A TimerLast week was the first session of my Wednesday night Roll20/Discord group’s new B/X campaign (a continuation of a long-running AD&D campaign that went on hiatus last year). Overall, I think it was a successful revival of the game with both the players and myself quickly falling back into both our roles and the lore of the setting.

There was one aspect of the group’s typical play-style that I was hoping to pre-empt, though: the tendency for the players to over-analyze and debate decisions both large and small so that the balance of any given session is 90% preparing for action and 10% action. While this interest in intense discussion can be taken as a compliment — they’re clearly invested! — it can cause things to drag out unnecessarily.

In the past, I had tried the typical solutions to such issues: random encounter checks, careful tracking of time, and depletion of exploration resources. These were useful tools, but not complete solutions because of two factors: (1) situations often arose where it would not be realistic to interrupt the party’s planning with random events or deplete resources (i.e., the characters are in a secure location with hours/days to spare); (2) any one player could prolong debates with clarifying questions about the situation directed at the Dungeon Master (sometimes relevant, sometimes not) even when a consensus to act had been built.

My solution to the problem of moving things along when the party could realistically be dilly-dallying within the game’s fiction has been to implement strict time limits on decisions on a meta-game level: ten seconds during combat rounds, one minute during exploration turns, five minutes while camping or in a secure location. Failure to make their decisions within those limits means time passes regardless of whether their debates and questions could have “realistically” fit within the time available to the adventurers. After all, we’re playing a game that makes abstractions all the time and there is no reason this approach shouldn’t extend to time management.*

That’s well-and-good as far as it goes, but I also needed to preempt any one player using up the party’s precious time — after all, if individuals are asking me questions and declaring intentions during an exploration turn, there is no way everything is getting resolved within one minute! It was then that the genius of the traditional party caller role become clear to me: a person that was responsible for facilitating a quick discussion of intentions amongst the party members — reminding them of the time limits and reining in those who would use up that time on their character’s individual actions — and then conveying that succinctly to the Dungeon Master for resolution. In short, the caller acts as the editorial voice for the party’s collective Id so that decisions can be made within the time limits imposed by the game’s referee.

The benefits of the caller role are numerous. First, having one person speak for the group means that the whole adventuring party will not be punished for any one player’s use of time unless that player happens to be the caller (in which case the party probably needs a new caller). I can also affirm that, for the Dungeon Master in terms of workload, this system is great: I just need to track the passage of time and not worry as much about making sure no one player is hogging the spotlight during what should be group endeavors. If the group decides to rotate the caller roll, it also becomes an effective way of gently pushing more passive players to step forward and find their voice.

It’s a bit early to give a final verdict on the system since it was our group’s first time trying both time limits and having a caller. There were definitely times the other players forgot to channel their intentions and questions through the caller and times I let them do so. Things did move along at a faster clip, though, and we will almost certain become better at this with practice. I’ll report back on this experiment after a few more sessions when it is clear whether this is a sustainable solution to issue of time management in game sessions.

* I’m not a complete monster, though. My time management rules allow for one thirty-minute discussion each game session without penalty for the purposes of detailed planning or debate during camping/party downtime.


An Ancient Dwarven Space Station Explodes

A scene somewhat similar to the explosive conclusion of the previous campaign

After a hiatus of more than a year and a half, my regular Wednesday night RPG group is returning to the campaign world in which they spent almost five years adventuring. During that time, they explored at least three mega-dungeons (depending on your criteria), had substantive interactions with nearly one thousand NPCs, thwarted at least four intricate conspiracies involving criminal enteprises and antagonistic polities, and finished things off by traveling to space where they defeated an ancient AI before it could open a stargate to unleash an armada of resource-hungry space dwarves. It would be accurate to say that I feel a bit of pressure to deliver an experience worthy of that legacy.

