8 votes

The next big D&D experience looks like a video game, and that's a problem

8 comments

  1. [4]
    PetitPrince
    Link
    So the whole "let's have a 3D version of D&D with live dungeon mastering and some environment creations" has already been tried way back in 2002, and it's called Neverwinter Nights. It has been...

    So the whole "let's have a 3D version of D&D with live dungeon mastering and some environment creations" has already been tried way back in 2002, and it's called Neverwinter Nights.

    It has been one decade and a half since I last launched this game or its sequel, but from my memories of the limited time as teenager I had playing and then building of some RP-based worlds, you can definitely have some great experiences even if the tool you have to use are rather limited.

    Random thoughts (also it's late and I haven't reread myself):

    • In the case of NWN, you cannot create zones on-the-fly (it requires reloading the game), but you can create "placeable" as you want (furniture, objects, creatures, graphical effects). Most of the time we played on known places (normal raidable dungeons for instance), but repurposed for another tasks in scheduled "DM event", e.g. all the creatures have changed and now this is the lair of a dark elf lord of doom(tm)).
    • It certainly skew the game toward more prep, but NWN had a good library of placeable, and it's flexible enough to evoke anything. Also, descriptions are still super important. It's not because it's 3D that it shows everything
    • Speaking of which: NWN is a 2001 game, and really have this low-poly esthetics that's basically PS1 low-poly but with better textures. I actually think that helps the suspension of disbelief because of its low fidelity; detailed enough to give you a good idea, but not so much so that you can still complete the picture in your head. For the D&D thing they specifically pointed out that they used a tilt-shift. I think it's a smart move to remind the player that what they are watching is just a suggested representation and not "the real thing". But now I wonder if nowadays tools are sufficiently good to easily create new content. For NWN it was easy enough that some people with no experience could create some pretty good new objects.
    • There's a recent video of a DM in the DM client showing how manage a given scenario. At one point he make the party fall out of a bridge, and crash on a mysterious cave with lots of debris (something that you definitely cannot do with the NWN engine). It's neat.
    7 votes
    1. [3]
      vord
      Link Parent
      There is NWN Enhanced edition out, complete with Switch crossplay. It holds up surprisingly well.

      There is NWN Enhanced edition out, complete with Switch crossplay. It holds up surprisingly well.

      2 votes
      1. [2]
        PetitPrince
        Link Parent
        Whoa, even on the Switch! ... No custom module though. Too bad, that's was (still is?) the bread and butter of what's interesting with NWN. But I do understand that player generated content...

        Whoa, even on the Switch!

        ... No custom module though. Too bad, that's was (still is?) the bread and butter of what's interesting with NWN.

        But I do understand that player generated content moderation would be a nightmare for Nintendo (my favorite modules were definitely adult oriented) (not a Dance with Rogue, but the Bastard of Kosigan).

        2 votes
        1. vord
          Link Parent
          PC still has custom, but yea switch is limited to curated content iirc.

          PC still has custom, but yea switch is limited to curated content iirc.

          1 vote
  2. [3]
    hungariantoast
    Link
    Emphasis mine. That's really the key difficulty in building a viable virtual tabletop program. I don't consider Roll20, Foundry, or anything else to be "viable" because they necessarily impose...

    So, I randomly thought it would be hilarious, epic, and thrilling to have Elbow use improvised explosives from his brother’s gunpowder stash to blow a hole from the surface down to the chamber where the PCs were fighting.

    Because I was using a writeable map with a dry erase marker, and simple tokens, all of the impact and thrill of this could just be done on the fly, with the theater of the mind doing the bulk of the “processing power” for the game. How easy will it be in this new 3D playspace Wizards will offer for me to toss in sudden changes and not have them feel like a wild departure? How quickly can I come up with secret rooms I didn’t plan for, because I might’ve been lazy in my prep, but in the moment thought of something that seemed interesting? How much do I need to know about 3D modeling to shape that? How can I describe a ratfolk blowing his way through the ceiling and potentially setting tons of shit on fire in an epic climax to a fight and have the 3D playspace keep up with the breakneck speed of improv?

    Emphasis mine. That's really the key difficulty in building a viable virtual tabletop program. I don't consider Roll20, Foundry, or anything else to be "viable" because they necessarily impose severe limitations on the "theater" of gameplay, even if the dungeon masters who use these programs don't recognize or feel those limitations.

    Fun fact, creating a "virtual tabletop roleplaying game program" is something I have been thinking about for almost a decade. I have written about my ideas twice on Tildes before (see below).

