Having not actually run a true urbancrawl for any length of time, I’m not really in a position to delve into truly advanced uses of the technique. But I do want to float out a few random thoughts I’ve had. This is stuff that I think will prove to be fertile soil for exploring in the future. (And if you have a chance to play around with these ideas in your own campaign, I’d love to hear your feedback on how it went in actual play.)
THE DIMENSIONS OF THE URBANCRAWL
First, I want to be very specific about what I think makes this model of the urbancrawl work by expanding on the metaphor of “dimension” in your urbancrawl that I’ve touched on before:
0th Dimension: This is the gazetteer. Recognizing that the gazetteer is a separate entity allows us to focus on the ‘crawl itself. You don’t have to explore the gazetteer; its contents provide context and backdrop and common goals for targeted movement.
1st Dimension: This is the basic “investigation” action. You could key an entire city with content and just allow basic investigation and it would work, but it would also be very bland.
2nd Dimension: By creating different urbancrawl layers, you allow the players to contextualize their investigations. This makes the city “come alive” for the players and rapidly creates a sense of the bustling metropolis; of a place where there’s always something happening just out of sight. (I suspect this will become even more true as the city grows during play and the various layers begin interacting with each other.)
3rd Dimension: Finally, adding depth to each urbancrawl layer and allowing PC activity to expose the hidden layers rewards player exploration. It can also be used to escalate the stakes and to increase the PCs’ investment in the setting. (In many ways it parallels the function that deeper levels of the megadungeon serve.)
URBANCRAWLS IN YOUR HEXCRAWL
The more time I spend playing around with this urbancrawl structure, the more excited I get about its potential. For example, you can use first dimensional urbancrawls to cleanly integrate villages, towns, and the like into your hexcrawls.
For most of these settlements, you can probably treat the whole town as a single district: When the PCs encounter a small village or town in the hexcrawl, the expected interaction is to look around and figure out what useful information it’s supposed to give you (i.e., rumors about potential adventures to be found in the wilderness). Alternatively, maybe investigating the village will end up triggering a local adventure (i.e., the whole town has been replaced by dopplegangers). In either case, the urbancrawl investigation action provides a default method for interacting with settlements of all sizes, even if it’s only when the really big, important cities range into view – Greyhawk, City-State of the Invincible Overlord, Minas Tirith – that exploring the city neighborhood-by-neighborhood and street-by-street becomes an interesting adventure in its own right.
To use a potentially ill-conceived dungeon metaphor: Most towns are like caves; they’ve got just one or two key entries. The big cities are full-scale labyrinths and can be chewed on for months or years.
MIXING URBANCRAWL SCALES
Most of the time you’ll probably want all of your urbancrawl layers keyed to the same map. But if the PCs become interested in the Dockside gangs, maybe you break up the Docks into specific sub-districts.
Similarly, maybe the vampires are only active in Oldtown. Or maybe there’s a gang war in the Guildsman District that you want to track street-by-street as territory gets swapped back and forth.
CROSSOVER NODES
Another assumption is that each node will only belong to a single urbancrawl layer, but it would actually be quite trivial to key the same node to multiple layers. These nodes would make the city feel more interconnected, but more importantly they would also serve as a mechanism by which the investigation of one layer can crossover into another.
For example, maybe the PCs have been rigorously pursuing the Halfling Mafia. If they end up raiding the blood laundering service the mafia runs for Count Ormu, however, that will tip them off about the local blood dens and possibly get them investigating the vampires, too.
URBANCRAWL TRANSMISSIONS
Technoir transmissions, as previously discussed, combine random content generators (for connections, events, factions, locations, objects, and threats) with explicit mechanics that generate a conspiracy as a direct result of the PCs hitting up their contacts in an effort to unravel the mystery.
It’s incredibly clever and extremely effective. And for dedicated groups, I think you can use the transmission system to add a fourth dimension to your urbancrawls: Tie the random content generators to your urbancrawl layers, seed the city with contacts for the PCs, and then let the system generate plot maps that bring the city to dramatic life.
I don’t have space here to fully explore this idea right now, but here’s a few preliminary thoughts:
- Connections, locations, events, and threats all probably double as items keyed to the urbancrawl layers.
- Many or all of the factions probably have their own layer on the urbancrawl.
- Objects are the one thing you’d have to create explicitly for the transmission dimension. (Fortunately, they’re also the easiest thing to create.)
For more complexity (or for groups who are new to the big city), add a mechanic that allows them to explore the city in order to make contacts. (Creating a dedicated contact layer in your urbancrawl or incorporating them into other layers seems like an easy solution.)
Finally, I’d be interested in adding mechanics to the transmission system so that performing generic or specific investigation actions would have effects on the plot map in the same way that hitting up contacts do.
SECONDARY INVESTIGATION ACTION
Another trick that Hite incorporates into Night’s Black Agents is adversary mapping: As characters explore the Conspyramid, they can map the relationships of the nodes on the pyramid. They can also use the Human Terrain and Traffic Analysis skills to peek at the generic structure of the map around the nodes they’ve discovered. (For example, “Someone has to be running the money to these guys.”) Additional investigation can then nail these structures down. Night’s Black Agents rewards the players for identifying sections of the adversary map by rewarding a dedicated pool of points for actions targeting that section.
In terms of our urbancrawl structure, we can imagine a secondary investigation action that the PCs can take to follow-up on the leads they gain from identifying, exploiting, exposing, or eradicating a node on an urbancrawl layer. For example, if they take out a blood den in Oldtown they could follow up with a secondary investigation action that might tell them where they can pursue their investigation:
– Asking around about the blood den you just rooted out, you hear that a lot of people wearing the livery of House Ormu were seen coming and going at odd hours of the night from that warehouse.
– Somebody must have been supplying those shivvel dealers with their product. And somebody must have been paying off the local cops not to look too close.
Basically, the idea here is that, when they perform the secondary investigation action, you would look at other keyed content on that urbancrawl layer and point them towards it. (Structurally you’re saying, “You should go perform an investigation action in district X.” But you’re contextualizing that into the game world.) Just like Hite, you could also incentivize this action by offering rewards for following up on leads. (A +2 circumstance bonus, for example, would work in D&D.) And I suspect that there may be richer ways of building on these secondary investigation actions.
RESTOCKING THE URBANCRAWL
When you clear out a dungeoncrawl, the dungeon is empty. You clear out a city and… what does that mean?
To a large extent, the layered approach to stocking your urbancrawl solves this problem. If the PCs wipe out Count Ormu’s vampires and clear that entire layer, there are still other layers of the city to explore. (And, of course, you can always add new layers to the city over time.)
One thing I am interested in is what actually restocking a layer (or a city) will look like in a campaign over time. For dungeons, this is a process I talk about in (Re)-Running the Megadungeon: “You keep the dungeon alive by using wandering monster encounters to simulate the activity of the complex. You partially repopulate the dungeon between sessions to keep it fresh. The result is that you can take 10 encounter areas, a couple of tables, and get dozens of hours of play out of it.”
I expect that a lot of those skills and techniques will transfer from the dungeon to the city. But I also anticipate that urbancrawls are going to evolve in their own unique and fascinating ways.
So that’s the next step of this journey: To bring the first urbancrawls to the table. To let them begin to grow and live. To unleash the unbridled creativity of the gaming table upon them.
I’m excited.
This is really exciting. I’m about to start an urban exploration adventure (The Asylum Stone from Paizo) and will twist things around to see if I can use some of this.
Cheers!
Kinak
I also am about to start an urban crawl of sorts – Razor Coast by Frog God Games. While my intention isn’t to deviate too far from the original structure it would be interesting to have these additional dimensions. Will ponder further on how this could be implemented.
Should anything worth reporting come up, either in planning or in game sessions I will of course feedback to you.
Thanks for this series! (And thanks to the patrons who made it possible?) There’ve been a number of things here that I’ll have to keep in mind for my own use when I get the time to DM again.
One tangential thought: your aside about hexcrawls made me think of a sort of halfway format in the form of a “ruralcrawl.” (I’m inspired by my time in rural Japan, where a “city” is as often as not just an administrative bundling of small towns scattered over an area, separated from each other by mountains and streams or circling the edge of a wide bay.) The overall map has little true wilderness, or at least no wilderness that you can’t pass through in a couple hours of walking, so most of your “encounters” are going to be with people. But the encounters are bundled up into little market-towns and farming hamlets scattered over the countryside, and probably interspersed with more-traditional caves, ruins, and lairs.