Fortunately, a combination of the choices made at the end of the previous campaign and in my work preparing for this continuation should help these new adventures feel like their own story rather than some kind of nostalgic echo of an epic that had reached its natural conclusion:

  • Within the campaign, the players’ solution to the main threat against their world — that AI-controlled space station and the imminent invasion by space dwarves — introduced a new problem on the same scale that is clearly their characters’ fault. To summarize quickly, the sentient sword that they carried for hundreds of sessions manipulated them into replacing the dwarven AI with Asmodeus, Lord of the DVLs (Dedicated Virtual Lifeforms). Asmodeus then preceded to use the stargate to open a portal to hell and unleash his hordes. Meanwhile, since DVLs powered just about all known magic, spells and enchanted items ceased functioning. The last decision the players made was to begin a search for an alternate source of magic to combat the evil their actions had brought into the world.
  • Outside the campaign, I decided that it would be interesting to try the B/X rules as reformulated in the Old School Essentials publications in place of the original campaign’s AD&D 1E ruleset. All the existing player characters were converted to the new system.

These two factors mean that this new campaign should be able to manage a full reset in terms of both content and form without losing the player buy-in established during their previous exploits. There’s obviously a clear goal to inspire and focus the returning characters — to fix what they broke — but there’s no obvious solution to that problem that would throttle player agency. Meanwhile, the sudden change to the in-game rules of the campaign world is echoed by the exterior ruleset change: the loss of magic is reinforced by the decrease in the power-level represented by the move from AD&D to B/X — it’s a brave new world in both realms!

Tonight is the first session of this new campaign, so it will be clear very soon if these factors help with the transition back into the Curabel campaign setting. For those interested in following these new adventures, check out the background documents on my campaign wiki — soon to be joined, I hope, by session summaries written by my players.

Temple Ruins

Temple Ruins – One of the Original Inspiration Images for the Curabel Setting

My plan to post my player’s session summaries with commentary hasn’t progressed much in the last few years. However, I have managed to put together a mediaWiki installation that includes those summaries — all but five out of two hundred and twenty-one! — along with much of the other campaign materials that were originally hosted on our group’s G+ community page. Here are the links to the relevant pages on the wiki, along with a quick description of what you can find there:

I hope to add the Excel spreadsheet of NPCs created by the players shortly (the DM version will need to remain secret a bit longer).

Readers might also be interested in the documents hosted on the wiki related to my group’s recently concluded Stars Without Number campaign and our upcoming return to the Curabel setting for a Old School Essentials (B/X D&D) campaign that I’m calling “Graveyard of Empires: After the Fall.” I will have more to say about both of those campaigns, especially the latter, in the coming months.

Glabrezu as illustrated in the AD&D 1E Monster Manual

Glabrezu as illustrated in the AD&D 1E Monster Manual

I have just uploaded a new Perl CGI script that generates a random demon using the rules from appendix D of the 1E AD&D Dungeon Masters Guide (page 194-5). Besides basic stats, with the usual information necessary for combat, this random generator also provides extensive descriptive information for the demons it creates with everything from style of head (e.g., duck-like, monkey-like, crocodilian) to body odor.

Considering the fact that demons are the very embodiment of chaos, the surprising combination of features created should keep players on their toes when tangling with abyssal forces!

Click here to give the Random AD&D Demon Generator a try!

It has been exactly one year since the last update here. I’m still alive (and healthy at the moment), still running a Stars without Number campaign that just reached sixty weekly sessions, and still working on my next OSR module to prepare it for publication.

There’s a slight chance I will be a bit more active here soon — the initial drafting and layout of the module is complete and I’m now considering how to handle the illustrations. I have high standards for artwork but no talent and little money, so aligning my expectations with those realities is slowing things down in terms of finalizing the adventure.

In the meantime, please continue to enjoy the various random generators here on the Dwarven Automata site; it’s a nice pick-me-up to glance at the usage statistics once in a blue moon and see just how many are making use of things like the weather generator tool. Thank you!

Just a quick post to let anyone following this blog know that it has not been abandoned completely. The Stars without Number campaign mentioned in the previous post has been progressing nicely for twenty sessions; the system has been easy to use and the challenge of putting together a sci-fi game (a first for me) has been refreshing. At the same time, work continues apace on my second adventure module. All the mapping, keying, adventure hooks, and faction information is done and just the random encounter tables, bestiary, and treasure description remain to do.