    I hate to be a pessimist, but I don't think this new offering by Wizards is going to succeed. I do not believe that computer science and software development are advanced enough yet to allow the creation of such a program. I think we're getting close to it being possible with advances in machine learning and procedural generation, but for a virtual tabletop program to actually add something to the "theater" of a game like Dungeon and Dragons requires that program to be able to deliver a representation of the theater, if not faster than, at least fast enough, and with equal "imaginative capacity", as the human mind.

    So, you know, something you could make in a weekend I guess.


    First old comment

    Dungeons and Dragons and the simulation of tabletop gaming.

    "But wait!" I hear you cry. "Roll20, Fantasy Grounds! These exist, give them a try!"

    "No," says I.

    It's so difficult to do, but I want it so bad, a virtual representation of the rules of D&D along with tools to facilitate detailed, on the fly generation of content.

    I want a nice grid with some cute ASCII characters to define players, characters, objects, just like a roguelike, and I want the DM to be able to quickly slap down cities, individual buildings, shops, shopkeepers, prices, inventories, descriptions, and all the other details they might need, and I want it all to be modifiable, scriptable, and interactive.

    Basically, think of a multiplayer NetHack, except, instead of dungeon diving, you're playing D&D, and your DM can just create the world you explore on the fly using procedural generation tools, as well as pre-building ahead of time whatever details they want, like using a level editor to generate a city before the game session, and then to generate specific shops, items, prices, furniture in that shop, the shopkeeper (race, description, etc.), and other details, but also to be able to design all that junk by hand.

    Needless to say, it's a massive, ridiculous amount of work. The technology and examples of most of the pieces of a program like this already exist, but putting them together and making sure they work within the realms of whatever tabletop game you're supposed to be playing, as well as work with other systems of the program, is probably exceedingly tricky. There's also just the fact that everything has to be cataloged and rebuilt inside the program. The "game" needs to know what a bag of holding is, what it does, and how it works.

    Then, once you're done rebuilding forty years of tabletop gaming, you have to write tools for players to be able to modify, script, and program things on their own.

    And really, what's the point? What's the advantage you gain that makes all of this effort worthwhile? I mean, when we are sitting at the (actual, physical) table I can just tell my players that they're going to be inhabiting a castle now. I can describe it to them, give them some bedrooms, a kitchen, stables, and a few other bits and pieces as it becomes necessary, and they'll understand it. I don't have to actually design the damn thing like I would in a game, and I don't run the risk of any gotchas from shit I didn't think about. Yet, the allure of this idea draws me in. It also isn't impossible to provide tooling and systems so that proper castle generation can be as fast as just describing it by mouth, it's just incredibly difficult. (The motto of this idea.)

    Still, you'd never have to bother with combat rolls, movement distance, or other nonsense again, but your DM does have to pay attention to the three-dimensional placement of everything so that combat doesn't get ruined.

    The biggest advantage though, and this is something you could only build, drop the rest of this idea, and still have a great program, is the procedural generation of just about any kind of content, right now that you could possibly want for running a session.

    How many websites and programs can you think of that generate a bunch of little things for tabletop games? You have city generators, non-player character generators, player character generators, world generators, religion generators, magic/spell generators, magical item generators, inventory generators, shop generators, a lot of damn generators, but I can't think of anywhere or anything that has all of them rolled into one single program.

    Shoot, just being able to generate just about anything I would need in a D&D game, on the fly, from a terminal, would be awesome. Forget all this "multiplayer NetHack, but it's D&D" nonsense, just give me the generation tools without the bloat.

    Second old comment

    Another idea I've been thinking about for a while and have also talked about on Tildes is a virtual tabletop program that specializes in the random/procedural generation of content for a wide range of tabletop roleplaying games.

    So, a big feature of the program would be that it's capable of generating content on-the-fly for a game master. Players opened a chest? The program can generate loot. Players enter a shop? The program can generate the shop's items, inventory, prices, the shopkeeper's character, and other characters who might work at the shop. Players want to travel to a city? The program should be capable of generating that city, including specific features of that city, like districts, points of interest, shops, population metrics, and major characters for the city (like captain of the guard). It should also be able to generate dungeons, encounters, player characters, non-player characters, specific types of buildings (like a castle), magical items, and all kinds of other stuff.

    What's wild is that, on their own, examples of software that can generate most of these things already exist, but they're all separate programs or exist on different websites. I've never even found an exhaustive compendium of all the online generation tools before, let alone a single tool that handles all or most of these use cases.

    Why? Because it's difficult to appropriately generate some of these things, and it's incredibly difficult to generate most or all of them in a relational way that makes sense.

    Still, that doesn't stop me from naively spending hours writing about the idea.

    Also, this wouldn't just be a program for generating stuff. It is a "virtual tabletop" program after all. So, in addition to generation tools, I imagine the program should be capable of simulating a tabletop with stuff like virtual spaces, miniatures, handling combat and movement, and all the nice things Roll20 or Fantasy Grounds try (but fail, in my opinion) to provide.

    Initially, I imagined that this program's "virtual tabletop" would look similar to a two-dimensional, grid based roguelike, where player characters would be represented by little ASCII "@" symbols on a digital grid, and they could move around and explore a virtual world designed by a game master. Basically multiplayer NetHack, but instead of being a roguelike, it's D&D, or Starfinder, or whatever.

    Having written about the idea more though, I think it would be a mistake (if someone actually built a program like this) to not go "all in" and have a not grid, not ASCII based virtual tabletop, but instead something more modern (though the ASCII grid could be an optional thing). If you've ever played RimWorld or Prison Architect, what I'm thinking is that the "virtual tabletop" would be, instead of a grid based table, something freely movable within from a top-down perspective similar to those games. That's probably an awful way to explain it, but hopefully you get the idea.

    I don't know, I probably sound like a crazy person, but I find writing about and explaining this idea to be difficult.

    6 votes
    1. Amarok
      Link Parent
      What would do for it is something like the verbal interface to the holodecks of Trek or the simulator of the Orville. The AI or whatever passes for a director inside the box has to be able to...

      What would do for it is something like the verbal interface to the holodecks of Trek or the simulator of the Orville. The AI or whatever passes for a director inside the box has to be able to improvise with the GM prompting it. What kind of AI training data would you get from all that GM interaction, I wonder? I feel like I'd also want some kind of intuitive... sculpting of the environment using my hands to iron out kinks quickly. If you've seen the Tilt Five (a quick demo) it seems like a step in the right direction from a hardware perspective.

      Networking multiple physical tabletops together virtually is a genuine upgrade to tabletop gaming, unlike roll20 and similar apps which are just poor shadows of it. It's extending one's game board/room into a shared, synchronized virtual space. It would certainly make finding reliable groups easier, or playing with friends across the country or the world. It's not hard to imagine extending this sort of digital tabletop into Kingsman territory. If you had the sense that everyone else was 'there' and you could visually interact around the virtual table that would sell me instantly.

      Telepresence is the internet's killer feature. I'd love to see that come to the tabletop arena as a melding of shared gaming tables. Wizards just isn't thinking about it the right way. They want to tap into the gaming market when they should be reinventing the segment they already have.

      2 votes
    2. FlippantGod
      Link Parent
      You might like Shmeppy. Dumbest way of getting a map up for an online group, turns out it's also the best way, in my experience. Edit: but I totally agree, a solution where everything can be...

      You might like Shmeppy. Dumbest way of getting a map up for an online group, turns out it's also the best way, in my experience.

      Edit: but I totally agree, a solution where everything can be generated on the fly is coming, very soon. I imagine it will still be grid based initially, for technical reasons, but the move to floating point, unrestricted 2d movement isn't that much more work comparatively.

      2 votes
  3. LukeZaz
    Link
    Honestly, I feel like this article is throwing the baby out with the bathwater. I've tried my fair share of VTTs over the few years I've played D&D, and each had their ups and downs, to be sure....

    Honestly, I feel like this article is throwing the baby out with the bathwater.

    I've tried my fair share of VTTs over the few years I've played D&D, and each had their ups and downs, to be sure. They never really interfered with the game or its immersion except in rare cases though, and to my knowledge they haven't really divided the community, either. I don't know why this would be any different. It seems this article wants to toss the entire concept of an extravagant, 3D tabletop out the window, without allowing for the possibility that it could be good.

    The latter portion of the article does at least have more salient points to me. The crux of how I see this is that a 3D VTT could very easily be good if developed well, but there's no guarantee it actually will be. The article mentions the possibility for microtransactions, as well as poor support for the game outside of it's new toolset, and those are very much valid concerns. These very reasons are why I won't be holding my breath for any of this.

    All in all though, I just think there's a lot of potential value for the game if it were set up to work well in digital environments as well, where the issues of math and ad-lib content can be solved by the computer so that the DM can focus on the story and roleplay. WotC could absolutely bungle this; there's plenty of examples of much-anticipated software turning out to be hot garbage, to be sure. I'd just prefer we not drop the entire idea of making D&D digital-friendly as a whole, when it could certainly work out very well if properly executed.

    3 votes