It would be like superimposing your idea of urban districts with their layered keys onto a wilderness map, and allowing access to both, if somewhat less intensely, at the same time?
I’ve been giving this an honest shot. I have a mixed systems approach that sounds more complicated than it is when I write it out. Basically there are two charts of results, Table One is a version of Jrients carousing tables that is citywide to start, Table 2 is a list of information about the Layers of the city in that District. A dungeonworldesque roll is made:
2d6
1-6 carousing mishaps a la Jrients
7-9 useful information + complication or district appropriate threat
10+ useful information on a deeper layer. This information can often be the basis for further investigation checks.
-if the search for information is ‘contextualized’ as per your earlier post, the roll is made 2d6+1 AND the information is most often from the intended layer.
-if the search for information is contextualized AND a connection is leaned on a la Technoir, its made at +2 and the information sought is always from the intended layer, but there is implication of the connection as described in Technoir.
Last Thought to achieve TL;DR status: Using the Technoir approach, you can integrate a city with a hexcrawl pretty easily by making locations/objects hexcrawl sites.
P.S. Fantastic Series
This was a very exciting essay series for me. It has stimulated a lot of creative juice and I now have a prototype urban crawl structure to play test. Your discussion of layers was the key for me and then I riffed off that. Here is what I’m thinking.
Encounter Layers
Encounter layers are a city wide collection of geographically keyed buildings or spaces that contain some detail encounter, or street encounters that can all be active or dormant at the same time. The spaces and encounters in a dormant layer will not be encountered by the PCs as they pass through the geography. Some buildings or spaces may be marked as “landmark” so the players will know they are there even if their layer is dormant. Other buildings or spaces may have been previously encountered so players will know they are there just as for landmarks.
Buildings, spaces, people or creatures may participate in multiple layers and will be dormant only when all the layers they participate in are dormant.
Example layers include a carousing layer (inns, gambling dens, brothels, boxing rings, theaters) or an information gathering layer(inns, gambling dens, market spaces, library, thieves guild) an armour and weapons layer (smithies, the local knights private armourer, the bazaar for lesser pieces).
Encounter Layer Types
There are three layer types based on how they relate to the PCs
The Incidental Layer – this is a layer about random things the PCs may unintentionally bump into as they move through or hang around in the geography. Some of the encounters or events may recur and may be reused in the encounter tables for several geographic locations. The Incidental Layer is always active.
Focus Layers – these are the layers built around what the PCs are focused on and will actively bump into, lodgings, provisioning, carousing, gathering info, finding employment, seeking hirelings, seeking treasure to heist or monsters to deal with, buying armour and weapons, converting gold to portable wealth and banking etc. (This feels like there may be a definitive list that can be re-used that may take a while to refine). Only one focus layer is active at any time. When the players switch focus then the current focus layer goes dormant and the new one becomes active.
The Belligerent Layers – these are collections of encounters that actively seek to bump into the players. They only become active when triggered by something happening in the city and an active belligerent layer should go dormant again in a fairly short period of time. This layer could include town watch groups alerted to the PCs and wanting to apprehend them, or thieves that have been informed the PCs are a mark and have alerted various agents to waylay them etc.
Precincts
The basic element of geography in which encounter layer elements may be found. Precincts have a character – the poor district, the wealthy district, the lane maze area, the Lord’s grounds, river side etc. Some idea of the amount of open space and ease of moving through the precinct should be noted. Perhaps there is a possibility of becoming lost or a difficulty level for finding the buildings, spaces or encounters the PCs seek. It is not necessary to have a detail street by street or building by building map of a precinct. It can be something like a wilderness hex with specific points of interest within it.
Time
A city changes as the day cycle…um…cycles. There are four basic time zones. Morning, day, evening, night. Either the keyed encounters change state depending on the time of day they are being encountered or the times represent a further layer separation e.g. The morning incidental layer is a different beast to the night incidental layer.
