How detailed to go?

[heregoeshell]

Sorry for the wall of text in advanced.

I'm building a city in my own fictional world with 3.5 as the system. I'm running into the issue of figuring out how much detail to go into with my preparation.

Backstory: Ran this cityscape idea a few years ago as a new DM, was completely unprepared for what ended up happening. I had more or less established the story as a choose-your-own-encounter style game, which worked relatively well. Eventually the game fell apart for several reasons. My issue however, is that as a new DM at the time, I was more or less totally unprepared for how to deal with any off rail choices my players would make, even though I was using random generators to develop npcs, stores, building floor plans, and what seemed to me to be non essential detail. I felt like a drag to my players because I was not prepared, and thus spent a decent amount of time trying to keep track of characters I wasn't familiar with, stores I didn't know, places I hadn't mapped out yet, and choices and directions on the part of the Players, that I wasn't generally prepared for.

The present: I am now rebuilding this campaign in much greater depth, but I don't know how much detail is truly necessary for my players to have a good time. As it is, I really enjoy detail building, but I'm not sure exactly how coarse or fine to make the details. I have many districts with lists of buildings in each district. Do I need to establish the exact locations of buildings with alley ways and other features? Do I need to have pre built traps and inhabitants for buildings and stores? Should I have inventories already established of each location? How much detail do I need to do into the inhabitants?

Ultimately, I'm hesitant to go into an insane amount of detail, because the idea of documenting 45 districts, with 25+ buildings in each, with inhabitants in the city of over 40k, and interactivity and relationships across districts and involving city politics and trade and governance. It seems incredibly excessive and like I may as well have just ported over a written version of Skyrim and called it good. Not to mention what happens if the players want to go to outside of city, and I haven't fully established the underground that is present as well.

Again, sorry for the wall of text.

[Venger]

Have you tried talking to your players and asking what they're interested in and what kind of details they want in a game?

Whatever their answer, don't make a list of 10,000 buildings, no one cares and it doesn't sound like you'd enjoy it.

[PacMan2247]

Have you tried talking to your players and asking what they're interested in and what kind of details they want in a game?


Can't stress enough how important this part is. Your players will need an outline of the setting to help them decide what sort of characters they want to use to interact with the world. While they're working out their characters, they'll (ideally) be coming up with motivations and interests, which can help you decide where to take things.

Something to keep in mind about cities is that they're nearly always broken up into different neighborhoods, and each neighborhood will have certain types of shops and services. Once you figure out what a given shop (or inn, or smithy, or whatever) looks like inside, re-use the floorplans at will- if your players even think to question it, you can say it's building codes. Maybe one neighborhood has the best blacksmith, another neighborhood's tavern has the best beer, and the greengrocer up the street has the best produce (but charges a little more because he's obviously paying the farmers extra to come to him first).

You can also create the appearance of a wide open, sandbox-y environment without actually creating a huge number of encounter options. If you want a merchant with a broken wagon wheel drop an adventure hook, or thugs in an alley, it doesn't have to matter where in the city the party travels- wherever they are, or whatever they want to do, you can set up something to be dropped into nearly any neighborhood. Want them to investigate a string of assassinations? Maybe they overhear off-duty guards in the tavern (or have the barkeep pass the tip), or maybe they see signs of a fight while they're out restocking on spell components or healing potions; these options assume that they haven't yet earned the sort of reputation that would have an interested party sending a messenger to them. You don't have to make a bunch of larger stories, just one that you can drop hints of from a variety of sources and eventually have the threads lead back to the center.

If a location is going to be significant, it's worth mapping (and once you've mapped it, keep the map in case it comes up again). For the most part, fairly vague descriptions of relative locations should do. Inventories can get difficult to manage, because your players will inevitably ask about something you hadn't considered. When they do, you can make your call based on whether you want them to have to work for it a little, or you can just roll a die and say "yes" on even or "no" on odd. Minor quirks can make NPCs feel less generic, and if you have a list of names and personality traits, you can pick them as you go and have the city develop itself on the fly.

As with anything else, your mileage may vary, but if these work for you, they can help free you up to pay more attention to the story and worry a little less about the details.