As usual, my goal in the coming months will be to post here more often. The realist in me, though, has a pretty good sense of the odds on that resolution being fulfilled.

Postojna Cave

Postojna Cave

The latest automated generator I have created for my AD&D 1E campaign is based on the tools for creating random caves and cave systems found in Patrick Stuart and Scrap Princess’s wonderful Veins of the Earth. Specifically, my script creates up to fifty random caves with indications of entrances, exits, distance between caves, etc. A DM can string these entries together to create a natural cave complex more-or-less on the fly.

My script is not a perfect recreation of the VotE system, though, and produces slightly more specific results in terms of cave dimensions and other features that I found necessary for my own use at the table. As Patrick Stuart described it when I shared this tool with him: “you’re kind a crazy because you’ve taken something specifically designed to work through the immediate intuitive spatial relationship of the dice as they are rolled in front of you and turned it into a raw-data readout, which is something I would never do, but, well people are odd and I’m glad you’re happy.”

If that sounds like something that would make you happy — or prove useful in your campaign — give it a try and let me know if you have any suggested improvements!


Sample Output (see generator page for explanation/details):

Cave #2
Width in Appropriate Units: 10″
Length in Appropriate Units: 17″
Height in Appropriate Units: 10″
Entrance Location: Roof
Largest Exit Location: West
Largest Exit Width Size: 7′
Largest Exit Height Size: 6′
Number of Other Exits (Each 1/2 Size of Previous): 5
Directions of Other Exits: Roof East Floor East Floor
Length of Exit Routes (Turns): 1 2 1 1 2 8


Mountains: More than a Speed-bump, less than Survivalist Porn?

Recently, my AD&D 1E campaign involved some travel and exploration within a mountain range, including scaling one particular peak in search of a dragon’s lair. My preparation suggested two possible approaches to mountaineering by-the-book: the dry application of a movement speed penalty (DMG) or a hyper-realistic treatment that assumes both a detailed area map and skill system that I don’t use (Wilderness Survival Guide). Faced with those extremes, I decided to put together my own system that offered a bit more than a speed bump but still adhered to the “keep it simple” ethos of early editions and the OSR. I make no claims to realism or balance, but it worked pretty well for a couple of sessions and can probably be modified to cover other “hazardous terrain” situations.

Mountain Terrain Assumptions: Movement and Time

  • Movement across mountainous terrain: 1/4 Speed
  • Climbing Mountains (Average): 4 hours to tree line; 6 hours from tree line to peak

Mountain Exploration Process

Below the Tree Line

  1. In addition to normal encounter checks, roll a d6 for each hour of the climb
  2. On a result of 6, the party encounters a special event; roll 2d6 on the event table below
  3. Roll another d6 to determine which turn within the hour the event occurs

Above the Tree Line

  1. In addition to normal encounter checks, roll d6 for each hour of the climb
  2. On a result of 5 or 6, the party encounters a special event; roll 2d6 on the event table below
  3. Roll another d6 to determine which turn within the hour the event occurs
  4. Roll another d6 to determine if the special event coincides with an encounter

