[Scourge of Ages]

Reading this discussion on emergent storytelling as opposed to just procedural stuff, I am reminded of the Mount & Blade games.

They provide a very solid framework of gameplay and rules, but also have established characters, places, and relationships all in play by the time you start. From there, you're free to take your character on any number of routes, finding interesting situations and making your own stories. All it takes is a little bit of imagination to impose your narrative over the events unfolding (much like we're all doing with the various LPs over the years).

[Sarkoth]

Reading this discussion on emergent storytelling as opposed to just procedural stuff, I am reminded of the Mount & Blade games.

They provide a very solid framework of gameplay and rules, but also have established characters, places, and relationships all in play by the time you start. From there, you're free to take your character on any number of routes, finding interesting situations and making your own stories. All it takes is a little bit of imagination to impose your narrative over the events unfolding (much like we're all doing with the various LPs over the years).

Someone said Mount&Blade ... oh the joy!!! So I'm not the only one loving that series around here.
Emergent storytelling indeed is one big thing for the future. But I believe it will stay in a niche. Growing, yes, but never quite overcoming scripted storytelling. For emergent storytelling, you need imagination. That is not always as easy as it was. If you don't believe me, try playing Wasteland1 again.

[Luis Dias]

I think your insights are all right of course. There's still a bit of a blur in my head to what is really changing here. I do think it's more than just creating tools that empower a developer to do his own game with a lot less resources at hand, although that's probably the majority of it. The idea that AAA games have to spend hundreds of millions of dollars into production with hundreds of top designers, artists and coders and then have to sell at least 5 to 10 million copies to get profitable is hurting the landscape of games massively. Top suits begin to take control of the game direction and no wonder most of the AAA titles have been borefests from planet Mediocria, dumbed down and homogeneously idiocratic. Hey, just like Hollywood!

I long for the creative destruction that will wash away this mess. And the only way to do it is precisely to create intelligent and thoughtful procedural (or not) generating code that will outsource the biggest workloads in the design and creative process. Again, it's a pipe dream, just like having a real strong AI, for instance, but I think that if a pathway of babysteps with great benefits all the way up is found and cared for, then it will become a reality.

This is why I compare this with Doom. 3D already existed in previous games, but this was the ice on the cake that showed how it could work and how it could scale to the fidelity we have nowadays to the point that you didn't need to be a brilliant genius to "get it". They all got it and started doing little things that complemented each other and improved the 3D experience one baby step at a time. And now we have Frostbyte, UE4 or Crytek, etc.

The only way it will work this way however is if a game like NMS or some ulterior title really blows away the mind of everyone who plays it long enough to make the point. And the fact that this game looks like fun incarnated is extremely important in that direction.

[The E]

Well, I'm with Mongoose on this one. Yes, this whole new area of game development is nice and exciting, but I don't believe it will wipe away everything else, or even that it should. Surely the gaming market is large enough to accommodate both big lowest-common-denominator games and small indie efforts?

I mean, I likes me some big budget AAA stuff. Games like Uncharted, or Tomb Raider, or Mass Effect, or The Last Of Us, or Saint's Row. I would be sad if that mode of storytelling would just vanish into thin air.

[Luis Dias]

Yes well I love Mass Effect. Love it. But the thing is, have you looked at their game design process? Just the VA is mind boggling.

Now they do want to scale that up. For instance they proudly state that the entire Citadel DLC from ME3 has more options, choices, dialogues, etc than the entirety of one of their previous games. Now consider the madness of that, and think about why was the plot of ME so tightly knitted, so hard to disentangle. Why ME2 suffered from all that "Now you are in a terrorist org and that's just great!" and had to continuously endure what was a railtrack experience with some tonal differences, when we were promised a world of "deep choices and deep consequences".

