Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

Emergent Narrative

Giauz Ragnacock

Scholar
Joined
Jul 16, 2011
Messages
502
In my other post I meant like how the GM has at least a rudimentary adventure planned and then the players (including the GM) fill it in with details. What you sound like your talking about sounds a little vast to comprehend (I'm so far understanding that you want a few hundred hours of unique randomly generated side quests with simple goals created from just the play mechanics). It sounds interesting from a design perspective, but I would like to test a pretty stable alpha/beta before I would know if I liked it or not. Push forward with making the game, BRO!
 

turrican

Educated
Joined
Dec 15, 2010
Messages
34
Location
Commodore HQ
Das Schwarze Auge (Realms of Arcadia) 1 did it right. It throws your party into a coastal region of Aventuria with the long term threat of an orc invasion at hand, somewhat similar to the Fallout waterchip idea. In a non-linear fashion the party has to obtain map pieces to figure out the location of a MacGuffin before this invasion will occur.

The map pieces can be obtained in pretty much any order (I do not even think all of them are needed), and there are a lot of non-mainquest related villages, quests and so on to discover.

Everything in the game is handcrafted, and it feels very reminiscent to the actual pen&paper Realms of Arcadia that I played back then. I doubt any dynamic, level-scaled hiking sim or rouge-like can achieve this.
 

turrican

Educated
Joined
Dec 15, 2010
Messages
34
Location
Commodore HQ
I believe you're referring to the Realms of Arkania series, correct?
Yes, my bad, for a long time I did not even know that the games were released under some other name in English.

Personally, I prefer Blade of Destiny to Star Trail or Shadows over Riva. While the successors improved graphics, automap, travel on the overland map etc., the explorable world became smaller and the non-linear plot progression was also streamlined. Hence my comment that Blade of Destiny is pretty much one of the best approximations of pen&paper RPGs on a computer.
 

Retinue

Novice
Joined
Jan 6, 2012
Messages
20
I've never played a Tabletop gaming before but to my knowledge, tabletop games do not have an emergent narrative. What they have is a Regressive Narrative. A world full of preset narrative that regresses based on what the GM created.

To this effect, what you are really saying is you want CRPGs to be more Life Sims.

and consoles/computers will always have trouble replicating the social aspect of tabletop

Setting aside the real/fake player paradigm, computers are much better at this than tabletop unless you include only quality tabletop experiences. Especially when satirical narrative is excluded, tabletop can degrade or upgrade. Computer rpgs are simply what you see is what you get setting aside patches and mods.

The main advantage tabletop gaming has always had over CRPGs is the ability to adapt and react to the players actions through the use of a Game Master/Referee

Not necessarily. FFTA has a game master (AI) that adapts to the player's actions but that isn't fun is it?

It really depends on the quality of the reactor. The mechanics alone provide the opportunity, not the reality.

Templated individual quests plugged into templated story arcs are the way that I believe can achieve this. An Individual quest would be something simple like get an object, deliver an object, get information, persuade someone etc… Something that is very simple in its purpose and is templated in ways like who is the quest giver, what is the objective, and what opposition is there (if any). The individual quests can then be plugged into a larger story arc. For example a story arc could consist of quests to retrieve a number of objects and then to have them made into a powerful artifact.

Close but wrong. For example, Depths of Peril have this and 7th Saga had this but are either of those possessing emergent narratives compared to your basic CRPG? No.

What you're basically saying is that given the choice to mod, quality mods would flourish. But on a more targetted variety. It's a delusion set forth either by your own bias as to how great you are as a GM or you're really that great and like a lablel, you fail to take into account the not so great.

The proof of this is your statement of quests. Quests are not part of emergent narratives, they are there to plug in the flaws of narratives.

In truth, quests are the original quest markers for quests especially as CRPGs can't have their sandbox if they don't have their distractions. Just because it's less obvious doesn't mean it still isn't a quest marker.

In contrast Life Sims (setting quests aside), creates emergent narrative from your desire to be either a maid, royalty, swordfighter etc. Even quests creates interdepency. What severity of warrior do you desire? What severity of outlaw do you desire? From there, flawed as they are, life sims produce emergent narratives where the key flaw is there's no easy to use diary for you to plugin your own emerging narrative.

Say for example all of your pieces of artifact are initially locked away deep in underground dungeon locations for safekeeping. If the player is able to investigate and get to them quickly it is simple dungeon crawling. But what if the player is dawdling and doing other things or a faction opposed to the player has caught wind of the player’s plans and recovers one of the pieces first. The quest for the recovery of the piece would then involve discovering where that faction has taken it and retrieving it from them. A magical or crafting inclined player might be able to follow an alternate quest of recreating the piece needed or jury rigging some way to make it work without that piece.