Special Event Table

2 Rock slide! All party members save vs. paralyzation or tumble 10′ (d6 damage); save again or continue tumbling d4x10′ (4d6 damage); repeat until dead or successful save.
3 Path split by chasm with no obvious means of crossing; width of chasm is 4d6x10′ (add another d6 for every result of 6); depth of chasm is 8d6x10′ (add another d6 for every result of 6)
4 Path split by chasm with weathered bridge (20% chance it is a rusted automaton collapsed to form bridge); roll d6 for each character crossing — on the second result of 6, save vs. paralyzation or fall unless secured (see above for width/depth of chasm)
5 Path obliterated by rock slide creating treacherous terrain; d6 for each character crossing and save vs. paralyzation for all fails or tumble 10′ (d6 damage); save again or continue tumbling d4x10′ (4d6 damage), etc.
6 Sheer wall impedes progress; thief or grappling hook needed to scale with rope required for others to follow (height of face: 4d6x10′, rolling again for each 6 result); if no means of ascent, an hour detour to find alternative route.
7 Sheer wall with overhang impedes progress; thief or grappling hook to scale (thief skill at -15%) and rope needed for others to follow (see above for height of face); if no means of ascent, an hour detour to find alternative route.
8 Cave with 60% chance of lair (roll on random encounter table immediately); 20% chance that cave is a shortcut to higher/lower area, cutting off one hour of travel.
9 Drop-off next to path leaving only a narrow ledge; roll d6 for each character passing and on third result of 6, save vs. paralyzation or lose footing; fall is 8d6x10′, rolling again for each result of 6.
10 Path split by swift-moving stream cascading down mountain; treacherous footing means d6 roll for each character crossing with every 2nd result of 6 requiring a save vs. paralyzation or fall (d4x10′ distance).
11 Exposed section of path where wind whips around mountain; cannot effectively communicate for d6 turns, otherwise safe.
12 Large rock dislodged from above path; randomly determine targets in character group and roll to hit as 7HD monster (can target any number of targets within 10′ of each other); 2d6 damage on hit and save vs. paralyzation to avoid fall (d4x10′ feet)

Notes on Table:

  • Improvise strength check if characters tied together with rope and some succeed and others fail required saves
  • Cross off events as used; replace table items as needed with new events between game sessions
  • Above tree line, roll for random encounter with special event (i.e., ambushes utilizing special event feature)

Some Considerations

All of this is meant to be random and loose — and that also means there will be a lot of minutiae to adjudicate in the moment. Still, I found this more satisfying than just making the trip take longer or needing to map out the entire mountain and place interesting features ahead of time.

Cover of Saddle-stitch print version

Luka Rejec’s beautiful cover in the flesh!

Having approved the proofs from both DrivethruRPG and Lulu.com, my low-level OSR adventure module is now available in print from those vendors. I am quite happy with this as someone who, while regularly purchasing PDF resources, only makes use of things for which I have physical copies. As someone who has worked in digital publishing for more than a decade and whose dissertation was written in XML before that was cool, my Luddite tendencies when it comes to books sometimes surprises even me — heck, I’ve never managed to read in eBook in all the years that I have been producing them for a paycheck!

Leaving aside issues of hypocrisy or technophobia, though, there are some real reasons to check out the print version of Automata Run Amok. Chief among these is that the map key portions of the module (see my previous post) were arranged on purpose to take advantage of the two-page spread. That explains the following features:

Interior spread from Saddle-stitch print version

Interior spread (Map Key)

  1. The map key portions of the module are oriented differently than the rest of the adventure (landscape versus portrait) and the spreads are flipped so that you can turn the book and read straight down across the gutter.
  2. All keying information appears on the same spread as the relevant mini-map (the map appearing in the top-left of the spread when you hold the book sideways).
  3. Visual information trackers and commentary for section of the map key always appear at the lower-right of the page spread.

While I have been happy with the comments on the layout of the PDF version to date, I think these choices show my print prejudice and my future publications will likely save these flourishes for a print-only layout since they add little to the reader’s experience when viewed in a digital format. Lesson learned!

Links to Print Versions

Detail of illustration by Luka Rejec

Let’s take a look under the hood … (Art by Luka Rejec)

Last week I published my first RPG product, an OSR module entitled Automata Run Amok designed to be an introductory adventure for low-level characters. I’ve been pleased with the reception so far and, barring unforeseen circumstances, there will be a much more substantive follow-up publication next year. Before going to work on that new project, though, I thought it might be good to discuss explicitly some of the design choices that shaped Automata. This post is the first in a series to tackle that subject and it focuses on that topic of perennial debate at the heart of most modules: map keys.

