Postcards from Kyrat, pt 1

Following up from a can of worms I opened on Twitter earlier, I wanted to expand on what I was talking about in relation to environment design, space and discovery.

It comes from thinking and chatting about Far Cry 4 and the volume of icons, micro-quests and collectibles in the world, which results in a map that looks like Otto Neurath lost his breakfast all over it. Now, I’m a big Far Cry fan, and I’m loving FC4 but I realised I was missing a real emotional connection to the world. Trying to figure out what the problem was, it really came down to the way I was exploring and navigating and how this was driven by all of those icons. I’m not alone in this, and FC isn’t the only culprit: there’s been a lot of similar complaints about the latest Assassin’s Creed games – also by Ubisoft, which doesn’t help the building sense of cookie-cutter design-by-marketing from a studio that a few years ago could really shout out about the innovation right across the board of its output.

Anyway, the issue is the conceptual frame of experience being encouraged, and it’s something we’ve really grappled with in the past – you could argue it’s one of the major design flaws of Amnesia: A Machine for Pigs, for example. Essentially, it boils down to the idea of media schema. Schema theory has been kicking around for an age, but is probably best approached via the psychologist Frederick Bartlett in the 1930s, who wrote about narrative and the way that we process information into stories via a set of pre-learned structures or schema. This then got picked up by AI thinkers like Roger Schank through the later part of the 20th century, and works its way into a new incarnation as affordances via JJ Gibson and Donald Norman, all of whom should be required reading for game designers. The basic principle is that you have these architectures in your head, and when presented with a situation with enough recognisable features, one of these kicks into play and frames your understanding and behaviour accordingly. In other words, you might think you are acting and processing events fully consciously, but the range of thinking, feeling and behaving is being mediated by a semi-automated response that you’ve developed to help you predict how things are going to play out. Media schema takes this concept, which is really derived from evolutionary psychology, and applies it in short-form to how you understand different media experiences. Scott McCloud talks about this in his brilliant book Understanding Comics (which is also required reading for game designers) in terms of the automatic projection of continuity of narrative between panels, and so the importance of a comic designer writing for the space between these panels. It's really the same process as when you take a zoetrope (or 60fps digital display) and turn it into moving pictures. In essence, when you engage with a piece of media, automatic systems of understanding kick in and mediate this experience.

You can then narrow this down as well – if you play a lot of FPS games, there are standard behaviours you learn in order to succeed at playing. There’s a smooth continuity between the biological process of stitching sets of static images into a moving whole, being able to understand the reality of television images without having to consciously consider them, the ease with which a heavy gamer picks up new skills in-game and the way we read a Far Cry 4 map. It’s part of the same process, and at root it's our mind farming out predictable scenarios to automated responses. Which is in many ways a fancy way of saying we learn stuff, right?

So here’s the thing. Your mind looks for the easy way out, because evolution is, for all its brilliance, a lazy bastard and so are you. Once you establish a mode of interaction, it takes effort to break or retrain that. You tend to automate your response because it’s more efficient, freeing up the mind for important questions like whether it’s possible to attach C4 to an elephant before driving it into an enemy outpost. But this means that all of the complexity of a world can easily be reduced or flattened to a series of push-button solutions, and that can have an impact upon emotional engagement with the actual content of that world.

In other words, if I’m playing Far Cry for collectibles, the way I engage with the world changes, as it’s now about reaching somewhere to achieve something. That’s very different to moving around for the self-generated reward of discovery. Played like this, Far Cry doesn’t present a world, it presents a resource for gathering rewards.

That’s not altogether fair of course. Far Cry presents an outstandingly beautiful and richly dynamic world that is amazing to explore for its own sake. The reward of seeing emergent behaviour unfold in front of you is just wonderful. But it’s undercut if you are ploughing from location to location, just to hunt collectibles. There was a period last night where I was just using the buzzcopter to zip backwards and forwards wrapping up Masks and Mana Wheels and Diary Entries and… and… and… and it was just Kyrat by air, the whole environment becoming essentially irrelevant as I’d flattened out play to being a lean, mean XP hoovering machine. Which sucks, because that’s the least interesting part of the Far Cry experience by a considerable margin. But it’s so bloody seductive, because it’s basically looting and grinding and if we pretend that as gamers that’s not like crack cocaine to our wizened, grubby little souls then it only proves we have the self-reflection of an Enron executive. The moment you start lacing a map with multiple collectibles, it flattens it because media schema start kicking in. 3 more masks for 25% off weapons is a much, much easier reward to quantify than “potentially beautiful valley ahead” – both in terms of communicating the targets of a game across a large organisation, and for the player. You can’t heat-map and metric up the insubstantial wonder of discovery, nor does it give you cheaper napalm. So not only is using this kind of design a useful tool to evaluate how a map is functioning in terms of sharing information about player behaviour, it’s also a very efficient way in which our schema can self-evaluate the benefits in terms of in-game decision making. And that’s potentially problematic, because of this flattening effect.

In simple terms, it confuses the destination with the journey, and that’s a killer bodyblow to experiential game design. It makes the experience secondary to a quantifiable series of cause and effect transactions – the mountain ceases to be a mountain but a means or an obstacle to an abstract value you are trying to manipulate by reaching a set of co-ordinates, which is a real shame when the mountain in question is a lovingly crafted and achingly beautiful expression of the artists’ vision and a super powerful tool for creating an emotional response in the player.

Different strokes of course, and not all players play for the same thing. And that’s the major issue facing AAA development. How do you protect a wildly divergent set of players from each other and the conflicting desires they have. Ubisoft used to be really good at walking that line. It feels, just from this one player's perspective, that they've struggled to keep that balance. It may be nothing more than feature-creep writ large, but I hope the trend is short-lived, because Kyrat has so much more to offer than that.

Anyway. That’s part one. I also started talking about the difficulty in making a game that doesn’t feature a lot of repeating gameplay in terms of how much environmental and other assets you can chew through in a fraction of the time you would in a ‘normal’ game. I’ll try and brain-dump thoughts about that tomorrow…