[BowStreetRunner]

When I DM, I actually prefer sandbox-style games (what you call 'choose-your-own-encounter'). However, they do have their own unique challenges. There are a couple of pointers I usually give others who are considering running in this style.


Think Modular: I'll explain this one with an example that fits your cityscape campaign. Go ahead and map out your city with lots of generic buildings. Lay them out like tetris-shaped blocks. Fill in the ones that matter right away. Then go ahead and put together a handful of generic templates for the other layouts. Do with the interiors what you did with the city itself - lay out the halls and rooms like tetris-shaped blocks but don't fill them in. Then go and create some rooms. For me, I would have printable sheets of tons of rooms. Plus printable sheets of building templates. If the players decide they need to go into a building that I did not already map out, I quick grab a matching template and stick some rooms down on it and fill it with NPCs that I have similarly statted out in a modular fashion. Everything takes me a couple minutes and we're ready to go. The more modular you are, the better prepared you are to throw something together on the fly. Have lots of random tables prepared to answer the questions you are going to need to answer for yourself. [What kind of building is it? Dice Roll says a cobbler shop. Hmmm... Elf shoemaker sounds like a fun idea. Go from there...]
Create Situations, not Stories: If you try to script out the events in the game, the players may ad-lib the story in a very different direction than you want to go. Instead, create NPCs with goals instead of scripted actions. Fill an encounter with people, places, things that they can interact with. Concentrate on how the NPCs and environment respond to the PCs actions rather than trying to railroad them into a certain planned plot.
Look at the Players' character sheets for inspiration: Pay attention in particular to their strengths (give every PC a chance to shine) and weaknesses (give them challenges that require them to go outside their comfort zone). Variety makes the game more interesting, and if you pay attention to things like creating situations that require particular skills, or that use particular abilities, you can give the decisions the players made when building their characters more meaning.
Keep a record of what happens in your game: I cannot stress this one enough. A linear campaign is written out heavily in advance. A sand-box campaign is written out heavily after the fact. It is too easy to forget that a particular NPC died or a particular monster was let loose. Write it all down.
Remember the 'Yes and' rule: In improvisational comedy there is a concept of never saying "No, that's wrong". If someone says something that you didn't want them to say, it's too late. Just say "Yes, and..." then build on what they started. Sand-box games can be similar. Look at what the PCs do and build on that. If the party decides the rogue is going to rob the supplies they need in the towns they visit instead of buying them at the shops, go ahead and start statting out Fences, City Watch, jealous Thieves Guilds, and so forth. Then go ahead and create links between some of these characters and your central story.

[Nifft]

Sorry for the wall of text in advanced.

I'm building a city in my own fictional world with 3.5 as the system. I'm running into the issue of figuring out how much detail to go into with my preparation.

Backstory: Ran this cityscape idea a few years ago as a new DM, was completely unprepared for what ended up happening. I had more or less established the story as a choose-your-own-encounter style game, which worked relatively well. Eventually the game fell apart for several reasons. My issue however, is that as a new DM at the time, I was more or less totally unprepared for how to deal with any off rail choices my players would make, even though I was using random generators to develop npcs, stores, building floor plans, and what seemed to me to be non essential detail. I felt like a drag to my players because I was not prepared, and thus spent a decent amount of time trying to keep track of characters I wasn't familiar with, stores I didn't know, places I hadn't mapped out yet, and choices and directions on the part of the Players, that I wasn't generally prepared for.

The present: I am now rebuilding this campaign in much greater depth, but I don't know how much detail is truly necessary for my players to have a good time. As it is, I really enjoy detail building, but I'm not sure exactly how coarse or fine to make the details. I have many districts with lists of buildings in each district. Do I need to establish the exact locations of buildings with alley ways and other features? Do I need to have pre built traps and inhabitants for buildings and stores? Should I have inventories already established of each location? How much detail do I need to do into the inhabitants?

Ultimately, I'm hesitant to go into an insane amount of detail, because the idea of documenting 45 districts, with 25+ buildings in each, with inhabitants in the city of over 40k, and interactivity and relationships across districts and involving city politics and trade and governance. It seems incredibly excessive and like I may as well have just ported over a written version of Skyrim and called it good. Not to mention what happens if the players want to go to outside of city, and I haven't fully established the underground that is present as well.

