Wednesday, May 28, 2014

World designs, go open or go hub?

Welcome to Empire City
In my mind, Infamous is one of the first, and last, open world games to truly present a compelling open world. Why do I say this? The world recognizes you. You're a savior or a tyrant, and the people respond to you in as appropriate a manner as they can find to. They gather together to fend for themselves against the gangs, as well as praise you, or they join together to protest your actions and try to kill you.

More than this action alone, they do this together. The people are represented as intelligent, rather than bumbling over confident fools that don't realize how powerful you are. They know you've got power, and they know they can't match it alone.

Now, you may say, other open world games do this too! They have police come after you, or they say something about your driving or clambering up buildings. You're right, of course, but that's such a mundane and limited response, it's hardly even noteworthy.

Infamous takes this all a step further. They integrate the people in a clever way that complements what's occurring in the world. Then, they subtly change how your character and the world appears to hammer it all home. You're influencing things, for better or worse, the world's getting cleaner or it's getting gritter, the people are growing fond of you or they're finding you absolutely intolerable and they're taking action.

Later installments in the Infamous series, unfortunately, seemed to tone down this level of influence, much to my dismay. They became like every other open world game.

Let's go to New York City, Empire City's inspiration
Supposedly the game's highest structure
Spider-Man 2, the tie-in game with the movie of the same name, was hailed by many as a fantastic game. Being the first open world Spider-Man game, it was especially praised for the freedom of movement as the renowned webslinger, compared to earlier, highly limited games (due in large part to hardware constraints). However, as the first open world superhero game, to my knowledge, it was also accompanied by many of the same issues as every other open world title of the time and to date.

In all honesty, I never completed the game, due in some part to these, as well as not enjoying it very much. I do, nevertheless, recall the flimsy heroics of recovering children's balloons, fighting rooftop criminals (as they strangely love heights, something even Infamous doesn't escape), and some other odd side tasks. What's problematic about all of this is nothing in particular, except that ignoring them is largely inconsequential. The world doesn't care either about your menial acts or your grander acts in the main story.

Infamous only barely skirts outside of this issue through its karma system, or its reputation system in other words. You beat up civilians? You're a jerk, you're bad. You save them or knock out criminals instead of killing them? You're a hero, you're good. Your menial actions are relevant, but your grander actions still aren't entirely as influential as you might hope.

Spider-Man 2? The people don't care what you're doing. They're scenery, they're nothing more than a background crowd, a backcrowd, to everything you're doing. They are only present to give the sense that you're navigating an inhabited city, and nothing more.

In Infamous, some of them are transformed into an old school chorus, as found in old Shakespearean or Greek plays. They respond to their current situation as I've already mentioned above.

Beyond the unresponsiveness of the world to your behavior, except in very minor ways, it also treats the world as an extremely diluted hub for your activities. What do I mean by this?

The diluting effects of persistence
All open world games are, by their design, large open persistent hubs of activity for you to experience, but by focusing on persistence, they dilute their world's potential in numerous ways. Here are a few of the ways...
  1. Aesthetic variance. By limiting a game's world to a single location, they can, if they choose a real location, improve its accurate resemblance. In the process though, they run into the same problem we have in the real world. Everything comes to look very samey. All buildings are, approximately, the same. If you enjoy the gameplay, it may not be a problem, but if you enjoy gameplay and world design, you may find yourself frustrated by never being able to take your gameplay out of the setting it's rooted in.
  2. Character capability. Once you've set your world, be it in a fictional city setting or other real world approximation, you've set the expectations for what your character's capable of. Unless you go to lengths to emphasize otherwise, such as in Prototype, Infamous, Assassin's Creed, or Elder Scrolls, players will expect to use vehicles and shoot guns, but not be able to enter or scale every structure or jump very high.
  3. Content variance. This follows entirely from a persistent world and aesthetic variance. It's not merely a case of the side-content being reduced to mini-games or limited to what makes sense within the world, it's that often, this content never takes you out of the hub. You are always stuck in New York City, Empire City, the Middle East, or the province of the moment in the Elder Scrolls series. By restricting content to the hub, and rarely, if ever, branching into a temporary world, the content is highly diminished.
In other words, a persistent open world grants much freedom in movement and emergent adventures throughout, but in the process sacrifices a great amount for the sake of aesthetic, functional, and event cohesion. This doesn't have to be the case, but this is often the case in my experience, and it makes sense on some interesting levels. By limiting the design to a single setting, you can better focus how it all fits together, and if your setting is largely homogenous on an aesthetic level, even better, as it requires less work to produce distinct structural models and textures. 

If done "well" then, an open world design could potentially prove easier than a more nuanced, intricate closed world (e.g. linear games, level to level) or hub world design. It's no great wonder in this case, really, considering the success of Minecraft and the rise of games with procedurally generated worlds. However, it should be noted, that a procedurally generated open world will almost always be easier to make than a handcrafted open world, and it's for this reason that, while an open world game could prove easier, it often does not since many would rather handcraft their worlds than work away at calibrating algorithms.

Although, if made really well, a procedurally generated world will be equally as difficult to make as a handcrafted open world, due to ensuring everything is generated as desired. (See games such as Limit Theory for a crash course in difficult procedural generation design work.)

Nevertheless, to return to the point, it is simply the case that despite the difficulty of handcrafting large worlds, it is often considered worth the effort because as appealing as a unique world each time may be, content implementation and player directing is much, much harder to figure out. There are workarounds that wouldn't be too difficult, but for most behind the creation of games of this type, they want a core story that can be followed, a structured setting that is easy to learn and navigate, and easy content distribution and mapping to ensure the player discovers it.

There's nothing wrong with any of that, except that they could do this just as easily in a sequential hub world design.

Okay, we get it, what's the deal with hubs?
As noted, all open world games are already hub worlds by design, the difference is that they are most like hollowed out, branch less tree trunks than anything else. You live inside the trunk, and you never leave, you don't even realize you're in a hub necessarily, because everything you can see is strewn about it. The only thing that reminds you of this is when you hit its borders, which often designers try their damnedest to hide because they want you thinking about what's before you, not what's around you.

Again, tree trunks are nice and everything, and arranging my nuts around it is fun for a time, but occasionally, I want to go outside...I want to see other trees, or at least step out on the branches. This is where what I promote comes in, which is nothing revolutionary or new by any means whatsoever, but the revisiting of sequential hubs and branches. What's this then?

It's Spyro the Dragon, Spyro: Ripto's Rage, and Spyro: Year of the Dragon, it's Crash Bandicoot 2: Cortex Strikes Back and Crash Bandicoot: Warped. It's the Sly Cooper series and Vexx. It's Dustforce and Bastion.

Each of these games has either a series of hubs with unique branches that follow the theme of their hub, or a central hub with distinct branches that in themselves act as hubs. This is fantastic because it retains the focus and cohesion of a single-setting large open world, but divides it to produce varying, and in my mind at least, more interesting, experiences. The major problems of such a design approach are the difficulty of realizing distinct looks for each hub and its branches as well as designing uniquely appropriate events for each, not to mention the loss of persistent crossover effects from hub to hub, but these are problems worth solving, rather than neglecting. Why? They can prove inspiring and they create more fascinatingly unique experiences to encounter than yet another Earthly natural landscape or cityscape.

Better yet, the lessons learned from revisiting this design approach could improve the open worlds of games today through more minute focus on little details, that an open world design may incidentally overlook.

In the end...
Whether you go open or hub world, it's best to recognize the necessity of minutiae and variance in both to improve the overall experience of the player. It may be overlooked, but most things, when done the best, often are anyway.

No comments:

Post a Comment