Do you honestly believe they can scale this up? Already in ME2 they had to "pimp it up" with awkward details (like boobs, "slut walks" and dumbed down gags and plots) to make people more "willing" to buy it (wink wink). Perhaps this is always a tendency in any capitalist endeavour, you always tend to sell out in minor ways then bigger ways (the whole ME3 is to me a big sellout towards the "safe zone" of "post-apocalyptic" landscapes), but I also think it's a side effect of the bloatization of the game design process.

Because I wonder. Mass Effect introduced really top notch tech regarding faces and eyes, etc.... but can it live up to LA Noire, for example? No, of course not. So should the next titles live up to those things? Well, regardless of what I or you might feel, I think they will try it, bloating even more the process (now it's not only about VA, or "arms and legs" motion tracking, it's also about tracking your ****ing micro-expressions and what not). If they don't, they'll be left in the dust in some way (compare the trailers of the new Witcher to the new Dragon Age... already DA looks "dated" ahhh shmucks. that can't bode well!). I think this trend is a suicidal one.

So I really do think they'll need a massive creative destructive blow from a different angle.

[The E]

Yes, but how does that translate into procedurally generated, mechanics-driven games taking over the entire industry?

There's always going to be a market for the big loud modern military shooter, or its descendants. There's always going to be people like me who will want to have a superbly crafted, entirely planned out experience a la TLoU or the other games I mentioned.
Now we have this new wave of game design, which concentrates on getting that big cool experience without having to invest tons of money into designing every single piece of the puzzle. There's certainly a place and an audience for those too, hell, I count myself in it, but does that automatically mean that all other forms of gaming should die? Or that the industry has to contract so much that it loses the ability to deliver the Mass Effects or Uncharteds or Gears of Wars?

Most of these projects, like No Man's Sky or Limit Theory or Enemy Starfighter are passion projects, made not to appeal to the broadest possible audience, but built to be the ideal experience for a small targeted group. Which is totally cool, it's not like we're doing anything different in this place, after all. But if all of gaming were to be like that, I fear that we would be losing a lot of common ground in the greater gaming community. Fewer shared experiences, as it were.

For me, it comes down to this: I don't care if a game was made by a lone man in a cave, programming stuff by tying bits of art and code to carrier pidgeons, or whether it was made by a big multinational corporation that spent the budget of a mid-sized city on it. All I care about is "Am I having fun playing this", and so far, I have to say that I can't honestly tell you whether I had more fun playing Hotline Miami, or something like GTA San Andreas.

[Luis Dias]

Yes, but how does that translate into procedurally generated, mechanics-driven games taking over the entire industry?

Oh no, I think it doesn't quite do it like that. It should be something different than either Stanley or NMS. It would be something that could be manageable, directable, but also surprising and un-railroading.

There's always going to be a market for the big loud modern military shooter, or its descendants. There's always going to be people like me who will want to have a superbly crafted, entirely planned out experience a la TLoU or the other games I mentioned.

Yes, well I caveated this particular speculation with the admission that games are and should be heterogeneous.

Now we have this new wave of game design, which concentrates on getting that big cool experience without having to invest tons of money into designing every single piece of the puzzle. There's certainly a place and an audience for those too, hell, I count myself in it, but does that automatically mean that all other forms of gaming should die? Or that the industry has to contract so much that it loses the ability to deliver the Mass Effects or Uncharteds or Gears of Wars?

Most of these projects, like No Man's Sky or Limit Theory or Enemy Starfighter are passion projects, made not to appeal to the broadest possible audience, but built to be the ideal experience for a small targeted group. Which is totally cool, it's not like we're doing anything different in this place, after all. But if all of gaming were to be like that, I fear that we would be losing a lot of common ground in the greater gaming community. Fewer shared experiences, as it were.

This is a whole separate discussion, one where I don't really disagree with at all. When I refer to a certain shift of paradigm I refer to certain waves of change that do affect several entertainment industries, whether we like it or not. For instance, DOOM forever changed the landscape of games; HBO changed the landscape of TV series; CGI changed Hollywood, etc., etc. And these waves of change always start from the ground up. The big projects always aim for something safely profitable, and wouldn't touch this "procedural generating" stuff ever. GTA was done by a really small crazy dev studio, for example, and it did change the gaming landscape too.

