Gold Canal Sources

My entry for 2019’s One Page Dungeon Contest, a high security prison sort of thing named The Gold Canals of Irid’s Vault was named one of the thirty winners.

As per past tradition here, winning means sharing the sources. Just like the final project, these are released under a Creative Commons Attribution Share-Alike license. The source files in the archive include:

  • The 3d model used to block out the basic shapes (Cheetah3d .jas format) and some renders thereof (png)
  • The flat, grid-based floor maps, which got transformed into isometric maps (Illustrator)
  • The resulting isometric sources (Illustrator)
  • The composite drawing with the maps and text (Illustrator)

Download here: (23MB)

Murky Dealings

Dealings mapI am entering Murky Dealings into the One Page Dungeon Contest 2014, even though it turned out a lot different than I had original intended it. As with my prior entries, I’ve left specific details deliberately vague. When I actually use a published dungeon (one page or otherwise), I usually only make use of the skeleton anyway, and replace the rest with campaign-specific stuff. I assume everyone else does the same thing, so just try to provide a feel to a place and let the reader fill in the blanks with stuff specific to their own game. (In the past, this approach has irritated some judges, but so be it.)

I also think this is more in keeping with the system neutrality which supposedly governs the contest, making it easier to, for example, use the map in a sci-fi game. Settings which don’t allow flying will make this dungeon significantly more difficult (would work well as a final challenge for a modern tomb raiding type game, for example). Another variation is to play with just how dark is “dark”. If you ran it in the Ptolus setting’s Utterdark, for example, everyone would basically be totally blind between the rooms. Even if it is normal darkness, the “see in the dark” magic of most fantasy systems will (intentionally) only be able to see one or two rooms nearest to the one you are in, sometimes none.

I continue to use vector software for maps, though this one also made use of a 3d modelling system to get the shapes right. This dungeon is significantly shorter than what I’ve done in the past, and better suited to a single night’s play. Like my 2012 entry, this is also another attempt to build a dungeon that is not a directional graph.

Seven Spindles… named a winner in One Page Dungeon 2012 contest

Seven Spindles and a McGuffin was named one of the 24 winners of the One Page Dungeon Contest 2012. I appreciate the honor, even more so than last year, because my entry this year wasn’t as good and the competition really kicked up a notch. Thank goodness they named more winners this time around!

At the time of this post, the main site for the contest seems to have fallen down, but there is a PDF containing all the winners on a different site. (And this guy reviewed them all, mentioning this entry as “bereft of flavor or charm”. Hoo-rah!) Once the site comes back up, though, check out all of the entries (over a hundred, I think). People really brought it this year.

In honor of this win, I’m releasing the source files for this dungeon (under the same CC licence as the PDF). This zip file contains one Adobe Illustrator file with the dungeon map and one InDesign file that links to the map and contains the text and other stuff. If you use these sources for something, drop me a link to the results in the comments.

Seven Spindles and a McGuffin

Seven Spindles and a McGuffin (pdf link) is my entry into the One Page Dungeon Contest 2012. Much like last year’s entry (which was one of the winners) this one is more about dungeon structure than compelling narrative. In fact, the “story” is so minimal in this one that the whole entry is really more of a sandbox than an adventure. I sort of like dungeons that way, personally, but mileage varies.

I had some some goals for the design of this dungeon (and some observations after building it):

  • Must use a vector-based map. (Why? Well, note that this map is infinitely scalable, but the whole PDF is under 300KB.)
  • Wanted a dungeon with a couple totally different vectors of entry.
  • Provide the flow control usually supplied by different levels of a dungeon, all using a single map. The center spindle effectively allows this to be a six level dungeon on half a page.
  • Take a regional approach, where the dungeon is described by section rather that detailing each room.
  • Put some rooms at angles to the grid, but showing the grid in their own frame of reference.
  • Subvert the idea of dungeon as node graph idea from last year’s entry. It may be possible to node graph this dungeon, but I’m not entirely sure how to do it, especially since the small spindles are placed at random each time the dungeon is run.
  • Minimize hallways (if you were digging your own underground complex, you don’t get much return out of the labor). Some sections do this better than others.
  • Because each of the spindles serves eight potential openings, it turns out to be critical that the spindles turn in 45° increments. If they turned in 90° increments, I’m pretty sure that you can get situations where randomly placing the spindles results in unreachable rooms. Turning at 45° increments avoids this, though you might get cases where some rooms can only be reached if you stay in the spindle when it rotates.
  • Use psychology against the delvers. For example, in a couple of places, there are short hallways with a normal door on one end and a secret door on the other. The normal door is in the “more secret” area, so the idea is that if the delvers are already in the secret area and go through the normal door, when they find the (obvious) secret door at the other end, their tendency will be to go through it (“it’s secret, it must be protecting something good”), which actually leads them out of the secret area. Not sure if it would shake out like that in play, but that’s the idea.
  • Loved the idea of the spindles periodically sealing and unsealing sections, so that air, water and such rush in or out when the spindle moves. Like, if you are in a room with water up to your ankles and, meanwhile, the tide has risen outside, then the spindle turns and the high tide rushes into the room you’re in. Probably should have done more with that notion, but it is a) tough to do in one page and b) hard to explain and use.
  • The overlapping technique used in the fissure section, where one room is on top of another with a ladder between them, could have been used more. I thought it might confuse people, even though it is a bit easier to illustrate with the rooms at angles to each other.

