead of just the two loculi.

The problem is, the more unique you make one tile, the less useful it is a a generic part of a dungeon generator.  It becomes very recognizable instead of a something unknown to be explored.  I also have to remember these were meant to be used on the kitchen table, so you can't have too many.  Though, if you have a big selection you could choose your own "deck" before play, so to speak.

Hexagonal Geomorph V

These two are for Ohio Metal Militia.  It brings us to ten tiles which is quite convenient:

Hexagonal Geomorph IV

Okay, once I got more familiar with the program and got my layers all set up, the rest of the tiles rolled out pretty easily.  Here are tiles 4-8:

And here is a pdf with all 8 tiles if that's more convenient for you than fooling with the pngs.

My intent is to have 4 printed tiles with these on them back to back.  You can print them back to back for the same effect or just use a d8 to try them out.  You can use hex paper to record the catacomb you generate.  Just write Tile #, Face # with a little tick showing where the face connected.

I apologize if any of the triggers are wacky, I was just pulling these out of my derrière to get the prototype done.  I'd be happy to hear any feedback about how these work out for you, even if you ignore the whole trigger thing and just generate some catacombs.

Update: I fixed a couple little glitches.  And, you know what, here have the svgs too.  That way you can hide the layer with the triggers if they bug you or apply filters to make the walls more rough looking or whatever.

Update 2: Whoops, I had Tile 7 with a Even trigger and it has no even entrances. I fixed the png.   I'll have to fix that in the pdf later, I'm away from home now.

Hexagonal Geomorph III

Okay, I felt myself getting bogged down with trying to make the walls perfect, so I stopped that.  I went back to a more abstract, game-boardy look for now so I can just get enough done to play around with them (I know myself).

So, here is Tile 3:

and I redid the first two as well:

And here is a key to the trigger symbols:

Traps function like pits in old school D&D; 1-2 on a d6 and its triggered.  But the solo player can see them and decides whether to risk them.  Eventually I should make a little chart of trap type (didn't I do this in a post once?).

For the rest of the triggers, take the number of the face you exit a tile, and the number of the face you're entering and mash them together.  If that number is odd, the odd triggers fire, if it is something like 55, double triggers fire and double odd triggers.

Specials are rarer events, they might affect a whole tile like a cave in, or indicate that it's web-filled or something.  I will need to make a chart.   Actually, I eventually want specials to be unique tiles, like chapels, or large ponds or something.  But I mess with that only after I have enough standard tiles done.

Triggers only fire once, when you first enter a tile.  All triggers fire at once, so if you got two encounters . . . well, maybe they'll fight each other.  I know knowing where the treasure and monsters everywhere on the tile are is not realistic, but it seems the most streamlined way to do this so someone won't game the system.  I'm guessing you could easily graft on more complicated systems if you wanted.

Hexagonal Geomorph II

Here is another of the catacomb geomorphs.  I thickened the walls a bit.  I was initially thinking of keeping things as simple as possible and saving ink for those of you printing at home, but I added a gradient to try and give the solid stone some heft.   I don't know how this gradient will screenprint, I do have pretty fine screen.  I'll have to see. 

I don't know, looking at it as I post the pic, the gradient looks awful "computery," too clean and regular.  I wish I could get some kind of spatter going.  Why does it seem like this would be so much easier by hand?

Hexagonal Geomorph

I haven't been making as much progress as I wanted to.  Some bad news: somebody broke into my car and inexplicably stole my little screenprinting screen I made 2 years ago.  I was planning on making 2 bigger new screens anyway, but it would have been nice to have it to do little stuff and test things with.

Some good news: I just got three new ink colors in the mail yesterday.  Now I've got black and all 4 browns the company offers.

I have been trying to re-learn what I knew about Inkscape 2 years ago.  Spent several hours today.  Here is an idea of what I'm shooting for:

The grays will be a light brown, the blacks a dark brown, printed on canvas.  The game boardy spaces are meant to simplify movement for solo play.  I'm thinking 5' per space and per inch.  That would mean a 60' light source would just show you the tile-length as you enter it.

My question for you: do you like the way the walls look?  (I fractalized the lines and then distorted the result with "torn edge") I wonder if I should add some hatching.  I'm not really an artist and working with Inkscape is difficult for me, but I am determined to forge ahead.

Terrain Generation by Hex Geo

This feels so familiar I may have seen it somewhere else.  Oh well, maybe not.  If you set each side of a hexagonal geomorph to a terrain type and then make a little chart so that it is more likely for similar hex faces to be rolled, you should be able to randomly produce a terrain that has some logic to it.

Here is a draft chart for rolling 2d6:
2     -4 from current hex face
3     -3
4     -2
5-6  -1
7     Same hex face
8-9 +1
10  +2
11  +3
12  +4 to current hex face

So, for example, water would be much more likely to be found next to more water or terrains that are common near water.  Here is a sloppy hex I threw together:

1. Forest
2. Grass
3. Water
4. Crops
5. Rocks/Mountain
6. Hills/Rough

I probably should have switched 5 and 6 around because trees on a mountain make more sense than crops.  But . . . a computer could generate a random terrain with more sophisticated algorithms much more quickly.