I kinda compare this NMS experience vs say Tomb Raider or TLoU with for instance the DOOM experience vs the "point and click" movie games from the early 90s. While Doom was really inferior to the "narrative" and "plot" and the whole "immersive" experience of us being inside a kind of a movie, DOOM offered us total freedom of space with our camera. And while "point and click" still remains with us, guess what kind of game was the more important regarding the gaming industry ever since?

For me, it comes down to this: I don't care if a game was made by a lone man in a cave, programming stuff by tying bits of art and code to carrier pidgeons, or whether it was made by a big multinational corporation that spent the budget of a mid-sized city on it. All I care about is "Am I having fun playing this", and so far, I have to say that I can't honestly tell you whether I had more fun playing Hotline Miami, or something like GTA San Andreas.

Well this is handwaving about fun. Which is fine, I guess. My particular version of "fun" is something that strives for different experiences. "Fun" isn't disconnected from my more intellectual parts of the brain, for instance. Yeah, Hotline Miami seems really fun and "edgy" but it bores me too. I played "that game" with the original GTA and GTA2. Fifteen years ago.

[Luis Dias]

PS: Curiously as a small evidence here's a small list of "upcoming" space sim games made in nov 2013 (yeah it included X Rebirth), and the thing that caugtht my attention was how so many of them mention this procedural stuff in really different ways. The quality and style vary quite a lot, but that's a really good thing.

[Flipside]

I think the thing is, there is a kind of 'Theory of Everything' about gaming out there, whether you are calculating a mathematical equation, or simply pulling a card off a shuffled deck, whether you are driving a Tank towards and enemy base or simply typing 'Go North', whether the challenge is roll the entire world into a moon sized ball within a time limit, or to rescue the Presidents' family within a time limit, there are certain parameters that must be, if not met, at least acknowledged.

However, a simple randomization of those values does not assure a coherent or enjoyable result, there needs to be underlying 'guidelines' to direct those parameters so that they work in tandem with each other and make a theme, there's always going to need to be some kind of programmer-defined control to the algorithm (hence the procedure in procedural).

The closest I've seen to 'mainstream' procedural generation is work by Will Wright or Sid Meier, which quite often using a mixture of pre-generated and semi-randomly generated content such as in Spore. I think the problem with larger distribution companies is that what they think they are is the equivalent to a movie studios producing blockbuster CGI movies that the player just happens to be involved in. There's nothing wrong with that at all, it's fun, but in those situations you can control everything, the music, the lighting, the dialogue etc to present a specific mood and draw the player in, to me it's not so much about the game itself, which is often simplistic and based on a very old gaming model, but rather the experience of playing it.

I do think we are a long long way from a procedural-based system being able to create that level of emotional investment in its own stories, there's a big difference between the two types that needs to be bridged. Players of story-led games expect the story to finish and be resolved at some point, procedural gamers basically want the opposite.

[Luis Dias]

That's why I brought the point and click movie games in the 90s. I remember playing some in my friend's house (I can't recall the titles) and they were amazing. Real actors, real sets, great adventures, really immersive, etc. Doom offered a really dumbed down experience in relation to that. But at the same time it was something so profoundly different and adventurous, that asked you to look at that world with your own eyes and movement and camera, that it completely blew away the other genre. The first was costly and cumbersome in the making, the second was trivial by comparison once you have the tech.

As you are right, the hard thing is to create the right tech, the right "procedures", the right design methodologies, the right level designs that come with that kind of tech. No longer about designing levels or particular puzzles but rules of levels and generators of puzzles. This is obviously the big question in these kinds of games, will the programmers be able to sucessfully create a "generator" of fun? Of diversity, not numbness, of creativity, not bore?