Random Encounters
Each layer and/or precinct or type of precinct can have its own encounter table. Entering a new precinct should invoke a random encounter check at whatever odds seem appropriate. If an encounter is indicated then another roll is used to see if the encounter is from the incidental layer, the active focus layer or an active belligerent layer. I’m not sure if the ratios should be fixed or flexible or what proportions to each. Experimentation required.
Thoughts
This mechanism might produce a lot of prep. A city becomes something like a 10 level dungeon where the players vertically teleport as they wish. However my first experiments don’t seem to be that onerous. Each focus layer has some simple shop spaces, and an “asking directions” mechanic with a few special encounters, and a number are across multiple layers.
Some things I had never thought of before and they are leaking out to my dungeon/wilderness crawls. e.g. Have you ever had your PCs do a midnight raid on your dungeon? Would it make a difference to your dungeon crawl?
Still working on things like places that are mini dungeons at street level in a precinct, and how things might activate belligerent layers.
I’m starting to pull together more ideas for a regular mechanic. One thing that feeds into it is situations around the city’s movers and shakers. They will produce nodes that cross cut the player focus layers.
An example I’m working on is between the local noble and local head of the church. The noble wants people to worship Ares god of war, the priest wants people to be more philosophical and follow Hera. In the incidental layer there are now encounters at the city gate and the market district with proselytizing clerics for each concern and some encounters where the clerics are clashing with one another. In the player focus layer for finding arms and armor there are encounters with clerics for Ares assuming that people arming themselves are good converts, while in the “find lodgings” focus layer there are some encounters with clerics for Hera trying to recruit the PCs.
Similarly the crime bosses in my city have a small network that looks for wealthy PCs seeking lodgings and in the gem and jewel smith district as potential marks.
These all end up on my random encounter tables and are not always triggered.
I also have some set piece scenarios that are triggered when the players randomly encounter clues about them. e.g. If in the poorer villein district the characters are asking directions as part of any focus layer they may encounter a ghost that will attempt to draw them to a house where the couple that murdered him now live. They are rogues who carefully do measured petty theft here and there to maintain a comfy lifestyle. The body of the original home owner is in the cellar, along with their “rainy day” treasure and some traps.
These experiments are specific but I’m starting to see something of a mechanics frame work emerge for generic encounter lists that can be filled with specifics as needed.
Active belligerent layer makes me think of that police attention mechanic from the Grand Theft Auto games.
Maybe active belligerents should have a specfic associated duration, corresponding to how long they’ll persist in the hunt between direct sightings of their quarry? 1d6 hours for an angry mob, probably no more than a week for anything but fanatical types and tireless hunter-golems. After the direct search has been fruitless for a while, they either abandon the matter altogether, or downshift to more casually ‘keeping an eye out’ and move to the incidental layer.
Adds a bit more bookkeeping, since you’d want to note when the last encounter was, but the benefit is increased engagement with systems for stealth and disguise.
Exahustilingly fantastic!
I have been wanting to do some sort of ‘crawl’ – maybe start in a village and go from there and see what happens but I want to do it as a solitaire venture and then blog/tweet about it.
I don’t mean one GM and one Player I mean just me, one person.
I’m not sure how to go about it but I think I could make it work.
I’ve thoroughly enjoyed this series of articles and it has really gotten me thinking. These are my thoughts (in no particular order).
Pacing is a big issue with an urban hex crawl. Players generally don’t have “let’s explore” as a goal when visiting a city and are quickly bored with a “you reach the next city block, what do you do?” approach to their game. They are usually in the city with a purpose (typically to access the commercial life of the city; to access specific services, rest and recover, find their next job, etc.).
To this end they will, generally, establish a base (permanent homes, or temporary rooms in the local inn etc.) and from there engage in a number of typical actions…
1. Access services
2. Research matters of interest, overhear gossip, and/or ferret out rumours
3. Check/establish contacts
4. Find work/accept jobs
Note, that “explore” (apart from a gazetteer like approach to learning about the city if it is the players’ first visit, or a specific role in the city that moves them around – beat cop, deliveryman, mailman etc) does not feature strongly in the above list. To my mind the arguably tame world of the city is less open to the exploratory behavior that is appropriate in the wilderness. Roaming the city like a bunch of teenagers in search of kicks just doesn’t work for me as a plausible motivation for adventurers.
As a result, a node based approach to city life is more natural, initially, than a hex based approach (at least I think so, from a player point of view).