Hex Geo Template

Here is a template for a hexagonal geomorph.  It isn't anything you couldn't do by hand if you followed my last post.  But maybe it will save you some time.

These are 6" face-to-face and I've marked the middle inch on each side.  I also put numbers on the faces for your randomizing needs. and a little line to label each tiles with letter/number.

And then I had to go through 400 hundred contortions to find the file type/printing options that maintained my images size, bizarre.  My printer refused to even print a PDF made with LibreOffice.  And it looks like something I or Inkscape is doing in non-standard in the svg because Firefox renders the numbers in the wrong locations.  Anyway, I'm hoping this PDF made directly from Inkscape will work out for y'all.  If you have problems let me know.

DIY, Analog, Hexagonal Geomorphs!

Making crap is fun.  Have a Saturday afternoon free?  International apocalypse bring down the internet? Make some geomorphs.

All you need is a compass, a ruler and some paper.  If you don't have those you can get by with two pencils and some tape.  (Seriously, that's what I just did in the mountains)

Okie Compass

Here, I'll walk you step by step toward geomorph goodness:

1) Draw a line that will be the length of your hexagon from pointy tip to pointy tip.  (I'll help you out here, if you want hexes that are 6" from flat side to flat side, they'll need to be 7" from tip to tip.)

2) Find the middle of that line (3.5")

This image is a little off,
line shouldn't be that long

3) put your compass point at that spot and set it so the pencil is just at the end of the line.  Scribe a circle.

4) Move your compass to the end of the line and scribe a half circle.  Where this crosses the first circle is two vertices of your hexagon.

5) Put your compass on the other end of your line and scribe another half circle.  This, and the ends of the line itself, will give you all 6 points of your hex.

6) Use a straight edge to connect your vertices.

Now, here is a beautiful animated gif by Aldoaldoz that will demonstrate this more clearly then all those words I just wrote:

There are a few more steps, though.

7) As you probably know, geomorphs only function if they line up correctly so we need to find the center of these faces.  This step will change depending on how large of a connection you want to make between tiles.  If you followed my suggestion above, your hexagon is 6" face-to-face and each face is 3.5" long.  I decided on a 1" connection just for the ease of using minis, but you could be perfectly symmetrical if you want.  So, for a 1" connection measure 1.25" over and mark a line on your face.  Then make a mark an inch from that.

8) Repeat this for each face.

9) Now you should have a nice, regular hex of just the right size.  Use this as your model.  Go to a sunny window and use it as a lightbox; place your model on the window, paper over it, and trace. If you want to make tiles out of something opaque, trace the first hex onto your new material, then cut.

10) Draw cool geomorphs on the hundreds of tiles you just traced and cut.  (Haha, you didn't think I'd do everything for you did you :)

 I like the way hexes feel like you can go in almost any direction.  They seem suitable for small scale paths and twisty passages because of this.  But that doesn't mean you couldn't draw a normal scale dungeon with lots of rooms per tile.  You might want to do this process on graph paper, though I like the idea of just giving yourself the freedom to draw without guides.

I also like the idea to try to use these to make a closer scale wilderness setting, with streams and paths that connect.

Any questions or ideas?

My Grand Project II

Here is a rough mock-up of how the hexes might work in play. You start at the red spot, maybe you've lowered yourself down a hole on a rope. Now, you can choose one of four directions to travel. You decide to head East. You reach the edge of the tile and need to determine what the next tile will be. So you roll 1d6 for which tile, and get a 2. Then you roll 1d6 for which side of tile two and roll 6. So you spin tile two around until the sixth side aligns with where you're at.

Now as you first enter this tile you take note of the numbers-- leaving side 2 to side 6 gives you 26. Uh oh, the first loculus has a symbol for monster and the "E" means any entrance to this tile with and even number triggers this monster. Your 26 being even, means you've got something to deal with. Maybe it's a skeleton, and you smash it with your trusty club. There is nothing more of note around so you need to decide which direction to go from here. You decide North West.

So, you go through the same process, roll a 3 and a 2 and plop tile three down with it's second side aligned with your passage. (If you had rolled a one here, this would be a dead end) So your number on entering the hex is 12. And that loculus to your left has a gold symbol that triggers on "Even." Sweet, you found some treasure. Maybe it is an ivory teetotum that you hastily put in your bag. You notice the other loculus on this tile only triggers on "Odd," so you dodged a possible encounter here, your 12 doesn't trigger it.

You only trigger the encounters on a tile once, backtracking or circling around doesn't do anything. But imagine, way back on your entry tile rather than heading East, you had decided to head North East. Follow the blue arrows. When you generate the next tile your roll of 3 and 3, results in an identical placement of the same tile. The catacomb map is essentially the same. But because your entry number is 13, odd, you trigger the monster but don't trigger the gold. That's Lady Luck for you.

