An experimentation in narrative gameplay

Design food for thought. Relatively long read ahead.

This is something I've been toying with since messing around with the NWN1 toolset (and now NWN2). The problem with those is that the RTwP system is kind of bland, and not something I really want to design around (some creativity could of course go a long way, but competent scripting would be required - I could learn it, but combat isn't my priority and certainly not the reason I'm making a mod).

Unfortunately, I'm not a scripter (which is, ironically, why I'm sticking around with the NWN toolset, since I've at least learnt the basics of NWNScript so that I may accomplish pretty much all I need to in that respect).

Now, I would have loved if the games were turn-based. I would have designed some modules around just that, no problem. Instead, perhaps out of 'necessity', I had to think out in what way I could implement gameplay that I didn't find fundamentally bland (given the limitation of being relatively unable to stray away from the default scripting of the game). A quick background: I'm not a 'hardcore' roleplayer. I enjoy good proper narrative and storytelling above roleplaying. That doesn't mean I don't appreciate roleplaying, and the two certainly aren't mutually exclusive. PST is my ideal game in that respect. It gives you some good choices to make (without simply using Bioware's triad of 'choices') while maintained a solid story frame. Fallout, on the hand, didn't grasp me as much, despite its greater roleplaying opportunities. Roleplaying *is* important, but I'm more concerned about roleplaying in the context of the game's story (though in no way promoting linear storytelling), rather than in spite of it. I enjoyed FO, but I enjoyed PST much much more.

And so, going back to the NWN toolset limitations, and looking back at conventional commercial RPG design, I pondered how I could implement gameplay that *I* would enjoy playing (again, against those restrictions). I came up with the concept of narrative gameplay.

Essentially, rather than using some sort of in-game gameplay system, I would implement everything via the dialogue system as a player interface. It's already there, I wouldn't need to create my own 'system', nor a new interface. In addition (and perhaps more importantly to me), I could put in some narrative directly into the 'gameplay' (I am, of course, not speaking of player-driven narrative, which FO can provide tons of, but it just doesn't grab me in the same way that the narrative of a good storyteller - hey, I like MCA - can).

What would the gameplay actually consist of? I would integrate combat, stealth, diplomacy (in the most basic sense) all via dialogue, driven by 'narrative'. I wouldn't be obsessing so much over stats, instead opting to emphasize the importance of *choices* within the game. This isn't really about character development, though, as that's a whole other discussion. For the sake of *this* discussion, let's assume that there is 'no' character development, and that the player character is simply average in all aspects.

For combat, rather than having statistics determining how successful you are (if I do implement stats, they'd be behind the scenes), it'd be more determined by choices during combat. I want every battle to be challenging, even simple 1-on-1's. Every battle has the potential to end your life if you're reckless. In that sense, combat may often involve avoiding getting killed, rather than killing your opponent. If you die, it'd be less of a 'unlucky dice, you die haha' but more of a 'you purposely put your neck on the line, good job'. It'd be text-driven ('duck behind the crates to avoid the archers' 'punch the idiot in the face' 'dodge to the side and try to parry the incoming spear' 'grab the chair and throw it' 'run away'), so the limitations of combat will basically be the limitations of the designer, rather than of something inherent in the combat system (in terms of how fighting and interacting with the environment goes).

Stealth would also be similar. Rather than being decided solely on your stealth statistic, your success in sneaking around would be determined with *how* you proceed with said sneaking. Hide behind cover, remain still, make some distractions, etc.

The real advantage of this (aside from being able to simply get to write) is that 'combat/stealth/diplomacy' can all be integrated seamlessly. You can try to negotiate your way out of dying WHILE fending off your opponent's attacks. You can create a quick diversion and get into hiding while fighting, you can punch an idiot in the face while trying to bargain with him, and so on and so forth. The only limitation is again based on what the designer imposes. Which is both a good and a bad thing, as you can't do 'everything'. The challenge is for the designer to provide a broad and flexible enough set of options so that the player can accept that. The advantage is that it'd be more narrative-driven (writing is indeed my favorite part of mod-making, but I don't want to stick to linear storytelling, I certainly believe interactive storytelling has a big place in game design, and is barely being exploited in our industry - and I really mean *interactive* storytelling, with actual roleplaying within said context). You would also be able to avoid the segmentation in gameplay that many games suffer from (a 'story' segment, a 'dungeon' segment, a 'boss' segment). Instead, everything could flow well together.

I realize how this could quickly exponentiate into a crapload of work, but in keeping the project small-scale, I think it could be viable.

I also realize I'm posting this at the Codex, and controversially implementing something that doesn't necessarily put roleplaying as topmost priority. Thoughts?

Isn't what you are creating effectively an adventure game? Or will the player choices only have a skill modification effect?

This project was trying to accomplish a interactive fiction / RPG hybrid.

I'm not aiming at specifically making an RPG or an adventure game. I just want to place emphasis on choices. Adventure games aren't stat-based, but aren't they for the most part completely linear? You're free to explore, but story progression is pretty set in stone.

My first impression is that it'd be similar to graphical interactive fiction. I'd certainly like to try it, but I'm not sure if it's even possible to create it so that it wouldn't feel too restrictive, with stuff like combat, stealth etc. Did you try to make a small module with using your idea, just to test how much work it takes?

From what I gather, the gist of it is that the whole game would be like the dialogue trees in FO2. Instead of just

A: Make a lame excuse
B: Make a good excuse (Excuse skill)
C: Actual reason (You did miniquest X)

only applying to dialog, it'd apply to combat and stealth too. You could only try to backstab the guard in the hall if your skill was high enough, and only succeed if it was higher still, and in any case there'd be no direct interaction, just choosing options. If it was like that (custom character stats determining your options/their validity) I'd probably like it. You could implement a myriad of skills and the game would basically have no limits to what the player could do except for the writer's imagination.

The main hanging point for me would be whether it could be made complicated enough to be interesting. If you either don't have alterable stats or too few of them, it'd be kind of boring to me. With no stats it'd feel too much like a choose your own adventure book, where you're basically just being baited by the writer to guess what he's thinking over and over again. You can't really play your character as you'd want because diving behind the cover from archers is only going to work if the writer lets it, so you didn't really have the option in the first place, just the illusion of it. And if the stats are too simple (Exaggeration: Attack stat + Talk stat are the only two.) then you'd just dump your points into one and always pick that option, making it a choose your own adventure with only the option of what your character is at the begining. With enough skills and the opportunities to use them, it could be very cool though. And the writing work could be segmented without much trouble to compensate for the extra work, having one guy do all the trap disarming dialogues, another do the combat, another the stealth, etc. If it was implemented like that, it's really just a question of whether you can do enough writing to make it worthwhile.

Well, I thought the options wouldn't be visible but you'd have to invent and type them.

Elwro said:

Well, I thought the options wouldn't be visible but you'd have to invent and type them.

Click to expand...

that would be more like work than entertainment. I don't miss the days of fishing for the correct action phrasing in text adventures

I've recently played a piece of IE called "LASH" and it certainly was entertainment. I can't remember any moment in which I'd curse the interpreter, unlike e.g. in Sierra's Quest games.

Coincidentially me and a friend decided today to program a game similar to your idea. He's got some basic experience in C++ and I've began to learn it, too, and we are planning to create a simple text RPG. It should have roleplaying with choices and consequences, but rather than stat based stuff it will be like you described, because implementing stats in a good way would be too complicated for us yet, although we might try it.

We haven't got a real concept yet though, only the basic idea of making a text RPG.

He's got some basic experience in C++ and I've began to learn it, too

Click to expand...

There are much better languages for writing a text adventure in than C++. Using c++ for that is a bit like taking a nuclear missile to a knife fight. Try a nice modern dynamic language like ruby or python.

Llyranor said:

I'm not aiming at specifically making an RPG or an adventure game. I just want to place emphasis on choices. Adventure games aren't stat-based, but aren't they for the most part completely linear? You're free to explore, but story progression is pretty set in stone.

Click to expand...

Sounds something like the work being done on "Interactive Storytelling" (as distinct from what most people mean by "Interactive Fiction"). Chris Crawford's book on the subject is an interesting exploration of the problems & possible solutions. The site http://www.storytron.com/ may be of interest too.

I agree. I used C and C++ for a long time, up until about two years ago. If you like bugs, handling your own memory management (odds are, you won't), tons of weirdness and lots of special conditions, then by all means, use it.

Its also complete overkill, as Oarfish said. Performance certainly isn't a concern in this case. If you use something like Python or Ruby, not only will you complete your project in a fraction of the time, but there'll undoubtedly be fewer bugs and headaches. Hell, use C# if you insist on using some C derivative. Too much of C++ feels like its jury-rigged, as its been pieced together over the last couple decades. C# took everything from C++, and rebuilt it from the ground up. I feel its a much better language, and the IDE (there is a version you can get free) is one of the best I've seen, if not the best.

On-topic - It sounds like a very elaborate Choose Your Own Adventure (that's not necessarily a bad thing). I suggest starting on a very, very small scale, and see how it works, whether or not you like the idea, and the amount of effort its going to entail.

Flux_Capacitor said:

It sounds like a very elaborate Choose Your Own Adventure.

Click to expand...

Interactive storytelling != branching.

Think of it more as a game system, in the same sense as say, an economic model built into a game. But creating something that generates a coherent, satisfying narrative on the fly is a lot harder than producing a profit & loss spreadsheet. ;-)

The intent is to carefully examine every part of the game and to implement options where deemed appropriate. Instead of making one logical 'good' choice, with the other ones being dead ends, I plan on crafting out a branching network of viable choices (some more, some less - depending on the actual situation). I don't want it to spin out of control by going overboard, but I don't intend to railroad the player into 'right choices', either. If there's no internal logic as to why the player shouldn't be able to duck behind the crates to avoid the archers, then there's no reason why the choice shouldn't be viable. 'Illusions' for their own sake so as to pretend to offer broader choices isn't something I'm aiming for. I'm not trying to be Bioware, and I have different priorities.

Like I said, I plan on focusing on the player's choices within the game, rather than those without (as in a character creation screen). There wouldn't be dumping any skill into a particular field and then only choosing from that field. This is why I'm going with the assumption (again, character development is a whole other discussion) that the player character is a blank state who's average in all regards, thus driving choices based on their appropriateness in any given situation rather on how high their stats in a particular field are.

In essence, I'm more or less trying for a 'gameplay = story' approach, rather than them having them be separate entities. On one end of the spectrum, you get your games where story takes a backseat, which is perfectly fine if the gameplay delivers. On the other side, you get those linear story games where segments of gameplay are (often awkwardly) spread throughout the game - in which case you could usually get a better story just by reading a bloody book. I'm hoping to utilize the gaming medium in order to reinforce the importance of interactivity in storytelling, all while not neglecting storytelling for the pure sake of the gameplay. I do believe 'gameplay = story' is achievable.It's ambitious, I understand, and I'm not a mindblowingly awesome writer (though I am mindblowingly awesome), but hey, it's an experiment, and it'll certainly be a fun project.

I realize it's a lot of work. The project is meant to be pretty small-scale. I do intend to make a 'demo' with a pretty enclosed environment, to test out how well a system like this can flow.

And thanks for the link, Klinn. I'll read through that.

The problem is that dialogs list options don't work well with gameplay. That is, if you have a list of options to choose then it's always going to be easy to choose the right one unless the player is very stupid. That's why dialog trees are overrated in general. There is no *gameplay* to them really. To put any in and make only options that make sense is very difficult and time consuming.

I'm actually developing an engine and builder for the type of game you describe, Llyranor. I've also posted about this last month.
So far I've finished most of the front-end... but if anyone wants certain features implemented, I'd be very glad to hear from those who actually plan to build a "narrative game".

As forthe seamless integration Llyranor mentioned (which I call "unified narrative interface"), I don't think it's necessarily a big advantage. Sure, it will make the games accessible to non-gamers as well, but I think it will seem too repetitive for those who don't enjoy writing as much as us.

I have thought about that kind of game and that's the type of game I would love, even more than a traditional roleplaying game. You don't have any artificial limits and situations like combat would be more dynamic and interesting. Eg 'throw a chair', 'jump out of the window' or whatever which are very common in fiction (and real life) but you never see in an RPG because they don't have the specific rules for that. Usually it leads to rigid, repetitive and boring gameplay.

But Bryce makes a good point and there is something I disagree with you about. What you should do is not have pure choice->consequence mechanics but have that variety of options and base the result on both stats/skills and a random element.

Game: guard is rushing at you with a sword drawn. You are unarmed. He is ten meters away and closing fast.

