|
Notes on Information Architecture When it comes to designing a game with many different layers of components and so on, it rapidly becomes critically important to figure out how all the information is surfaced to players and indexed so that it may be findable in the future. Particularly for a megadungeon which has multiple points of entry, large and small zones, and denizens, the amount of information that is generated in prep, execution, and aftermath of a single session can be rather daunting. This is one of the main dilemmas, I believe, of megadungeon and other large scale design with a focus on game play. Part of designing in a ‘mega’ context has to do with being constantly in a flow of creation, either iterating on ideas or making new ones as play proceeds and new nuggets of information are generated. Invariably, there will be multiple cohorts of players in how they engage with information at the various phases of the game. When you present narrative information, it is unlikely that all of it will be retained, and unless there is a mechanical system in play to encourage players to document what it is they found, then the end result is what the Megastrata project ended up with, which is document soup. Information management and design is not a unique subject to the game design space, but the majority of people do not find great reward in tracking down information that they might not even know exists in a variety of information systems. Here’s a brief examination of what was used in the Megastrata Project and some takeaways: - Discord. Previously used for active discussion and semi-static documentation, this actually was just a really dumbass idea because Discord is awful at information presentation. It’s great for holding conversations, but in mixed player cohorts and no specific guidelines for discussion, conversations tend to veer off into irrelevancy or get to the point where information is lost in hundreds of lines about all kinds of other discussions. This penalizes players who are less active for whatever reason. - Google Docs. This has the opposite issue of Discord in which information is contained in one or more docs which are not well indexable or searchable unless you create a landing document for everything, which must then be manually updated with new information, creating a large upkeep gap. The amount of information generated in various documents also becomes less accessible over time, because it’s unclear when a key term is leveraged in one document to another, and things that would be assumed common knowledge, even if they are in centralized documents, actually get ignored or flat out missed. - Confluence. This is moving in the right direction. A centralized wiki with a robust search index and cross-referenced multimedia pages is something that is great, but Confluence itself is a paid piece of information management software (or cloud hosted), which means unless you are running an actual business, maintenance costs over time get somewhat unwieldly. That said, it is enterprise-grade software with all the ups and downs that entails. Disclaimer: I work on and support Confluence as part of the company that produces the software. - Trello. Aside from tool porn, this is useful as a measure of workflow transience but not actually practical for most to all cohorts to participate in. Meaning, it would be great if people actually leveraged it, but everybody needs to leverage it, and if they don’t it’s easy to miss changes because each information node (cards and lists) is transient by design. - Tabletop Simulator. This is as close to a physical pinboard that you can get that displays the information to people, but it requires a host and session to be active for people to review information, which is not accessible if the host is not available, someone manipulates the data by flipping the table or weird physics interactions, etc. - WorldAnvil or similar. This takes the idea from Confluence and provides a web-accessible wiki that has more specific tools for information management with a generally cheaper license. The editor is more task oriented but also a bit confusing, which means anybody curating the information has a learning curve compared to the other details noted above. - Roll20. If you’ve ever tried to have multi-tenancy in Roll20 and multiple people updating documents at the same time, you know how much of a nightmare this is, in addition to a confounding UI. Asset management in particular is abysmal unless you do things in a very specific way. It also requires some level of money if you want to leverage certain features, but I don’t consider that personally to be a limiter, it’s mostly the ease of use and access that drives me up a wall. - Physical notes. Mostly self explanatory, but not tenable for sharing in an online medium. It would work fantastic in a live in-person setting as long as the information was accessible in a common space in between sessions. Ultimately, I think unless the information architecture and design is in multiple systems it rapidly decreases accessibility and increases frustration proportionally. The most cohesive framework would be WorldAnvil or Confluence, and based on price and specific features for game design, WorldAnvil and its similar derivatives seems to win out even if its editor is not that great. In failing to consider the information design and presentation this also means that things that happen rapidly get forgotten as soon as there are delays in gameplay, or easily missed if there are major things like multiple Mysteries and so on. I think having a truly centralized information hub is the most key thing that can be offered, but the lift while the game was running was simply too great to undergo not just one or two but three information rearchitecture and refactoring.
|
#
?
Jul 5, 2021 07:37
|
|
|
|
| # ? May 24, 2024 23:59 |
|
|
ive been running ptolus for a dedicated group for the last year, and other than the banewarrens (which theyre only about 60% of the way through and are currently locked out of further progression in because someone else has the key and is waiting for the heat to die down) they havent actually done much dungeoneering at all. they took one attempt at it, got kind of owned but still won, went back to the surface and never really went back. which is funny because they got a lot of treasure from their one incursion into the dungeons, but theyre always like "we feel so poor" and when i reply "well you could delve. 2 of you are registered paying members of the delvers guild" they just fart around instead. in fairness, theyre pretty locked into many, many threads. ive been keeping every possible ticking clock in the setting running, along with a few i added myself. its funny to see them completely no sell the plot hook that would take them into the night of dissolution, but 1000% excited for the debut of a musical starring the grandchild of the citys mafia boss in 9 in game days from where they are now. how did you get your players to actively seek out things in your games? i feel like a lot of the time mine expect important stuff to just come to them, when the point of this setting is for them to be in this place where they can dictate terms. if they wanted to stop adventuring for a few years and open a pub, then sure. but they just sort of go "uhhhhhhhh" waiting for someone important to show up and tell them what to do. though in fairness, 3 of the 5 are pretty new to roleplaying.
|
|
#
?
Jan 24, 2022 13:45
|
|
|
In a mega-dungeon setting you just need to be clear that nothing is going to come to them. The mega-dungeon is an entity that waits for things to come to it, and if they're hard pressed for things to do in there since there's stuff going on outside of it, that's kind of not really about that place and situation anymore, it's more of a general advice and game. In a megadungeon context if you get "the heat" you should be able to go somewhere else in that space to go explore and loot while that heat dies off. It is narratively interesting to have a dynamic and random encounter where the heat instead increases but also has corresponding additional rewards, like abruptly bringing the story to a head. If they need someone to tell them what to do, that someone should say: "Go to the megadungeon". If they don't want to go, it's time to hit pause on that game and talk out of character what the table wants and to realign expectations, because passive playing in a megadungeon is absolutely not going to cut the mustard. Instead of plot hooks, give them things with clear reward structures, all of which require exploration in the megadungeon or ultimately lead back to the megadungeon. From a referee standpoint there is nothing actually wrong with players being fixated on tidbits instead of "the main game", but there should be mechanical reinforcement to spur players to doing what they want, and positive mechanical reinforcement, eg. rewards to do so is a cornerstone of that. Also, as the referee, you can just unlock poo poo. It's a narrative that you have a direct hand in shaping. Just say that thing that's locked is unlocked and play to find out what happened after that. "I thought there was only one key to unlock this thing. Why is this open? Uh oh."
|
|
#
?
Jan 24, 2022 15:05
|
|
|
If the PCs are actively seeking out things that aren't murderhoboing, it's a blessing. It sounds like you need to put something they've just gotta have down in the dungeon.
|
|
#
?
Jan 24, 2022 16:02
|
|
|
At the end of each session, ask them what they want to do next session. Present a small number of options. Two ticking clocks is an interesting choice. Thirty ticking clocks and there's realistically no way the players can ever get to them all, so why bother? It requires "telescoping" the schedule, but things the players are interested in should generally happen soon, rather than forcing them to wait a certain number of in-game days. This may not be suitable for the kind of plot heavy, clock driven game it sounds like you're running. But based on your post it sounds like the players may not be interested in a plot/clock heavy game either.
|
|
#
?
Jan 24, 2022 18:28
|
|
|
Another possible dynamic, which is not specifically related to a megadungeon, is that your play group may lack a member willing to step up and make decisions for the team when nobody has a strong opinion about what to do next. Some folks may feel like just telling the rest of the group what they're going to do next is rude; if everyone feels that way, a group can wind up lacking decisiveness. There are tricks and tools you can use to sort of press the players to make a decision: things like, rotating a Party Leader role; reducing choices to a small number with "nothing" not being an option, the way a choose-your-own-adventure book does it; having a "threat" pool that gains tokens every time the players waffle, and it triggers a dangerous encounter whenever the pool is full; just charging the characters rent that they must adventure in order to not wind up out on their ears; or perhaps just the classic out-of-character table discussion in which you explain the party need to get better at reaching decisions or the game won't work and then asking them for suggestions.
|
|
#
?
Jan 25, 2022 02:20
|
|
|
Hi I was linked to this thread in GenChat and am digging it up again. Megadungeons have gotten into my head as of late. It may be the Delicious in Dungeon anime, which clearly uses that as a structure. I'm also fond of an old Knights of the Dinner Table arc revolving around them running the "Biggest drat Dungeon Ever!" (likely a spoof on the then-recent World's Largest Dungeon) which turns out to be a broken mess because Hard Eight rushed it out to meet a deadline. (And while not specifically about megadungeons, the 4e Underdark supplement does a great job of presenting a setting element that's literally built for such things.) Whatever the origins, it feels like a neat design challenge. I used the old World Builder's Guidebook to roll up a straightforward D&D realm and the lore is gonna come from there. I have no group or campaign for it, though I will be using 4e for the mechanics. Part of the appeal to me, going back to that, is just that it is a nice novel twist on the traditional campaign structure. Instead of going around the countryside clearing dungeons either for profit or for story reasons, you find one place that is a nexus of evil and danger and opportunity and slowly uncover all its secrets. There's some of this in the various "dungeon crawler" RPGs that have existed since the likes of Rogue and Wizardry. And I think also one of the fun things is how a megadungeon changes as you go deeper. It's not just one place- as you delve further there are many regions and conflicts going on, the landscape changes, there are multiple little communities that have sprung up, etc. (I'm not sure if I will have a mechanism to go back to the surface or if rather they'll find "friendly" areas along the way. Both approaches have their ups and downs.) There's a lot to do and it'll be slow work, I think, but I'll share what I come up with.
|
|
#
?
May 12, 2024 00:16
|
|
|
Maxwell Lord posted:There's a lot to do and it'll be slow work, I think, but I'll share what I come up with. I guess I'll respond to you here instead of GenChat then, but I've been obsessed with the 3d6 Down The Line podcast's AP playthrough of Arden Vul lately. Arden Vul is a really amazingly well-crafted megadungeon (if very intimidating at ~2k rooms and 1122 pages long) and being able to go back and forth between the book and an example of how it actually plays out in practice is super fascinating for anyone into megadungeons. I like the idea of a 4e megadungeon a lot, although you'll probably want to be doing a bit of re-examining core principals as you work on it. A lot of what's really cool about megadungeons is the interconnectedness of everything and the gradient of danger that exists within it. The healing surge system and so on within 4e can be really well suited to a megadungeon setting (although you need answers to the whole '15 minute adventuring day' issue), but the AC treadmill could easily create situations where only a sliver of the dungeon is reasonable to explore at and given player level (either too easy or too hard). I'd suggest looking at the math on all that and possibly even just making a mini homebrew where you remove the treadmill entirely. I'd double check the results, but the math is pretty straightforward, so it might be as simple as subtracting level from everyone's AC/Def/Atk scores. I'm real excited to see this thread opening up again. I ran a megadungon named "Palace of the Sun King" here on SA/elsewhere for a while in ~2013 (no relation to "The Sun King's Palace" mentioned in the other thread). I've been missing it lately, but my sensibilities have changed a lot in the decade since, so in the past month or so I've started working on a replacement. I'll probably make an intro post about it and the process I'm using to make it a bit later--maybe some fun design cross-pollination can happen.
|
|
#
?
May 12, 2024 17:29
|
|
|
A 4e megadungeon sounds like a good fit for the paradigm aldentefax describes at the beginning of the thread. A system of appropriately scaled realms with encounters that can be completed by parties of various levels, rather than a naturalistic dungeon where you can run into stuff you have no chance of beating.
|
|
#
?
May 12, 2024 18:06
|
|
|
Maxwell Lord posted:Hi I was linked to this thread in GenChat and am digging it up again. The World Builders Guidebook (from 2e D&D right?) is so good. It's what got me into generating everything using tables and then getting into making a megadungeon to populate below it. Which way did you work through the book, local to world or world to local? If you like to generate a big world to populate first, I recommend using How to Host a Dungeon. It's a solo game where you create a large-scale dungeon and its ecology with dice rolls. It's a little clunky and it took me a couple tries to get something I liked but I use what a generated as my megadungeon.  This is the result I was most satisfied with, and I use it as my megadungeon world map. I also wrote out a history of major events during the creation of my megadungeon so various factions exist with pre-existing enemies in the dungeon itself. I can now work on various parts of the dungeon when I want and having different environments keeps me from getting bored. I've been using this for several years and come back to it every once in a while. I like having the mega dungeon world map from How to Host a Dungeon because it allows me to keep the same dungeon but use multiple systems. I started in 5e D&D and now use Pathfinder 2e.
|
|
#
?
May 12, 2024 18:25
|
|
|
mellonbread posted:A 4e megadungeon sounds like a good fit for the paradigm aldentefax describes at the beginning of the thread. A system of appropriately scaled realms with encounters that can be completed by parties of various levels, rather than a naturalistic dungeon where you can run into stuff you have no chance of beating. That's fair--I might be a little in my own head on some of this megadungeon design stuff. Like, one of the big dynamics in the Arden Vul podcast I was talking about is that there's this constant progression where the ways the players play is constantly changing; at first they're going into mid-level zones as level one characters and treating it all like a heist game that's all stealth and trickery, but then as they level up they start going back to those zones able to stand toe to toe with their old enemies and start toppling rival factions. Level inappropriate foes can be really fun to engage with--they're just not combat encounters--and then it feels real satisfying to deal with them once you're actually strong enough. You're right that that's just one way of flowing all this, though, and not one that lines as cleanly up with 4e's default style of play (although it's not entirely impossible). There are whole other ways of taking advantage of the megadungeon structure that probably fit better. Anyway, here's my dungeon and process so far: The Thrones of God Concept The Gods sat atop humanity for as long as recorded history, demanding tribute and worship from their great peninsula-bound temple complex. Five years ago, a small local revolt shocked the world by overthrowing their divinely-appointed ruler, then snowballed into a global multi-species revolution against the holy empire, during all of which the gods remained strangely silent. A week ago, massed armies from across the continent stormed the Palace of the Gods, breaking down the gates and swarming within. After several days the sounds of battle, unearthly howling, rumbling explosions, and much stranger noises died down, with no survivors from either side emerging to claim victory. Now, a second wave of visitors begin to arrive outside the sprawling temple complex--treasure-hungry adventurers, fallen-behind rebels, reactionary loyalists, and other assorted misfits, each with their own reasons for wanting to delve deep into the dungeon and uncover its secrets. How I'm Writing It A lot of my dissatisfaction with my old dungeon came from the fact that I was writing it one floor at a time, with only a vague sense of what would be on deeper levels. This time I'm creating everything in sweeps--making an outline of the entire dungeon and the high-level concepts at play in each zone, then filling each zone in bit by bit piecemeal. I probably won't start running sessions or meaningfully keying rooms until I have at least a paragraph of concept-text and a handful of points of interest and notable NPCs written for each zone that'll exist in the final dungeon, to let me do more foreshadowing of later zones in earlier areas. I probably don't recommend this way of writing a megadungeon to anyone making one for the first time; it's a lot of work before you can run your first session, and the possibility of realizing that some core part of how you're structuring it doesn't work the way you thought it did is relatively high. I'm just setting a challenge for myself to go for maximum interconnectedness (the gods should all be felt strongly in all parts of the temple, even if their actual forms are obviously going to be extremely deep within). Part of that interconnectedness is meant to be physical--I'm trying really hard to create a map that balances a ton of interconnection between zones with genuine roadblocks that make areas difficult to get to at first. That means lots of natural landmarks on the surface that mess with traversal--steep cliff-faces, raging rivers, etc. Things that allow sequence breaking (especially as the players gain knowledge and abilities), but that are difficult to reliably and safely cross at low levels. I started with a real rough sketch of the surface:  Then I started brainstorming surface zones. I've been using Obsidian--it's incredibly useful for big projects like this. I just make a page per zone I want the dungeon to have and leave any thoughts I have in the moment on pages as I have them, and also generate a bunch of 'scrap' pages where I can dump random shower ideas/randomly generated encounters that aren't appropriate for the place I made them (more on this later). I then use its Canvas view to try to start determining which ones should be reachable from which other ones.    Finally, I made a photoshop (well, GIMP in my case) file for the map. I'm not going to worry about exact room placement for a while, but it's still useful to try to nail down zone placement, so I drew an outline of the plateau and blocked out where each zone will be with big color blobs, then used a black line to mark which zones will be easily accessible from which other zones. You can probably figure out which zone is which if you compare this to the obsidian canvas view above.  It's a fair amount of work, but it's been extremely useful having these guides for me as I start brainstorming what the zones will actually contain. Starting with a map of what you want to connect to what before you actually start drawing the map definitely is resulting in more interesting maps than just free-handing it like I did in my previous megadungeon. I have a few other techniques and things I've been doing to help me come up with ideas (stocking megadungeons gets extremely tiring, and it's easy to fall into ruts if you don't go in with a plan), but I'll save those for a later post.
|
|
#
?
May 12, 2024 20:07
|
|
|
Thread necromancy! The design of megadungeons creating ecosystems that create situations without easy answers I think is part of the juice that I tend to enjoy. Resource management that is also informed by reputation is something that is a point of consideration (and also a core part of Dungeon Meshi). Finding the natural language to think about dungeons then translating them into mechanics will always be a balancing act. Delicious in Dungeon’s world design is very much inspired by Wizardry, and also by extension the earlier dungeon scoping of various forms of D&D dungeon design from the 70s and 80s, which traces lineage back to Castle Blackmoor. If I were to go and crack open the Megastrata again I would still rely on the old generative systems but add more detailed information for keys and locks to create that kind of metroidvania or Zelda type feel. Some of the design exercises to “create a bunch of rooms and then figure out what the throughline is” can work as long as there is an internal logic to them since players mastering a dungeon segment only to discover something that changes their perspective on the many other parts of the dungeon is juicy. The artistic pursuit of design and discovery for whoever is making that dungeon is also juicy! I struggle with using other content as is often due to the information density but appreciate them as functional works of art.
|
|
#
?
May 13, 2024 00:51
|
|
|
|
| # ? May 24, 2024 23:59 |
|
|
gurragadon posted:The World Builders Guidebook (from 2e D&D right?) is so good. It's what got me into generating everything using tables and then getting into making a megadungeon to populate below it. Which way did you work through the book, local to world or world to local? I started with Continental Geography and rolled up an Island Continent- didn't use any scale maps so I haven't fully got to grips with the size beyond "probably smaller than Australia." And this is a good segue to sharing some of the lore I've got. Basically a couple of thousand years ago this part of the realm (which is about the upper-middle, plains bordered mostly by mountains) was ruled by orcs. The Great Orc Kingdoms lasted up until a violent cataclysm- the "breaking of the world", a bunch of earthquakes and volcanic eruptions in the region. Most of the orcs took it as a sign and left for their original homes on the coasts, and a lot of the blame fell on King Rugash, who had built something called The Everforge in the side of a mountain. It's not clear what happened to Rugash and eventually a human-dominated kingdom rose in the area. (There are still orcs around, they're just folks, mostly assimilated into the local culture. Similarly dwarves, elves, etc. have "homelands" elsewhere on the map but they definitely show up here.) So the discovery of the Everforge, buried under volcanic rock, is gonna be the thrust of the thing. The mountain's in an otherwise kinda sleepy region, the legend is old enough that not everyone was sure where this was or if it was ever real, and there are plenty of other orc ruins around. The Everforge alone won't be the whole dungeon, it turns out whatever Rugash built ended up tapping into something much bigger and I can likely start drawing on 4e's Dawn War for material. Start out trying to yoink some ancient treasures, end up in the Primordial Chaos trying to stop the world being unmade. Probably should have a few steps in between. But anyway that's where I am right now.
|
|
#
?
May 13, 2024 08:48
|
|