So it is complicated to devise, but I think once you get it, it would be very quick to play.

There are many things I still need to work out. In making this mock-up, for example, I had to come up with a symbol to represent treasure. A dollar sign wouldn't be appropriate for my European friends. So I settled on Sol just now, but maybe there are better symbols. What about Events, like cave ins, how would you symbolize an event? Also, how many loculi do I want on each hex, how many would be too sterile, how many would be too cluttered?

Once I at least have a solid draft I'm satisfied with. I think it would be cool to implement some kind of website where people could upload their own tile designs. To let you all design tile "decks" to share with each other with CC licenses or something. It's all sort of a dream right now, but one can dream, no?

My Grand Project

I spent hours of my youth rolling up characters, wizards and gnomes and thieves. Oh, how I longed to play them through some dungeon or explore tangled woods. I never seemed to be able to find people similarly inclined. I tried the random dungeon charts in the back of the 1e DMG, but they were so random as to be nonsensical to me. I played through every solo module and gamebook I could find, which were nice, but once you run through them, then what?

So about ten years ago I sat down to methodically devise a way to play solo D&D.

I had found a design textbook in the boxes of my father's college books out in our backyard shed. I was intrigued by the way it talked about combining the functional features of a product and this was the biggest influence on this project.

I took stock of all the things you needed to determine in play, e.g. monsters, treasure, hazards, and the map itself. I decided the map could determine some of this. [I think I'll need to post separately on the many iterations of prototypes I made, I just want to get something posted today.]

I had several breakthroughs and many setbacks. But eventually, my system worked out to being a 1) Catacomb, because I felt the smaller, more windy features would be easier to create than a typical dungeon, at least at first, and 2) It would be hexagonal, the largest regular polygon you can tile with, but also easy to use with the most common of dice, the d6.

Yes, yes, you might be thinking, there have been hexagonal geomorph systems before (there must have been, but I don't know of any personally [well, except the brief mention in the Dungeoneer's Survival Guide]). But what I think is cool about my system and may be useful to others is the way the map helps generate encounters.


Here's how: On the geomorph (that seems to be the term people would most understand, but is it a trademark?) above is a loculus, a place in the catacomb wall where a body has been sealed. Well, it was sealed, but that was centuries ago, perhaps it's been broken open and a spider lairs there now, perhaps the skeletal remains of its occupant still clutch a glowing dagger. In all the loculi of these geomorphs will be a two symbol code, the second designating Monster, Treasure and, maybe, Event. The first symbol is the key, and determines the likelihood of the second of triggering.

In play, you'll randomly determine the next hex and its location. For now, let's say you'll roll two differently colored d6. The first will indicate which of the six available hexes to lay down, and the second die will tell you which face aligns with the passage you are in. Those faces will generate a two digit number.

Now, back to our loculus, the first symbol will refer to the two digit number that your entrance to the hex generated. This symbol could reference Odd/Even, Double, Odd Double/Even Double, or a Specific Number.

If we have two tiles with all six sides accessible, the odds for the above categories work out to 50%, 16.7%, 8.3%, and 2.7%. This gives me, as a designer, a range of probabilities for things to trigger. Each tile will have different triggers, and each time you enter the tiles you'll be coming from different directions, so it will be impossible to game them, knowing which tiles have better treasure etc. It will be largely random.

Now the problem comes in, when the tiles don't have all 6 sides accessible. My example above cannot be entered from side 6 or 4. So the chances of an Even triggering are down to 25%, and the chances of Odd are up to 75%. Doubles, Specific Numbers are all affected. It would be easier to make the tiles all accessible on all six sides, but that makes the catacomb infinite, and I don't think we want that.

So, what this means, is I have a lot of work to do, hand choosing the triggers for each tile so that they trigger in appropriate ratios. First I have to finish making the tiles. And I'm having a bear of a time learning to make SVGs with Inkscape. Each step leads to new revisions. I've just decided, for example, that the hexes will be 6' across and that I will hand draw gameboard-like squares in the passages as above.

I have always intended to give the whole set of these away once I'm done, as downloadable SVGs (so you can edit them and resize them to your heart's content), but I think it would be cool if I could make something I might be able to sell in addition. Don't get me wrong, I don't think there is even a tiny market for something like this. But I guess I'm thinking of the product I would want if I weren't making it myself. I'd like to offer that to people and be able to cover costs.

Anyway, printing these tiles on paper will work, but I think having them printed on something more sturdy and durable would be even cooler. To that end, I began learning how to screenprint this summer! Who would have thought that all my thinking on this project with a little boost of excitement from the OSR would end up in me learning a completely different craft!?

Still learning, but I did produce a prototype that let's me know that this is possible. I can actually do this.


Disregard the hex grid there, I just wanted to see how fine a line the screenprint could produce. Fine indeed, it turns out. Also, the blemishes there resulted from me brushing on the emulsifier when I shouldn't have, so it will look better next time.