Emergent Gameplay

[Thompson]

But that's the way the real world would interact, you know; that's what most emergent experiences are right? I'm mostly with Lork on this one: some elements of the game combine based entirely on the player's actions, intentional or unintentional, to produce an outcome that wasn't hard coded in by the designers. Maybe they foresaw a similar outcome, but it will seem like a unique event. In the case of Far Cry 2, the game is constructed in such a way as to be conducive to such outcomes because of the volume and consistency of interactions. I believe it was Remo that said its brilliance was down to a lot of the player's experience being "so plausible". I'd probably call it plausibly implausible, the kind of stuff that doesn't usually happen but theoretically could.

The most common one is driving a car off a bridge whilst looking at the map in Far Cry 2. It won't happen to everyone, but that's a situation made possible by several aspects and sub-aspects of the design working in concert with player input. Maybe we can break it down into broad but largely accurate points.

A) Design

1) Driving (play mechanic) - cars are pretty fast and the handling is fairly tight.

2) Physics (passive mechanic) - this allows the wood on the bridge to be smashed when driven through.

3) The map (play mechanic) - sufficiently large to obscure the player's view of the road.

Intended input

4) Destination (mid-high level interaction) - player is heading to objective in vehicle.

5) Current location (mid level interaction) - player is navigating the immediate route, which happens to be a bridge.

6) Orientation (low level interaction) - player is viewing map in order to effectively navigate to 4).

C) Emergent interaction

7) Loss of concentration - some of the above factors combine to make the player lose concentration. The "emergence" happens here - in some ways it is "immersion", because the player's unintentional mistake results in something that could plausibly happen were this scenario replicated in real life. It's not the outcome, but this that is emergent because it's the eye of the storm of contributing factors. The player and the game have "fused" to produce something. Not sure how well I'm communicating this, I'm struggling to describe it well. In my rhino example there may be several emergent interactions.

D) Outcome

8) The player ends up flying off the bridge and crashing into the stream or, if very unlucky, river valley below. This has not been "scripted" into the game.

You see where I'm coming from? I'd add an extra qualification that it has to be both uncommon and relatively significant to be "emergent". It needn't be unexpected, because the player could deliberately interact with a game in such a manner as to produce an outcome that was expected in real world logic, but isn't often part of game logic. This has possibly never happened, but is theoretically possible.

All of this is IMHO, of course, but I'm confident that most people would agree.

[DanJW]

I agree. The emergent gameplay in that example comes from the interaction of two separate game systems:

1: interactions between cars and wildlife and

2: mission objectives, completion of, tracking of etc

At no point was a designer expecting you to use wildlife in this way to complete a mission. But it so happens that you can.

The emergent narrative is the linear sequence of minor unplanned-for (by the player) events, that in retrospect make a logical pattern (that was unplanned for by the developer).

Edited June 10, 2010 by DanJW

[morslok]

Maybe my view is a bit simplistic, but I tend to think of "emergent" game elements as being the unintended (to the player, not necessarily the developers) consequences of player interactions.

For instance, intended consequences would be shooting someone to kill them, or stealing a car with some passengers in it and the dumping the car in the water to kill them, whereas unintended consequences would be hitting a rhino with your car and then having that same rhino stop a caravan that you happened to be about to ambush. The act of hitting the rhino was a player interaction with the world (or game systems if your prefer), and that interaction had the unintended consequence of stopping the caravan.

I don't think it is a important whether something was intended to be done by the developers or not, how can we know exactly what the intentions of the developers were anyway? It's like art (if you'll forgive the analogy), you don't know every facet of the artists inspiration and direction in making his art, the only thing that is for you to know is the emotions it evokes in you.

And of course the best examples of emergence are the ones that make great stories

[Hermie]

Morslok makes a good point, "intent" is a bewildering word. The developer can have intended everything in the world, but you will probably never know just what. To bring back what someone cited as the most "pure" emergent game, Nobi Nobi Boy. Keita Takashi has said as much that he intended for players to make up their own fun in the game. Does that make it "unemergent" after all?

I guess "scripted" or maybe "directed" could fit better, but neither cover it completely.

I like gdf's breakdown that "one game mechanic + a separate mechanic and/or player interaction = unexpected result" is kind of what we are discussing here. And I guess some also include that a new set is established by the player, like the Japanese guys running away from the extra life.

Take Team Fortress. In the original, players discovered that with grenades and rockets, they could use the blasts to propel themselves around the map (to be fair, it happened in the other shooters too, but I don't really think it was "intent" (there's that word again)). But when TF2 came out, this was included as a "mechanic", meaning rocket jumping and sticky jumping is something the demoman and soldier are expected to do, and maps are designed so that some places only they can reach. But now, modders are making maps that focuses exclusively on rocket jumping, where the goal is to traverse the map by flinging yourself around, bouncing from wall to wall, trying to get as far as possible, while also honing your skill for actual play. Emergence?

[gdf]

Yeah, that would be emergent.

Maybe we should draw a distinction between Far Cry 2 style designed or intended (lololol) emergent play and out of the blue, "check what we can do or make with this mechanic" emergence. Both are valid and awesome.

Thus, in the case of Noby Noby Boy, I'd say whether Takahashi designed the game that way or not - and he did, of course, because the whole purpose is one of a sandbox - is pretty irrelevant to its emergent value. Intent by the designer plays a role only in defining which type of emergent systems are available, but often these are distinct from rocket-jump style examples anyway. Emergent play is where the medium really comes into its own, allowing its inherent non-linearity and reactivity to, well, emerge.

Am I still making sense? I have no idea at this point, but I'm glad that this discussion is taking place, video game theory is still in its infancy and that means there's a lot to chew over.

[Lork]

Yeah, I know this thread is over a month old. I've been busy, but I still think this is worth talking about.

What I was getting at in my earlier post is that rather than focusing on technicalities like whether an event was intended or not, it's more useful to talk about the reason people bring up stories about emergent gameplay, which is that they're good stories in the first place. If we examine what makes those stories interesting, we can create a context by which to judge, and hopefully come to a better understanding of the elements that make them up.

For example, using a puddle of oil to set a splicer on fire in Bioshock seems mechanically identical to any good emergent event, but it's not interesting. Why is that?

I think it's because, while there are certainly a lot of independent systems interacting with each other in complex ways to produce a logical result, each individual element seems single minded the point of banality to me. I can't think of any purpose for the puddles of oil other than to set splicers on fire, so I can't bring myself to care when it inevitably happens. It lacks that crucial element of surprise that you'll find in any good story.

This is definitely a less scientific approach than people typically use to talk about gameplay, but I think game design could benefit a lot by being viewed from a more liberal angle sometimes.

Edited July 21, 2010 by Lork

[Nachimir]

I think you've nailed it :tup: