[Steve Ince]

Originally Posted by Fairygdmther

How nice to have one's opinion backed by competent authority!

Has someone else entered the room? (but thanks)

Originally Posted by Martin Gantefoehr

I agree that, in principle, making ALL events playable seems like the "natural" way to approach things in an interactive game. However, on the development side of things, providing dozens of gameplay options for a complex, fast-paced story event that's supposed to have 30 seconds of screen time, is something that might be feasible over at Valve, but not in many others development houses.

I do understand that the cut-scene may be seen as the best way to get over key dramatic story events, particularly when you have limited resources at your disposal, but shouldn't we be striving to deliver as much in an interactive manner as possible? Perhaps we should be looking at the overall game and story and modifying the gameplay and interface mechanics so that we maximise all sides of the interactive experience. If that means adjusting the way we design and deliver our interactive story to be as seamless with other gameplay as possible, then surely that's the ideal solution?

[Martin Gantefoehr]

If that means adjusting the way we design and deliver our interactive story to be as seamless with other gameplay as possible, then surely that's the ideal solution?

Hey Steve.

Yes, of course. I guess that's the kind of design we're all trying to work towards at the moment.

As I said, I'm not advocating overly cutscene-y games. My only point is, when one plans to convey something as important as a central (or even THE central) plot point to the player through gameplay entirely, then it puts the game at high risk. It's very resource-intensive, very hard to design for, and any flaw might ruin the dramaturgical value of the scene altogether.

So, if one can't provide that kind of gameplay with given resources, then in my view, a good gameplay build-up to a to-the-point cutscene is much preferable to some pseudo-interactivity that amounts to random clicking or buttom-mashing, as I've seen it in a variety of games lately.

[Jackal]

Surely there's no "ideal" when it comes to a creative endeavour? Aren't there just different approaches? Regardless of what motivates (or requires) it on the development end, using cutscenes as a "reward" is still a viable use for them in games. Making better dinners doesn't mean we don't still want dessert. Using them as a change of pace for the player is also reasonable to me - a non-interactive break, of sorts (admittedly, this is less of an issue with adventures, which are pretty passive already).

I certainly agree developers should be looking for opportunities to create gameplay, rather than ways to escape them, and there's probably been an over-reliance on cutscenes in many cases. But going to the opposite extreme doesn't seem like an improvement to me. Frankly, Half-Life 2 wasn't a better game for having no cutscenes, though it's trendy to say so.

[Steve Ince]

Originally Posted by Martin Gantefoehr

Hey Steve.

Hi Martin. The book's done and with the publishers. Nutso is a few weeks away from completion.

So, if one can't provide that kind of gameplay with given resources, then in my view, a good gameplay build-up to a to-the-point cutscene is much preferable to some pseudo-interactivity that amounts to random clicking or buttom-mashing, as I've seen it in a variety of games lately.

Most definitely! I realise that I'm perhaps asking too much of developers in general and it's certainly not something that we can just switch on. My own plans in this respect will be some way off seeing the light of day, but it's going to be something I work towards over the coming years.

Originally Posted by Jackal

Surely there's no "ideal" when it comes to a creative endeavour? Aren't there just different approaches? Regardless of what motivates (or requires) it on the development end, using cutscenes as a "reward" is still a viable use for them in games. Making better dinners doesn't mean we don't still want dessert. Using them as a change of pace for the player is also reasonable to me - a non-interactive break, of sorts (admittedly, this is less of an issue with adventures, which are pretty passive already).

Any ideal I talk about is not one that's restrictive in terms of how gameplay is defined or created but more about the philosophy behind how you approach the creation of an interactive whole. It's concerned with how a developer thinks about the way they can get the most out of their gameplay ideas.

I certainly agree developers should be looking for opportunities to create gameplay, rather than ways to escape them, and there's probably been an over-reliance on cutscenes in many cases. But going to the opposite extreme doesn't seem like an improvement to me. Frankly, Half-Life 2 wasn't a better game for having no cutscenes, though it's trendy to say so.

But if you look from a different perspective, it was a better game because it was filled with good gameplay.

Actually, HL2 did have cut scenes, after a fashion. There were many instances where NPCs would launch into conversations or set-piece attacks would happen. It's just that the player could choose to stand and watch them or move around and do something else. The weakness with this approach is that the player is still not interacting with these scenes and because you can walk your character away, it's possible to miss important story-related exposition.

[insane_cobra]

Now, I must admit I've read absolutely zilch of this thread (seemed like a been-there discussion that keeps popping up every now and then), but there's a great lecture on interactive narrative by Ernest Adams from last year's GDC up on Gamasutra.

He doesn't come up with any definitive answers, of course, but it's a very interesting listen nevertheless. I can definitely echo his final sentiment that game narratives eventually need to become "thinking" and adapting entities.

One of the first steps could be breaking down set pieces to events, actors and stages. The same event could take place on different stages, with different actors and intelligently modified accordingly. Then, for instance, if you'd miss something you're not supposed to miss, the game would find you and force the event onto you, but in a way that would feel natural and pre-scripted. I believe something like that's already being utilized by certain games, but not to the extent I'd like to see.

Anyway, enough blabbering, go ahead and listen to it.

[Steve Ince]

Originally Posted by insane_cobra

I can definitely echo his final sentiment that game narratives eventually need to become "thinking" and adapting entities.

Or do they?

While this sort of narrative may well suit certain types of games (MMORPGs for instance) the actual way the narrative works will be dependent on what the developers want it to do. In a comedy game, for example, some of the humour may well be derived from the developing situations that arise from the way the plot unfolds. If the narrative is a "thinking" one, at what point does the game stop being funny because the situations are no longer being developed.

Most adventures have a narrative that's revealed by the player working through the gameplay. If the narrative is "thinking", how will this affect the gameplay - will we find that this "thinking" has backed the game into a dead end?

[insane_cobra]

Originally Posted by Steve Ince

Or do they?

Okay, maybe I worded that badly, I didn't mean to say that all future games must be like that. Diversity above all, of course.

While this sort of narrative may well suit certain types of games (MMORPGs for instance) the actual way the narrative works will be dependent on what the developers want it to do. In a comedy game, for example, some of the humour may well be derived from the developing situations that arise from the way the plot unfolds. If the narrative is a "thinking" one, at what point does the game stop being funny because the situations are no longer being developed.

When I said that narrative should be a "thinking" entity, I didn't mean that AI routines should govern how the story unfolds and make everything up on the go (though even that may be possible in distant future, but what use for us people then?). Instead, writers/designers design the story (linear, branching, whatever...) just like they do now, but they don't detail everything unless absolutely necessary. AI then makes sure that player passes through all important plot points while also allowing for greater freedom then possible in most current games.

For instance, let's say your character gets accused of committing a crime and story demands that he ends up in prison at some point. Unless you voluntarily surrender, cops come to arrest you. That can happen almost anywhere in the game world, in your apartment, at work, while shopping... But game acknowledges all those different stages and slightly modifies the scene to make it look like it was always supposed to happen right there and nowhere else. You're not forced to be at specific location for plot to advance, it'll find you.

Or say you can choose which characters to befriend from a pool of optional NPCs. Key scenes then play out slightly differently depending on your social relations. Also, if plot demands that you meet with one of them and you refuse to go to their place, they'll come looking for you.

Your point is well taken, though, some situations need to be meticulously scripted if they're to make the intended impact. So that approach wouldn't be applicable to all cases - or, as previously noted, to all games - but often it would.

Most adventures have a narrative that's revealed by the player working through the gameplay. If the narrative is "thinking", how will this affect the gameplay - will we find that this "thinking" has backed the game into a dead end?

As I just explained, ensuring that it doesn't happen would still be designers' job.

[Steve Ince]

Originally Posted by insane_cobra

When I said that narrative should be a "thinking" entity, I didn't mean that AI routines should govern how the story unfolds and make everything up on the go (though even that may be possible in distant future, but what use for us people then?). Instead, writers/designers design the story (linear, branching, whatever...) just like they do now, but they don't detail everything unless absolutely necessary. AI then makes sure that player passes through all important plot points while also allowing for greater freedom then possible in most current games.

Clever AI would be a little wasted on that. I can make sure a player passes through all the important plot points in a game by the use of conditions - setting variables and testing for them. The game could have a number of branches that could be played in any order with resulting consequences. Not only would I not have to spend a fortune developing AI routines, I would be in control of the scripts, how the characters react to the player, etc. That's not to say that the writer/designer couldn't still do that with AI controlling things, but why bother?

For instance, let's say your character gets accused of committing a crime and story demands that he ends up in prison at some point. Unless you voluntarily surrender, cops come to arrest you. That can happen almost anywhere in the game world, in your apartment, at work, while shopping... But game acknowledges all those different stages and slightly modifies the scene to make it look like it was always supposed to happen right there and nowhere else. You're not forced to be at specific location for plot to advance, it'll find you.

You don't need a "thinking" narrative for this, just a set of conditions to be fulfilled. It could be that when a timer reaches 10,000 the player character is arrested and could be anywhere when the timer goes off. The downside of this is that you either have to set up the arrest scene for all possible locations it takes place or make the locations so generic that a fixed arrest scene can be played out anywhere.

What you've got to ask is what you'll be getting from such complexity and the need for additional resources, much of which won't be seen by many players. Costs are already spiralling through the roof on many big games and it's hard to justify content that only "might" be seen, particularly if that content is expensive.

[insane_cobra]

Originally Posted by Steve Ince

Clever AI would be a little wasted on that. I can make sure a player passes through all the important plot points in a game by the use of conditions - setting variables and testing for them. The game could have a number of branches that could be played in any order with resulting consequences. Not only would I not have to spend a fortune developing AI routines, I would be in control of the scripts, how the characters react to the player, etc. That's not to say that the writer/designer couldn't still do that with AI controlling things, but why bother?

Because it would require too much effort to script every possible combination. And given enough variables, you couldn't plan everything out anyway. But maybe AI is again bad wording on my part, it wouldn't have to learn or anything like that.

You don't need a "thinking" narrative for this, just a set of conditions to be fulfilled. It could be that when a timer reaches 10,000 the player character is arrested and could be anywhere when the timer goes off. The downside of this is that you either have to set up the arrest scene for all possible locations it takes place or make the locations so generic that a fixed arrest scene can be played out anywhere.

It would definitely require some clever coding, but with procedural programming techniques becoming more popular (Spore, next gen LucasArts technologies, SpeedTree...), I think it's already within reach.

What you've got to ask is what you'll be getting from such complexity and the need for additional resources, much of which won't be seen by many players.

We obviously don't agree on the importance of having a unique experience. Many people value it more than simple calculations would lead you to believe so I'm positive those resources would not go wasted.

Costs are already spiralling through the roof on many big games and it's hard to justify content that only "might" be seen, particularly if that content is expensive.

That's why I'd aim it at non-gamers who are not yet spoiled by amazing graphics and similar distractions. Take a nice engine from several generations ago, come up with some striking stylized art (Parappa the Rapper wouldn't look any nicer on a PS3), and invest the money you save that way on what you believe is truly important (non-visual tech and extra content).

I know I probably sound like a hopeless dreamer to someone like you, and thousands must have gotten these same ideas many times before, bur I think the climate is changing and soon there might be room for something like this. Maybe I'm wrong, though.

[Steve Ince]

Don't get me wrong, what you're saying is an excellent way of developing content and narrative in many ways. What I'm trying to question is whether the approach is the right one. I'm just very wary of anything that distances the end product from the creative individuals. One of the reasons that so many games look the same has nothing to do with the talent of the artists but that they all use the same tools and similar rendering pipelines. If we start applying this to narrative we'll end up with games that have forumulaic stories.

There are many screenwriting books that tell you that your screenplay must be 120 pages long and the inciting incident should happen on pages 6-10 and that the whole thing must be split into three acts. The good screenwriters worry less about this - if they followed the forumula we wouldn't get films like Seven, Witness, Talk Show, Fight Club, Memento, or Raiders of the Lost Ark (which actually has seven acts).

[insane_cobra]

I couldn't agree more, but in what I'm suggesting the story would still be structured and written by a writer. The "narrative engine" just makes sure the story is told while taking into account player's actions (player's actions don't change the story in any significant way - unless it's written as a non-linear story in the first place - they only personalize it).

As for possible cross-game similarities, every game could use its own unique narrative engine just like every game can use its own unique graphics engine or its own parser etc.

[MoriartyL]

Sounds like Craig Perko's ideas for storytelling. I'm with Steve here- it sounds like a waste of time.

[Jeysie]

Actually, I like Insane Cobra's idea myself (though I admit I don't know if it's feasible technology-wise or not).

I play tabletop role-playing games regularly, and it occurs to me that a session being run by a good Game Master has the sort of narrative that would fit the ideal of "interative fiction".

I mean, Cobra pretty much described the computer version of a GM running a campaign. You take a setting - a medieval kingdom, for example, with buildings, NPCs, etc.. You set up an overarching plotline... we'll say that the players are hired by the king to do the old rescue a kidnapped princess from a wizard in a tower near the kingdom shtick.

And... that's it. How the players proceed is up to them. Maybe they do the whole "bash in the door, kill things, grab the girl" tactic. Maybe they spend time in the town trying to find out any weaknesses the wizard might have that they could exploit. Maybe they plot a way to storm the tower with stealth tactics. Maybe they *join* the wizard themselves. Etc. Etc.

The puzzles spring from having to get around the environment, and you can solve them whatever way makes sense. The princess' cell door is locked, so you can break the lock, pick the lock, steal the key from the guard, etc.

Also, a GM that doesn't want to waste a lot of time will only spend time detailing out things that will matter to the players. And a GM who isn't out to "punish" the players will make sure that if there's some sort of info/items the players need, they can find creative ways to slot it in if the players take one path instead of another. Like I said, Insane Cobra pretty much described it already. The players themselves are creating (or at least personalizing) the story.

Obviously this sort of thing is not completely technically feasible... for a computer game you have to pre-script many things (like PC/NPC dialogue, at least), and even a flesh-and-blood GM would have trouble pulling off this sort of concept in an ideal sense.

But to me, if you're going to make any pretense at adventure games being about exploring the environment and "living the story", then to me it seems like tabletop RPGs are a good place to do some research - especially the "diceless" systems (since adventures don't typically have much, if any, randomness or combat).

I've often thought that if someone could find a way to distill the tabletop RP narrative experience into an adventure game, I'd be in gaming nirvana.

Peace & Luv, Liz

[Kurufinwe]

Originally Posted by Jeysie

I've often thought that if someone could find a way to distill the tabletop RP narrative experience into an adventure game, I'd be in gaming nirvana.

Daughter of Serpents tried to do that, long ago. But, ultimately, it would have been too vast, and complex for the time, and so the final product didn't really fulfil its promise -- but that's indeed something that may warrant further examination now.

[insane_cobra]

Originally Posted by Jeysie

Obviously this sort of thing is not completely technically feasible... for a computer game you have to pre-script many things (like PC/NPC dialogue, at least), and even a flesh-and-blood GM would have trouble pulling off this sort of concept in an ideal sense.

Exactly. That's why I had a much more down to earth implementation in mind, but I'd say you got the basic idea.

This Daughter of Serpents game looks interesting, thank god for abandonware sites.

[Intrepid Homoludens]

Originally Posted by Jeysie

You take a setting - a medieval kingdom, for example, with buildings, NPCs, etc.. You set up an overarching plotline... we'll say that the players are hired by the king to do the old rescue a kidnapped princess from a wizard in a tower near the kingdom shtick.

And... that's it. How the players proceed is up to them. Maybe they do the whole "bash in the door, kill things, grab the girl" tactic. Maybe they spend time in the town trying to find out any weaknesses the wizard might have that they could exploit. Maybe they plot a way to storm the tower with stealth tactics. Maybe they *join* the wizard themselves. Etc. Etc.

That sounds a lot like the upcoming Assassin's Creed (preview ), except that in this game you're hired to kill the princess.

Assassin's Creed plays out over a series of different cities in the Middle Ages, such as Jerusalem and Damascus. Altair must infiltrate each city, locate and eliminate his target, and quickly escape from pursuing forces. One of the things that makes Assassin's Creed unique is the flexibility the game gives players in accomplishing their goals. For starters, there are no pre-planned paths that players have to take through each level.

[Kurufinwe]

Originally Posted by insane_cobra

This Daughter of Serpents game looks interesting, thank god for abandonware sites.

Don't get your hopes up. Basically you start the game by creating your character: you choose his/her gender, nationality, age, personality, then the job (archaeologist, traveller, PI, sleuth, occultist, mystic), and finally the abilities (papyrology, Arabic, toxicology, demonology, clairvoyance, etc.).

And then, supposedly, the game will depend on both your character and the choices you make. Except that it really doesn't. There are 3 main versions of the story depending on your job, but that's it. Your abilities influence the course of the game, but not much. For instance, at one point, you can meet a demon. If you have 'demonology' and 'the powder of kljkldsjk' (did I mention it was a Lovecraftian game? ) among your abilities, then you'll know what to do, and use the powder to control the demon. Otherwise, the woman who's with you and happens to know about demons will tell you you need the powder, and tell you to go and look for it, and what to do with it once you've found some of it in the room. Not a tremendous difference... (and that's one of the cases where abilities matter most; most of the time, it's just a couple of different lines during a conversation).

On top of that, people keep telling you what to do, and there are no puzzles to speak of, so you can finish it in a couple of hours on your first run (so make that 6 hours for the three paths, though at least half of each path is the same as in the other ones).

So really it's not very good. Still, it's not bad either. The story isn't bad, and there's nothing painful to play in it. When we played it with my parents at the time of the release (1992), we didn't really feel robbed.

Oh, and before anyone asks: no, it's not the same as The Scroll, and yes, the Internet is great for spreading misinformation.