Does that exclude the possibility of hex based adventure encounters? I don’t think so.
Items 2, 3, and 4 on the list can point the players at specific hexes, while travel to and from services can be interrupted by hex encounters.
Further if we spread the services our players access out over the city and turn each into rumor nodes in their own right, we can point players to nearby hexes and ensure that our players come into contact with plenty of hooks and reasons to visit various points of interest around town.
Once the team begin to see the adventure possibilities within the city, they may indeed start to think of it as a place worthy of exploration in its own right. But probably not at first.
In my view, and by way of resurrecting an old and underutilized old time gaming mechanic, it helps to think of the city as a series of rooms in a dungeon (stay with me, I know this seems trite) with a series of doors and connecting passageways represented by rumors and job offers.
Here are my untested thoughts on creating an urban crawl.
Like you say, a city has a surface life (its commercial purpose as a hub of goods and services), beneath that is its political and regulatory structure (how the peace is kept and who’s in charge), and beneath that its secret life (the underworld, monsters, secrets, cabals, and hidden things that make for adventurous encounters). This is probably one of the best takeaways ideas I got from your articles.
So, firstly, I’d prepare a hex key according to your suggestions and key each hex to the surface commercial, then the deeper political, and then the deepest hidden, life of the city.
I’d spread the commercial services most likely to be accessed by players around the city and, most importantly, give each a collection of rumors linking to the hexes in immediate proximity around them and the political and secret life encountered within them. I’d make the proprietors and customers of these services the source of these rumors (either through a planned dropping of rumors or through a random table); “Ah, good to see you again and what can I do for you? Oh, and by the way, did you hear about the odd goings on in Thibalt Street recently?”
By keeping these rumor nodes spread out, I am hoping to give the players reasons to pass through hexes along the way. An interruption mechanic would be necessary to determine whether the players find a hook on their trip, eg. “On your way to the apothecary you pass along Thibalt street and notice a surprising lack of bustle, many of the shops are shuttered and closed, and surly looking individuals in dark clothing stand on the corners of the streets and alleyways giving passers by reason not to linger.” The players may or may not react to the events they encounter but the hooks can easily be placed before them this way.
Now I’d also spread some potential employers out across the city in the same way and set them up to become contacts and sources of work and rumors in the city). The players can be hired for specific jobs this way (potential employers can send word to them at their base of operations) eg. “I’d like to hire you on behalf of my friend Rurik the merchant who is being driven out of business by an extortion gang in Thibalt Street”. Once more they can be fed rumors about the nearest hexes, and can encounter hexes on their way to and from the job interviews and the jobs themselves.
Okay, with the commercial layer of the city is complete, I’d want to link the jobs and rumors to the political life and secret life of the city and create new rumour nodes and job sources for areas as yet unpopulated in the hex map. Hubs that specialize in rumors regarding the political life of the city (such as city hall) and the secret life of the city (such as secret guild location) are also appropriate.
Lastly, I’d add encounters based around the secret life of the city, political, and job based rumors, along the paths players must take to reach any of these rumor and job hubs and link rumors and jobs back to them.
If any hexes remain unpopulated, I’d fill them and link them back to their nearest rumor nodes.
These “word of mouth” linkages act a bit like the doors and passageways of the traditional dungeon, and, like in the traditional dungeon, can provide a hex with numerous points of entry from different locations.
Looking back over the above I can see that managing this complexity (especially managing the retirement of a rumor that can be found on multiple nodes) would be difficult. A master list of rumors, their target hexes, and the hexes the rumor can be collected from would probably be essential.
So, if all goes well, this exercise would result in a hex map and network of rumors, jobs, and plot hooks spreading out across the city from the players home base that will hopefully link them to a nearly inexhaustible array of hex encounters if they choose to follow them up. How are they discovered? Well the shopkeepers, service providers, and potential employers have to communicate them in the pursuit of their usual activities (buying, selling, provisioning, hiring etc.) either from a list (marking off each as it is delivered in turn to avoid repetition or via a randomized table). Jobs are merely a form of explicit rumor hook with a predetermined reward.