Platonic vs. Aristotelian Map Keys

There seem to be two schools of thought on keying maps for published RPG adventures. In more traditional products, like many of those created in the early days of the hobby by TSR, map keys read like the authors were transcribing the platonic ideal of a play-through for that module: copious text lays out the details of each location with the assumption that every Dungeon Master aims for the same tone and style of information management, that the DM has carefully read and at committed much of the product to memory, and that the group consists of well-behaved players who will dutifully listen to and absorb those details before acting. Everything needed to run the location is present, somewhere in the text, but there’s very little indication that the writers concerned themselves with the realities of variable humans interacting with their creations in real-time. An example of this would be the key entry for the Earth Temple in The Temple of Elemental Evil:

Example Page for The Temple of Elemental Evil

May Zuggtmoy help you if you need to know who and what is in the Earth Temple quickly while running ToEE …

The alternative to this ‘Platonic ideal’ approach is what I guess could playfully be described as Aristotelian — an empirical approach to map key information design based on observation and direct experience of what really happens at the gaming table. There have likely been examples of this throughout the history of the hobby, but the explosion of small publishers and other hobbyists dissecting and philosophizing about RPG design has intensified the drive to innovate the map key. Of course, like anything based on observation and direct experience, this innovation has led to a number of competing and idiosyncratic approaches. I will not even attempt to describe this variety except to say that, generally, these methods favor the hierarchical arrangement of information based on its perceived importance to play and the use of formatting to provide visual cues so DMs can quickly parse what is being presented. My work in Automata Run Amok borrows shamelessly from what I feel to be the best examples of this ‘Aristotelian’ map key design, distilling the practices and recommendations I have encountered over the last few years into what I hope is something solidly useful if not innovative.

Dissecting a Map Entry from Automata Run Amok

So, what does it mean in practical terms for my module to employ ‘Aristotelian’ map key design? That will be easier to explain by grounding the discussion in a specific example from the publication and then explicating how it embodies my design philosophy. Let’s look at the first entry in the key, which describes the public part of a tinkerer-wizard’s shop that has been overrun by out-of-control automata:

Map Key example for Automata Run Amok

Some Aristotelian Key Entries (Click to Enlarge)

This makes an excellent example for three reasons: there is quite a lot of stuff for characters to investigate, its default state is rather static in terms of opposing NPCs and monsters, and it integrates some non-key elements of the module like its random tables and timeline. Even before diving into those details, though, notice how certain items are called out using color or styling (i.e., italics and bold-face) while the key entry largely consists sub-headings and bullet lists to create a clear hierarchy of information. Comparison with the entry for the second room shows that this layout is consistent so that, once the Dungeon Master learns the module’s visual vocabulary, he or she knows everything necessary to parse subsequent entries on-the-fly.

Now let’s walk through each element of the entry:

  • Room number, name, and dimensions (larger red font): As the only non-black font in the module, the red titles for each entry clearly divide up the key’s entries instead of relying on spacing alone to convey this information. This allows the DM to find the entry’s beginning quickly and jump back-and-forth between the text and map without losing his or her place. The number’s purpose should be self-explanatory, while the name of the room simultaneously provides a shorthand indication of what kind of location is being described and makes sure there’s enough text for the red font to stand out. Finally, the dimensions are given in the top line since the player group’s mapper (remember, this is an OSR module!) will need this information. My habit when Dungeon Mastering is to give mappers the information they need early in the description unless there is a logical reason — such as a huge chamber or attacking monsters — that would prevent someone from quickly eyeballing the size of the room. That allows them time to sketch and revise even if the party is prone to moving along quickly from location to location.
  • Brief evocative phrase (italic font): Quite a bit of the debate surrounding RPG map keys involves boxed text and I tend to agree with the argument that blocks of read-aloud for the DM are largely useless or even counter-productive at the table. It constrains the tone of the game to whatever the module author thinks is appropriate for a “typical” group and assumes a level of patience and attention to detail among players that I have rarely encountered in an individual let alone an entire party. However, ditching boxed text does limit the module author’s opportunities to convey the non-visual elements of a room that would be immediately evident — its smell, sounds, and general feel. My compromise is this evocative phrase, which describes the flavor of the room yet still allows the DM to couch that information in whatever tone or style fits their gaming group. For example, this particular room has “shelves of arcana and pungent herbs” — in other words, it’s a crowded shop of curios of mysterious origin suffused with the smells of magical ingredients. My opinion is that a DM doesn’t need more than this to describe the room in appropriate terms according to the needs and attention span of their player group.
  • Subsection detailing room occupants (if any; bold-face header with bullet list of details): Once the DM has been given the top-level information about the room’s dimensions and feel, the module’s key consists of subsections presenting information in a hierarchical list from most to least important. Since the first thing players (and DMs) will want to know is whether a room contains living creatures to either speak with or kill, those are covered in the first subsection entitled “Occupants.” The shop room in the key above is typically unoccupied, but this subsection is still present because there is a chance of encountering monsters (monkey automata) or NPCs (robbers) as the adventure’s timeline, as outlined earlier in the module, advances. When appropriate, this description of the room’s occupants will also cover the likely reactions along with references to the module’s bestiary. A good example of this comes from the alchemy chamber later in the module:

• The first time the party enters the tower, a single monkey automaton (AC:6; HD:1+1; #ATT:2; DAM:d6; Bestiary 1) is in this chamber collecting the last two acid vials (the remainder are already piled on the catwalk near the ladder). Hearing the group enter, it will climb the wall and perch above the door – making a last few squeaks in the process
• When the first party member enters the room, the monkey will attempt a surprise attack and toss one of its acid vials (2d4HP direct damage; 1HP splash within 1’). Any counterattack will cause it to scamper over to the other side of the wall and toss another vial; a successful counterattack will cause it to leap onto the catwalk (which provides 25% cover from missile attacks)

  • Subsection listing and describing exits (bold-face header with bullet list of details): Once the players know whether there’s anything in the room, the next thing they need to know is where the available exits are located and whether those exits are barred, locked, or otherwise impeded. This is also the next thing the party mapper will need to know once the outline of the room is sketched and therefore appears towards the beginning of the entry.
  • Series of subsections listing items of interest (bold-face headers with bullet lists of details): The last thing a DM must provide players is a list of miscellaneous items within the room that can be interacted with or investigated. My intention here is that only the headers would be shared during the initial room description and that the additional information detailed in the bullet lists would only be disclosed if the party spends time (i.e., expends resources) to look more closely at the item in question. So, in our example, the DM would let the party know that there are shelves covered in items, a counter, and a closet in the shop. In most cases, I imagine, the party will be interested in seeing exactly what’s on the shelves and choose to spend a turn investigating those; at that point only they would learn that these contain objects “artfully arranged to look cluttered with no labeling,” including several large crystals (one of which is valuable), spell components if a magic user is doing the searching, and something special from the random magic item table. Similarly, if the party or member thereof spends a turn investigating the counter, they will find the shelf underneath with the book and money-box.

So that basically demonstrates my Aristotelian style of map keying, which leverages visual cues and the hierarchical presentation of information to make running the adventure at the table a smoother experience requiring less preparation. The page layout supports the structure of the key entries, with each two-page spread in the module containing all the entries relevant to the map (an innovation lifted from Maze of the Blue Medusa) and trackers with check-boxes and blanks for the DM to note the passage of time, defeat of enemies, and other changes to the location (suggested by my editor and illustrator, Luka Rejec). When printed out, all of this information would be visible at once and (hopefully) easy to interpret even with the reality of excited players, imperfect recall of module details, and a tone that in my experience inevitably oscillates between epic and farce.

Quick Summary of Map Keying Practices

  1. Consistent ‘visual vocabulary’ across all map key entries
  2. Hierarchical presentation of information (from possible threats to miscellaneous objects to investigate)
  3. Front-loading of information for mapper (dimensions and exits)
  4. Use of ‘evocative phrase’ as substitute for boxed text to convey non-visual information not specifically associated with items in the room (i.e., the ‘flavor’ of the location)
  5. Subsections for each item in the room with which players can interact; bullet list details only shared with players if they expend resources (time, light source, random encounter rolls, etc.)
  6. All key entries appear on same two-page spread as map that references them along with visual tracking tools to keep track of monsters defeated and similar things

[If any of this sounds interesting, please pick up a PWYW copy of Automata Run Amok at DrivethruRPG; if you have any feedback — including reviews — please let me know!]