It still won't have much of an emergent narrative effect. Play the demo for Empires and Dungeons and you'll see what I mean if you don't already.

What is used to fill the templates can be based off the player’s previous actions and decisions. The opponents they face can be pulled from things the player has interacted before such as reoccurring opponents or loose ends from previous experiences. The same thing can be done for quest givers and helpers. Rewards can be made so the player has an interest in doing the quests and penalties for not doing/failing the quests can threaten things the player has a positive interest in.

Again, 7th Saga/Depths of Peril. Still not much of an emergent narrative. GMs get away with this because even a great GM have flaws with connecting a narrative without some form of swap. Only instead of palette swap you have motive swap in the form of recurrance.

You could even have story arcs plug into each other like a story arc about slaying some terrible beast that needs a specific weapon/artifact could plug our example story arc in the place of retrieving the weapon/artifact.

You can but that would be a story arc. A bonus optional "good ending". VN makers call these shrine unlocks or something.

The idea is basically this: Create an action/unlock an action either on the next playthrough or the next arc.

The thing here though is that you are just basically sidestepping the creation of a sequel.

To create an emerging narrative, the player must be the reason for not only who the beast is but how and what form the beast came to be. Even with this, you won't guarantee that players won't hack it per their desire to unlock all paths in replays.

This was the problem when you used a walkthrough or replayed Dragon Valor and Phantasy Star III.

To truly sidestep this, you need to create the opposite of a chess world which is a Go world.

Problem is many perceive chess to be complex and wrongly used, the complexity of Go ends up becoming Othello or worse, kills the power fantasy player from reaching the utmost power which just causes them to restart.

Another game mechanic that would be useful would be some sort of investigation mechanic. To help determine what kind of information you are able to get as opposed to the quest giver telling you everything you need to know. It could encompass things like research, social interaction, and spying/thieving. If it were set up as some kind of game mechanic as opposed to individual quests doing that it could provide the player more options/flexibility on how they want to handle something and fewer things that need to be specifically written.

Investigation mechanics are cool but again your GM experience is blurring your idea of emergent narrative because you pre-know a certain form of outcome.

In true videogame implementation, investigation mechanic has little to do with emerging narrative.

It would also be helpful to have game mechanics in place that allow the player to dig deeper and get more information from a scene (provided they have the skills in place to do it). This should help with the repetition since it makes it more like a game mechanic and people tend to notice repetition in game mechanics less than in specific descriptions.

Basic background impression should do. For a hint of this, see Choice of Romance and play up to Choice of Intrigues.

A rich person would have an impression of a rich person npc which in turn would create a rich person quest.

Comic Book Hero also has an interesting take on this. You as a new hero must befriend certain heroes to unlock the locations of villains as one option. Another option is to seek specific villains with a low percentage of success. Finally you can just go on patrol.

As a mechanic, it's unoriginal and basic. Potentially though, what you really want to do is step away from investigations and again fall back on Life Sims. Let the Life Sim create your alignment and focus and the only real aspect you should switch and switch the classes concept of Life Sims and turn it into quest ladders. From there you will get your emergent narrative but I won't say you would like the final result. See the complaints for the flash sandbox game Caravaneer.

Even with a walkthrough if the game designer can't handle the emerging narratives, all he has created is a sandbox but the worse part is that regardless of his talent, if his audience can't be receptive to emergent narratives all he has created is a sandbox.
 

NewFag

Educated
Joined
Nov 16, 2011
Messages
97
Fuck my eyes. Too much TL;DR.

Would it be possible to put a list of difference scenarios into a randomizer, randomly assign a new goal, path, obstacles, during each playthrough, and tweak the fine details with more randomness? In other words, create all the puzzle pieces, but procedurally generate the final picture. Choose ten pieces from a bucket of a thousand.

New Game (Minor objective pulled from a hat. By accomplishing objective, events lead to a main quest. Which is also random)
Main Quest Part 1 (Quests can be radically different, but sorted into general themes, also randomly chosen)
...
Main Quest Part 10
Conclusion (Every playthrough would be different, except some quests or NPCs might overlap depending on chance.)

Certain systems would be put in place to "adapt" to the player. If the character is overtly strong in combat, insert a quest from the nightmare-combat bucket. If the game notices most of the skills chosen are non-combat, go with more combat-light speech/persuasion quests. Etc. If the PC is rich, spawn some jew NPCs to lower his fortune by a notch. Or make a communist revolution and have the PC's wealth be taken away.