Again, sorry for the wall of text. That's barely a wall, you're fine.

IMHO it's fine to over-prepare a bit if it gives you more confidence when you're running, up to the point where it's not cutting into your sleep / exercise / work / other leisure activities. It's fine to over-prepare as long as you're enjoying the work, and you don't burn yourself out.

Regarding maps: possibly good to have one with details that cover whatever sort of chase / fight / etc. you foresee happening within the next few sessions, and to give the players a general idea of far it is in general between various locations of interest. The latter can be accomplished with a very vague map. There are a lot of historical city maps online; I've stolen quite a few over the years. Stealing a map is wonderful.

Regarding preparation for player actions: prep a few different kinds of building. Like, here's two poor tenement apartments. Here's one warehouse in the Docks, here's a different warehouse by the Western Barbican. Here's a pub with an inn upstairs and a brothel across the alley. These are not specific places -- they're just available to be used when you need a place of that type.

When you need a floorplan, use one of the prepared plans. Expect to throw most of them away, so don't put too much effort into them. Again, stealing from the internet is wonderful. As you use them up, prepare a few more. Obviously, if you know the next session will involve a conflict in a specific location, prep that.

For NPCs, do the same thing: pre-generate 20 names, with random trait(s) next to each one. When you need an NPC, just pick the next unused name on the list. In some ways, it's fun for you as the DM to be surprised -- "who would have though I'd use the limping overdressed gentleman who smokes a pipe as the bakery owner?" -- but unlike on-the-spot random generation, these NPC names have floated around in your head for a bit, so you're using something with which you ought to be a bit more comfortable.

For warehouse contents, it's probably counter-productive to track what is where, because it won't stay there for very long. Instead, I'd suggest a more generic system:
- Did a cargo shipment just come in? Probably some well-stocked warehouses near the docks and/or gate.
- Did a cargo shipment just go out? Probably a few (mostly) empty warehouses.
- If the warehouse isn't empty, then just roll randomly on a table which has a few perennial goods and a few seasonal goods.

For the 45 city districts, my advice would be:
A few specific landmarks. How do they look / sound / smell?
A few general traits. These could be in stark contrast to the landmarks, or not.
Who lives in the district? What do they think of the city overall?
Smells. Anything notable in this district?
Sounds. Waves & gulls? Blacksmiths? Shills? Elysian thrushes?
Do the people in this district have any strong opinions about their neighboring districts?
Is there one (formal or informal) governmental structure, or are there multiple factions?
Does someone "run" this district? Who are the leaders / spokespeople / "big men" / bosses? If a new business of some type were to open up, whose approval would be needed? How does this "boss" relate to the legitimate city government?


... er, my advice would be to use fewer than 45 districts. Having 20 districts is probably overkill for even a moderately large city.

Maybe start with a smaller city, say 6-12 districts, and see how it goes. Refine your process and apply it to a larger city. Experiment, adapt, escalate.


Overall
- Always keep in mind that you're going to throw away a lot of what you prepare.
- Steal what you can, including from yourself: recycle unused content.
- If prep work turns into work work, do less of it.

A DM's prep work is fuel for the fires of adventure, not fine art for display.



Have you tried talking to your players and asking what they're interested in and what kind of details they want in a game?

Humans are terrible at introspection.

Surveys are difficult to write because people don't respond to questions about preferences in isolation -- you need to be aware of a lot of unconscious bias effects when you want to get data out of people. Here's one discussion of a few of them: http://www.pewresearch.org/methodology/u-s-survey-research/questionnaire-design/

Bottom line is: unless you are good at the art & science of writing survey questions, you're unlikely to get a useful answer to this sort of question by asking directly.


There is a very good indirect way to ask, though...

Build the city together with your players. This is "session zero", before PCs are formally created.

The Dresden Files RPG has a fantastic chapter on cooperative city-building. That book is well-written and fun to read, especially if you've read the relevant novels, so I recommend picking it up if you have even the most passing interest in that franchise.

If not, you can find the process outlined in various other places online. Here's one that looked pretty good to start with: http://www.rickneal.ca/?p=599

The players will tell you what they're interested in seeing as they help you build their own doom the world around them.

It's fun for me as a DM, and it seems to improve player engagement in the setting.