I think last year’s entry was stronger, but when I went to the well this time, this is what came out.

Escape… named a winner in One Page Dungeon 2011 contest

Escape From the Lost Laboratories was named one of the fifteen winners of the One Page Dungeon Contest 2011. No idea what the prize will be, yet, but it’s an honor to be named.

The contest page offers a PDF containing all the winners, but also take a look at some of the other 70+ entries. I loved the idea behind Mystery of Godzina House, for example.

In honor of this event, I will be setting up DivNull Lark aimed at giving the Lost Laboratories a bit more more flesh, and some love to open source systems. Stay tuned.

One page dungeon contest entry, 2011

Escape From the Lost Laboratories (pdf link) is my entry into the One Page Dungeon Contest 2011. It is, perhaps, a bit more free form in its room descriptions than some one-page dungeons. Since the rules specify that the dungeon needs to be system agnostic, I tried to give just enough detail that readers would think “Ooo… I bet this is how you’d represent that in [insert game of choice]”, but not so much that two different people would do it the same way.

When reading through it, it should be pretty obvious that the experience your players have will be highly dependent on their access to teleportation magic. The title assumes they don’t have any, so that their only choice at escape is to work through the dungeon. If, however, they can teleport on their own, the focus of the whole adventure changes pretty drastically, shifting to more about exploration and investigation of what the place is and how the players might use it for themselves. Some parties might get sucked into the network, take a quick peek, teleport out, and never think about the place again. Others might do serious exploring and be more interested in finding out about its builders, and so on. (If tamed, it would make a pretty great “home base”, for example.)

I am open to feedback on this dungeon, so post it if you have any. Though I have already submitted this entry, the contest allows resubmission with updated versions, so I have until the end of March. Also, feel free to post how you’d flesh out the rooms for a specific system. If anyone is really clamoring for it, maybe I’ll build a Pathfinder version once the contest is over, with MapTool maps and such.

At a more “meta” level, this dungeon makes explicit the notion that all dungeon maps are really just directed graphs. The graph is plain to see in my entry:

One of the reasons dungeons remain popular is that their directed graphs not only represent the geography of the dungeon, but also the flow of the narrative. The dungeon is a tool to control the pacing and sequence of the story and place payouts such that reaching them requires certain challenges. You can see some explicit examples of this in a thread on this topic at Story Games. This thread has a number of opinions on what traits these graphs need to make dungeons “fun”. I also provide side by side comparisons of the maps of some classic dungeons (Tomb of Horrors, Keep on the Borderlands, White Plume Mountain) with their directed graph representations. For example, here is the graph of White Plume Mountain, which clearly shows its “three silo” design:

White Plume Mountain directed graph White Plume Mountain map

As a quick example of how graph analysis can be used, take a look at room 2. The graph makes in painfully obvious that the adventurers will be moving in and out of that room multiple times. So, it would be good to make that room memorable somehow or, perhaps, contain some kind of trap that needs to be dealt with each time through the room. The revised version of White Plume delivers here (see PDF at link above), with a challenge that is similar general each time through, but still different enough in the specifics that it doesn’t get annoying.

One thing I don’t spell out in that thread is exactly how I built these graphs. The key is a program called dot, which is part of the GraphViz package. Once you get that installed, you need to make a .dot file to represent the dungeon. This is just a text file that follows a specific format. These can get complicated, but for turning dungeons into graphs, we only use a really small portion of the format. It starts with a shell definition that defines the file as a directed graph, and contains a single graph attribute for spacing out the nodes:

digraph G {
	ranksep="0.4 equally";

Let’s use a smaller example from one of last year’s One Page Dungeon contest winners: map three from Antti Hulkkonen’s “Den of Villainy!” (reproduced to the right through the magic of the Creative Commons license). At each room on a the map, enter one line for each connection the room has. For example, room 1 on the map leads to rooms 2 and 9. So, one line is "1" -> "2" and the other is "1" -> "9". Do that for each room. Then save the file and open it with graphviz (or open a command line and run dot on it). That’s the basics of it. You can also add some embellishments to each line, like using a different color for secret doors and so on. The result might look like this:

digraph G {
	ranksep="0.4 equally";

	"8" -> "3" [dir="both"];
	"1" -> "2" [dir="both"];
	"1" -> "9" [dir="both", color="blue", style="dashed"];
	"2" -> "3" [dir="both"];
	"2" -> "4" [dir="both"];
	"3" -> "5" [dir="both"];
	"3" -> "6" [dir="both"];
	"4" -> "5" [dir="both"];
	"5" -> "13" [dir="both", color="blue", style="dashed"];
	"6" -> "7" [dir="both", color="blue", style="dashed"];
	"9" -> "10" [dir="both"];
	"9" -> "12" [dir="both"];
	"10" -> "11" [dir="both"];
	"11" -> "12" [dir="both"];
	"11" -> "13" [dir="both", color="blue", style="dashed"];
	/* Mark the entrance rooms */
	"1" [shape=Mdiamond];
	"8" [shape=Mdiamond];

The graphviz package will handle the layout of the nodes automatically. Sometimes it does a better job of this than others (and there are tricks you can use to make the output better), but as these graphs are just to visualize the dungeon, they don’t need to be perfect. In this case, the result looks like this:

The flow of the dungeon becomes more clear in this graph, with the one whole branch only accessible through secret doors, a main interconnected section, and two key secret rooms (which, significantly, are furthest away from the entry ponts). All very rational.