Probably would take way too much coding and effort to be feasible though. Just a thought.
 

kris

Arcane
Joined
Oct 27, 2004
Messages
8,852
Location
Lulea, Sweden
I've never played a Tabletop gaming before but to my knowledge, tabletop games do not have an emergent narrative. What they have is a Regressive Narrative. A world full of preset narrative that regresses based on what the GM created.

You actually even made me uncertain and forced me to look up the words for all their meaning. PnP clearly is closer to a emergent narrative, although I am a bit unsure what an "regressing narrative" really is in this aspect.

But as far as this goes it is different from group to group.
 

deus101

Never LET ME into a tattoo parlor!
Joined
Aug 18, 2010
Messages
2,059
Project: Eternity Wasteland 2
You actually even made me uncertain and forced me to look up the words for all their meaning. PnP clearly is closer to a emergent narrative, although I am a bit unsure what an "regressing narrative" really is in this aspect.

But as far as this goes it is different from group to group.

He see's modules and setting as "goto instructions" like a diy adventure.

How he lables them as preset is beyond me, since what part of the setting and module you have to look up is based on a string failures and successes.
Nor is the setting and campaign that fleshed out, it simply set the states of various actors, the GM must fill in the blanks which also leaves him room and the players room to improvise.

How much and how well this is done is based on human factors though, but you should be able to overthrow regimes and suddenly start a war inside a bag of holding damnit.
 

Giauz Ragnacock

Scholar
Joined
Jul 16, 2011
Messages
502
Retinue: I BROFISTED you because you've gotten a lot better at being concise. Just asking, but is English not your first language? Also, use obscure terms like "regressive narrative" sparingly unless you can give a little parentheses () description next to it and people will understand you even better.

Finally, when you want to quote someone type in "BLANK said:" and then hit the QUOTES (") button on the white bar of the poster reply bar to give you a quote and end quote box to copy and paste stuff in between. You can even go back and edit this into your post to make it easier to read.

Other than that I don't have anything to add to the discussion, yet.
 

Tel Prydain

Augur
Joined
May 31, 2010
Messages
123
I like the idea of it, but I’m not sure I would like how it would turn out. I think that the danger of procedurally generated content is that it tends to feel soulless.

I think that, as the honourable Blobert said, the inherent structure of some RPGs do offer the opportunity for emergent narrative, even if some of the specifics are pre-planned. Fallout really is great example, where you can choose to follow a ‘quest’ or alternatively forge your own path with the gameplay mechanics.
Even Fallout New Vegas manages that… you can follow the path of your would be assassin (which will give you a wider range of options later on), stumble upon him or ignore him altogether.

I’m open to the idea of procedurally generated content, but I would imagine it being a supplement to the pre-planned content.
 

DraQ

Arcane
Joined
Oct 24, 2007
Messages
32,828
Location
Chrząszczyżewoszyce, powiat Łękołody
Templates
I think that in order to work best, the templates should be as loose as possible and try to avoid relying on explicit scripting. For example if you want to have a competing faction intercept a McGuffin, instead of scripting so0me ambush or something, you'd just trigger mechanics assigning a band of people to this quest and they would trigger their own submechanics responsible for what means do they use, while using as much as possible unified mechanics.

Some form of world simulation would be necessary to help provide pieces to plug into the quest/story arc templates. It would also make the world not seem like it is just waiting for the player to interact with it. I don’t think it would need to be very in depth. Just enough so that the player can see that some things are happening independently of them. You would have to be careful not to simulate too much and making the player feel like they aren’t really impacting things.
I disagree. You'd generally need quite in depth simulation to have stuff interacting in non-trivial manner.

Another game mechanic that would be useful would be some sort of investigation mechanic.
For example you could have an act that might need investigating drop a clue object, linked to that act, and with attached appropriate in-world object that would serve as this clue. If the attached object got destroyed by its own or other mechanics, the clue would destroy itself. That's if you want the knowledge to be trackable by the program (for example you want to use evidence to persuade the law), for most applications just generating clues with no linking to the event would suffice.

I think you would have to drop traditional dialogue.
Topic-based (where/what is...?) and action-based (persuade/intimidate/flatter/insult/present evidence/...) would work just fine. Ocassional dialogue trees might be used if the situation is unique or the dialogue can be made generic enough.

Overall, :bro: .

Retinue: I BROFISTED you because you've gotten a lot better at being concise.
Too bad he got even worse at making sense.