What about “on the way” encounters. Each hex should have a “% chance of activating” mechanism that would expose the hook as players pass by. These should be pre-rolled as soon as the players’ destination is known, so you can say “You notice three things along the way…” rather than presenting the players with a smorgasbord of interruptions on the way to a destination with no end in sight.
This would all be fine and highly workable enough on its own, I think, if cities were static environments. This, however, is not the case. They are dynamic environments, have a great deal of traffic passing through them, can experience sudden reversals (plague, famine, fire, etc.), and can be subject to major political shifts etc. Is there a way to incorporate this dynamism into the hex based design of the city? I loved the suggestions you made regarding the conspyramid and so on, but it seems a little complicated (or at least I didn’t quite get it) and I’d like to make an alternative suggestion.
At one level, much change can be left up to the players (let them be the pebble that starts the ripples of change moving across the pond) and that is fine, the city should have a reactive life (one that responds to player actions). But I suspect the experience would be richer if it had its own pro-active life (a history that unfolds without player intervention). Stating that simply begs the question of how to build that into the game design however.
Perhaps giving each hex a timeline of action that covers its history (each new change having occurred between visits by the players) would work, but this seems to require massive amounts of, potentially wasted, ongoing work with little pay-off.
Since actors create change, I wonder if perhaps creating factions within the city (like a set of small conspyramids) with conflicting agendas and creating an overarching timeline of action and outcomes would cover it. Big agendas, like these, are rarely interrupted by the players (though attempts to interrupt those agendas have real consequences for the players; new enemies, vendettas against them etc.). The players can kill the head of the evil council of magic attempting to take over the city, but his right hand man is always there to take over etc. making the macro-narrative quite robust. Note, this is in relation to a macro history you are establishing to give the city a life and change outside of player agency (and one that is only tangentially interacting with the players). If player’s become centrally embroiled in the big macro-narrative of the city, then they should absolutely have the agency to impact and redirect it. I would just have to keep my timeline handy and rewrite it on the fly in light of the consequences.
A key to major NPCs, their agendas, and the likely consequences (for the players) if those agendas are thwarted, might be useful (though such information could be included at the hex level I suppose).
Before each session I would need to check on the state of the city according to the Macro-Narrative/timeline and adjust any hex encounters to reflect those changes. Knowing what nodes the players wish to visit in the session and what rumours and hooks they will encounter as a result should, potentially, make this manageable (but still requires a bit of thinking on one’s feet).
Would this work? I’ve no idea as I’ve not tried it yet, but it does seem to offer a way to avoid prepping plots, while responding to players’ desire to engage in typical node based city activities, factoring in a dynamic environment, and leaving open a wilderness style hex crawl option suitable for an open table. My only remaining concern is how to deal with the tendency a party has to split and each member visit their own specific services individually (something that cities seem to encourage far more than other adventuring environments). Individualized encounters tend to put everyone not involved in them to sleep. Thoughts?
Two short(ish) thoughts.
First, on crossover nodes, it seems to me that this should be the norm, not the exception. The nature of a city is that everything is interconnected. So almost every node of a layer should know something useful about a node in another layer. It might be mutual communication. It might be due to secret investigation, or a desired possession. Whether players use it or not (then), it’s there.
Second, a small peeve. Each layer seems to be unique and comprehensive. There is one thieves guild, one vampire rulership, one … well, you get the idea. Every successful organization has its peers, its competitors. I suggest that when building an urban layer you make two or three. Yes, multiple layers are dense, but that’s a city. And conflict is opportunity is story.
Those noted, thank you for this series. I’ve enjoyed it, and I’ve learned from it.
Have you done an Urbancrawl with this yet? Can you show us what some of these layers would look like in actual prep notes?
> Every successful organization has its peers, its competitors. I suggest that when building an urban layer you make two or three.
@Kirk: I think it’s reasonable to have multiple different factions that are directly competing with one another on the same layer. Say there was an external thieves’ guild attempting to gain a foothold in a town currently dominated by the local Halfling Mafia. They could exist on the same “organised crime” layer, vying with one another over territory and influence.
Think of it like wild animals marking their territory – they don’t care about the territorial markings of other species, but they sure pay attention to markings from their peers.
(And of course, just because they don’t care about the territories on other layers doesn’t stop them from interacting with individuals from other layers, since they all coexist in the same physical space.)