Option 1: shout and point behind him [bluff skill, 30% chance for average skilled character]
success -> get option of free strike at opponent or chance to run away next 'turn'
failure -> he doesn't fall for it and swings the sword - less chance of escaping now
Option 2: kick a chair in his way [random roll, 70% chance success]
success -> the guard trips and stumbles over the chair. Next turn can try to grab his knife and stab him, run away (90% chance success), kick him in the face or whatever.
failure -> you time it badly and he dodges, barely slowed. He gets an attack but still with a penalty compared to Option 1 failure.
Option 3: charge and attack (resolve consequences with usual combat mechanics - roll to hit/damage etc) [combat skill - for average characters say 20% chance of hitting and wounding gaining the advantage for next round, 30% chance he parries and gains the advantage. For a more category more skilled character it could be 30% and 25%]
success -> gain advantage next round, easier to hit/wound again or do something else.
not success (ie you miss but not too badly) -> you're evenly balanced. Get combat options without any bonuses or penalties.
failure -> you are at a disadvantage and he swings again (parrying will be at a penalty), much harder to run away or pull any of the tricks you could have earlier

Obviously it means more work but I think some elements of that are necessary.

Call it "Text 'Choose Your Own' Adventure w/ RPG elements" if you want. It would still beat the shit out of what passes for roleplaying these days.

We are actually using it in AoD in certain fights, so I think it's a good idea. Then again, I always liked text adventures. I can post some stuff if you guys are interested.

Fell free to give your own examples.

I'm not throwing away the concept of character stats; it's just something I'm putting in the backburner for now. The game doesn't start you off as an incompetent Lvl-1er. You have at least 'average' skills all-around for the most part. I'm not too keen on an actual 'character creation' screen per say, though. I was either aiming for a more minimalistic approach, or simply use it more subtly behind the scenes.

For example, you'd start at as your average 'blank state'. As you'd make your choices within the game, it'd take that into account and develop your character according. For example, if the player constantly engages into direct combat, that's because the character is a good warrior. The difference with that and the character simply *becoming* a good warrior as s/he levels up is that the character doesn't exponentially grow in sheer power. The character simply becomes what the player had in mind given that s/he choose such choices for the character, eg. the character is just what the player expects the character to be based on those choices.

I think this can work, and I'm aiming it this way because I don't want to use a conventional RPG system where the player grows fairly quickly in power. I want to effectively keep it 'low-level' (or 'average-level', anyway) - in which the player can still get killed if s/he becomes reckless while fighting a single thug.

Hory said:

"unified narrative interface"

Click to expand...

Haha, I was also thinking of it as the 'narrative interface'.

Yeah OK, go ahead if it doesn't ruin any surprises. Nice to hear that anyway.

Llyranor: once you've finished I'd be interested to see how you implemented it in NWN. I've had some RPG scenario ideas kicking around but don't have the skills to implement them. Don't really know the NWN1 Toolset yet.

What's the setting?

It sounds like you're working toward something similar to the highlight of Space Rangers 2 - the wacky text adventure moments. It's going to be a lot of work, but I think it ought to be worthwhile. If nothing else, there's a lot more texture and flavour to something like:

Reeling from your previous blow, the thug drops his guard briefly, but it's enough for you to plant your mace squarely on his cheekbone. From the sickening thud of the impact, and the way his body crumpled, you doubt he's getting up again.

Click to expand...

...as opposed to...

[PC] hits thug in the head for 16 damage. Thug is killed.

Click to expand...

Obviously, one is more labour intensive than the other, but if you think of it in terms of how many "animations" a writer can turn out as compared to a 3D animator, it's dirt cheap.

Well, this is a hobby, not a job. And I'll easily take #1 over #2 any day, so it's win-win for me, even if it is more labor-intensive.

As for the setting, I don't think I'll stray away much from NWN2's standard setting. I'm not going to be implementing any custom content for its own sake. I don't think it'll end up being a serious mod, but I certainly plan to implement serious design into it.

I suspect the whole premise of the mod will revolve around killing Deekin.

Llyranor said:

I suspect the whole premise of the mod will revolve around killing Volourn.

Click to expand...

Fixed. It will make your mod way more popular.

I'm not sure people would really want to have Volourn killed. Deep down, you all love him.

Llyranor said:

I'm not a 'hardcore' roleplayer.

Click to expand...

But you like it when this roleplayer goes hardcore on you