:rpgcodex:

I think that, as the honourable Blobert said, the inherent structure of some RPGs do offer the opportunity for emergent narrative, even if some of the specifics are pre-planned. Fallout really is great example, where you can choose to follow a ‘quest’ or alternatively forge your own path with the gameplay mechanics.
The only "little" problem is that you're not using gameplay mechanics. You're following scripted options, statchecks and so on. If you try something out of the box it will only work if the devs happened to to think about this particular approach as well, if they didn't - tough luck, no matter how logical and brilliant was your idea.

I’m open to the idea of procedurally generated content, but I would imagine it being a supplement to the pre-planned content.
I'd go the other way around - make a good procedural sandbox, then spice it up with as much handmade and storyfag content as you can fit in.
 

Gregz

Arcane
Joined
Jul 31, 2011
Messages
8,594
Location
The Desert Wasteland
The only answer to this dilemma is LARPing with yourself. :bounce:

In all honestly we shouldn't be so ashamed to do it, I mean...what is reading a novel but imagining a bunch of words and equating them to the feelings and actions of real people in our heads?

I like a good combat ruleset with lots of depth, plenty of lore as ambiance, and then I let my imagination fill in the blanks.

To 'solve' the problem the OP describes you need self-aware level sentient AI. As the OP pointed out, the Dungeon Master changes the gameworld to the mood and needs of players on the fly, it's completely unrealistic to expect a computer to be able to do that, even in the distant future. (I.e. we will probably never see something like that in our lifetimes)
 

Giauz Ragnacock

Scholar
Joined
Jul 16, 2011
Messages
502
DraQ: Well, I felt Retinue gave it a shot. I hope my suggestions can help him, but everything comes down to him settling on specific questions instead of this all over the place with his words stuff.

Back on topic: I think I agree with Tel Prydain as far as algorithmicallly-generated content would be soulless. There is no fine crafting to add any context to the quests, just a shallow "the task at hand." If I truly wanted to dig into CRPGs, I would not expect to keep squeezing new game content out of them (except for maybe making a future replay nuanced). If I want a new adventure I can wait until I get an entirely new game at Christmas or for a birthday, check out a few interesting reads at the library, or take a chance by ploping a few bucks for something on GOG.

On another note, has anyone tried just like "free-roleplaying" (no real rules just pretend) with one of those near-Turning Test-passing chatbots (I remember reading an article that claimed two chatbots did pass it, but I have no idea what came of it)? How complex of a role-play can you get? Was that fun?
 

DraQ

Arcane
Joined
Oct 24, 2007
Messages
32,828
Location
Chrząszczyżewoszyce, powiat Łękołody
The only answer to this dilemma is LARPing with yourself. :bounce:

In all honestly we shouldn't be so ashamed to do it, I mean...what is reading a novel but imagining a bunch of words and equating them to the feelings and actions of real people in our heads?
Do you LARP that characters done something differently than described in the novel and different stuff happened if you're displeased with the course of action as written?
Maybe cross out undesired part of the text and write your own replacement on the margins using console pencil?

To 'solve' the problem the OP describes you need self-aware level sentient AI.
You don't. Real life provides perfectly plausible emergent narrative without.

DraQ: Well, I felt Retinue gave it a shot.
He did, but he missed. His explanation of "regressive narrative" was just as cryptic as the term itself to me.

Back on topic: I think I agree with Tel Prydain as far as algorithmicallly-generated content would be soulless. There is no fine crafting to add any context to the quests, just a shallow "the task at hand."
Well, you can put handmade content in an emergent, procedural sandbox. As much as you want, you just have to limit scripting and railroading to an absolute minimum, and specify end conditions well, otherwise the storyline will break if you, for example, were sent on an epic quest to retrieve missing key to princess' chastity belt, but chose to disintegrate the belt instead with magic.

If I truly wanted to dig into CRPGs, I would not expect to keep squeezing new game content out of them (except for maybe making a future replay nuanced).
But have you ever thought of some clever way to solve in-game situation, but couldn't?

Emergent narrative is the answer to this problem, because instead of predefined solutions it relies on specified goal state and a lot of simulative mechanics between you and that goal.
 

Giauz Ragnacock

Scholar
Joined
Jul 16, 2011
Messages
502
Simulative mechanics are fine when they really add to playing the game like a puzzle like that. Using the example of a locked, thick metal door on the outside of an "impenetrable" fortress. The lock can be picked if you have mastered the proper set of lockpicking techniques and have the proper tools. The lock has a damage threshold of 3000, so use any combination of techniques to break the lock (however, something like Fire Ball will only begin building damage slowly after you have heated the lock up with many consecutive Fire Balls). The lock also just so happens to be made of a certain material a magic stone you might have found can wipe from existence. Blow a poison cloud into the lock and use a Lightning Bolt to charge it enough to ignite the gas and blow the lock off. You could also look around for clues and get information out of people to find out if you can get a key to the lock without much hassle.