It's like designing not the thing itself, but the meta-thing. It's a different kind of design, but nevertheless should be done with the same spirit.



I dream of the equivalent of Mass Effect that will be truly exploration based, intelligent and interactive.

[Flipside]

Just wanted to add, the other reason large companies are less likely to get involved in this sort of thing is that their bottom line counts on people buying their new releases. If you create a game that is capable of generating content to a high level, then you are lengthening the play-life of that game by a vast amount, which in modern commercial thinking is considered a bad idea.

I'm not certain I agree with that position, since a lot of these companies rose to prominence primarily on one or two titles (Rockstar, Bethesda, even Blizzard - not that all these companies are necessarily prime examples of this, Blizzard DO use procedural generation in Diablo for example, and Bethesda are obviously incredibly fond of their entire series), but that seems to be what the general marketing opinion is for a large swathe of the entertainment industry.

[Luis Dias]

This game is a stunner, absolutely beautiful. Congrats to Hello Games for these shots I only hope the game lives up to them!

[Colonol Dekker]

Is on PS3 as well?

[The E]

Nope. There's no way to make a game with No Man's Sky's scope within the confines of the PS3 (and there's also the small issue of Hello Games being a small indie dev without the ressources to make a port to a different hardware anyway)

[Colonol Dekker]

The PS logo threw me, I understand

[General Battuta]

Just wanted to add, the other reason large companies are less likely to get involved in this sort of thing is that their bottom line counts on people buying their new releases. If you create a game that is capable of generating content to a high level, then you are lengthening the play-life of that game by a vast amount, which in modern commercial thinking is considered a bad idea.

Not really true.

[Mikes]

Procedurally generated content and hand crafted / storytelling based content are entirely different niches that can and will hold their own.

While procedurally generated content can provide enticing "playgrounds" they have about as much a chance generating an "enticing story" on their own as a chatbot has writing a new bestseller. I.e. None.

Saying the one would replace the other sounds as silly to me as saying people will stop reading books because they are playing Lego now.

i.e.: Entirely different experiences.

[Phantom Hoover]

Yeah, procedural content is not all that threatening because players do get bored with it unless it hits a very specific note.

[Dragon]

Procedurally generated content and hand crafted / storytelling based content are entirely different niches that can and will hold their own.

While procedurally generated content can provide enticing "playgrounds" they have about as much a chance generating an "enticing story" on their own as a chatbot has writing a new bestseller. I.e. None.

Saying the one would replace the other sounds as silly to me as saying people will stop reading books because they are playing Lego now.

i.e.: Entirely different experiences.

Note, there's nothing stopping you from writing a hand-crafted, complex story set in a mostly procedurally-generated world, with locations and such partially randomized. Case in a point: TES: Daggerfall. An absolutely titanic world (the size of England), huge, convulated storyline and thousands of procedurally generated sidequests. I haven't seen anything like it since, but it's right at the "golden mean" of procedural/handcrafted dilemma. I think that's the way to go, TBH. High-quality procedural-generation with restrictions put on it so it always generates a world in which the plot can happen. Important locations can be hand-crafted, though one needs to be careful as not to make it too obvious. Generic, randomized NPCs mixed with unique ones are already a common approach, and this can easily extend into the whole game world.

That said, I don't know of any game besides Daggerfall that would fully take advantage of incredible power of this approach. Maybe because actually creating such a game would be a huge challenge, and Daggerfall also exemplifies potential pitfalls of this approach. Namely, it's very hard to test, balance and anticipate every possible interaction of world and story, which can create unexpected situations. Also, to fully take advantage of such a world, the story should not be lineal and allow the player significant freedom. Daggerfall pulled that off (the main quest tree is as wide as it's tall ), but it's hard and somewhat unrewarding for the devs (you know that most people probably won't see half the work you done due to branching).

[An4ximandros]

PS4 Went ahead and took one of the few games I had any interest in years hostage. Damn it, by the time it goes PC, if it does, it will probably be beyond my interest horizon...