On the other hand, actual emergent narrative, as in building a pseudo story that might be coherent and possibly amusing by just throwing out random stuff like this guy needs saved, escort him, retrieve this item, fedex this, talk to her, protect the castle, lead the invasion, etc.--- I just don't see it. Dwarf fortress would probably be the poster boy for emergent narrative, but the stories that come from it depend greatly on the writing ability of the person who posted the story, otherwise there's really no fulfilling narrative to experience. All that remains is what you could come up with entirely without the game.

No, I would still like a writer to put a lot of thought into making the game feel different from other games with their particular writing, settings, characters, etc. I guess the only thing I actually disagreed with you on was the use of the term "emergent narrative."
 

lefthandblack

Arcane
Joined
May 5, 2006
Messages
1,287
Location
Domestic Terrorist HQ
I'm thinking of something very similar for what I'm working on. In thinking about it, I have come up with a couple of commandments that would have to be established in order for it to work:

1) Damn overarching "main quests". Damn them to Hell.

Instead of a MAJESTIC main quest, come up with a well fleshed out backstory that is supported by your game's world/atmosphere. The factions aspect of your world would be the most important component. You can write quests that allow the player to interact with these factions working in concert with the backstory, but never let the player become THE CHOSEN ONE. This leads directly to commandment two.

2) The player is just "some dude."

I often find myself enjoying the side quest's in CRPGS much more than the main quest. The main quest in alot of cases just feels like an anchor dragging down my enjoyment of the game. Think of the beginning of Freelancer, your very first quest objective is GET A JOB. Brilliant. Ok, player, here is the world> these are the factions>this is how they relate to each other>
this is what's happening in the world> Now go forth and do stuff.
 

Giauz Ragnacock

Scholar
Joined
Jul 16, 2011
Messages
502
To each his own, lefthandblack. To me your idea (READ: the factions every other super-immersion guy on the internet wants) sounds just so similar to many others and broad.

This does not decry the idea, but I think you will need a lot more bullet points than what all the other ones have (factions, you are just a small part of the world, thousands of handcrafted and randomized quests, 1 million items, 20 character classes, etc.) rather than just go and say that, "Well my game does them better" (than Depths of Peril, the upcoming Age of Decadence, TES games since Daggerfall, Dwarf Fortress, and probably quite a few others that have done all the things you want to do).

Whatever, either you or darkpatriot do, just my opinion, your games have to stand out from each other (setting, some darn well-written characters if you can, in-game era/technology, a range of motivations to play your game not limited to joining factions and building the most min-maxed characters possible, and whatever else can pleasantly surprise your consumers). Just don't stop with, "Well, we do these bullet points the best."
 

lefthandblack

Arcane
Joined
May 5, 2006
Messages
1,287
Location
Domestic Terrorist HQ
This does not decry the idea, but I think you will need a lot more bullet points than what all the other ones have (factions, you are just a small part of the world, thousands of handcrafted and randomized quests, 1 million items, 20 character classes, etc.) rather than just go and say that, "Well my game does them better" (than Depths of Peril, the upcoming Age of Decadence, TES games since Daggerfall, Dwarf Fortress, and probably quite a few others that have done all the things you want to do).

Yeah, those bullet points are just the ones that I've determined to be necessary so far. The one thing that I think is paramount is to avoid stupid situations like you have in the Codex's favorite game where you are dicking around with one faction or another while the whole world is being invaded by demons. I don't want to be the chosen one, maybe I'm happy just being in the thieves guild.

So how could you have the player do what they want to do and avoid the situations described above? Let's take the thieves guild example. Instead of going through a bunch of quests for them that have nothing to do with the backstory (demon invasion), write quests related to the backstory that make use of the talents of someone that is in the thieves guild. Maybe you have a powerful wizard that is allied with the demons, have a quest to steal the wizards magic foozle thus using your unique skillset in a way that directly ties into the backstory while still allowing the player to do what he wants to do.
Since the player is NOT the chosen one, there is no disconnect between completing this quest (for a handsome reward from the royal treasury) and the very next day robbing the royal tax collector, since you are just some dude who is out for himself and only helped the war effort because there was a personal gain in it for you.
 

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom