Wednesday, September 17, 2008

How Game Design is like Architecture


When I was talking about Too Human last week I tried to make this point that game design is more craft than art. I've been thinking on this idea for a while now, and I'm going to try to make some sense of it by comparing games to another art form, architecture.

Some kinds of art have a basic functional aspect. Because buildings, as objects, have this function of providing shelter from the elements and serving all these other practical needs, there are certain constraints on their form that precede any expressive capacity. You have to make sure no water pours thorough the roof, and that it doesn't fall down. Windows are probably a must, along with doors.

In functional media like architecture, the craft consists in finding elegant responses to the engineering challenges posed by the role of buildings in our lives. So if you're in Manhattan and you need to maximize the floor-space-to-ground-rent ratio, the steel-frame skyscraper is a graceful solution to this problem. When you get into more specialized spaces, you have to account for the fact that people will be viewing art, listening to music, or watching topless women dance in your edifice and build to suit. Because we've been constructing buildings since the dawn of man, most of these basic design problems have been ironed out long ago.

This isn't to say that architecture is mere craftsmanship, and the visual aesthetic appropriate to buildings mere ornamentation. The form we impose on our living spaces reflects, like other arts, our ideals concerning human culture-- our ambitions, our obsessions, the whole form of our common life. (John Ruskin wrote this great piece in which he said that the predominance of wrought-iron fences told you everything you needed to know about the degradation of Victorian England at the hands of industry.) Architects are often consciously guided by their convictions about beauty, or domesticity, or the cultural logic of late capitalism. Sometimes the demands of function go by the wayside, but the pursuit of significance is generally constrained by the need to keep the weather out.

I think the demands of interaction make games an intrinsically functional medium too. One of the things that makes games different from most other media is that they are intrinsically objects you have to interact with. You do things to them, they do things back to you. It's how they work. Some other forms of art are like this to some degree-- even paintings have this way of drawing your eye along a certain path through the scene. But in video games it's more extreme; you have to do something to them if you want any art to happen.

This is why I think that the craft of game design-- by which i mean mostly this whole business of finding elegant solutions to the problems posed by interaction-- is so important, maybe even predominant, compared to the more artistic or expressive aspects of game design. Before a game can set about saying something interesting-- narratively or otherwise-- it has to hold the roof up. Producing a non-buggy code is only the first hurdle (though I'm assured it consumes great proportions of development time). There are a suite of basic questions: How am I going to optimize the real estate offered by the controller so that the player can get around in the world with ease? How do I convey the mission-critical gameplay information-- her life, her ammo, the map, whatever-- to the player? What do you do about the camera? How long does the player need to play at a stretch, and when do they get to save their progress? Coming up with good answers to these questions is a prerequisite to doing anything else, and since we haven't been making games for millenia, and the basic structure of games is incredibly diverse, we still see many games with shaky foundations.

A great many games-- like Too Human-- are the story of innovative ideas dashed by the neglect of craft. I wanted to love Dead Rising (open-world zombie apocalypse set in a mall? Yes, yes, a thousand times yes), but the save points were so far apart and the hordes of zombies so thick that I could barely make any progress. The suicidal pathfinding AI made escort missions nearly impossible. Aiming guns introduced a barely usable first-person perspective. I know a lot of people made it past these issues and ended up loving the game, but for me I could never get past the obstacles posed by these design problems. (As a fan of the Metal Gear Solid series, I am capable of overlooking a litany of insane design choices; I don't even want to imagine what it was like to play MGS3 without the Subsistence camera.)

In a post last week, Michael Abbott compared the new game Spore to the iPod, which “put an immaculately designed and easy to use system in the hands of people and let them have fun with it,” and I think this is high praise indeed. Well-designed games make us forget the technical impediments to the enjoyment of art, and this is more than half the battle.

18 comments:

mtvernon said...

Good lord! You make it look so easy.

I tried writing on much the same topic earlier this week, but failed to mention the feedback loop. Sure we do things to games, but they do stuff to us, too. More directly than in just about any other medium. We're not simply spectators, we're participants. And this means that, in addition to the functionality hurdle, developers with an artistic bent have to find a means of creating space for the audience. Players shouldn't be so overstimulated as to completely miss the message.

Damn but game designers have got an awful lot to contend with.

Charles said...

I think you're overestimating just how much there is a 'right way' and a 'wrong way' to do things in game design.

Take your example of save points. There's really no tried and true method for how often a player should come across a save point. Some games allow you to save at any time, some don't allow you to save at all. A game like Resident Evil makes it very hard to save because it's trying to invoke certain emotions (fear, desperation) in the player.

Game designers often approach things as problems to be solved, but only in the sense that they're trying to fulfill particular, and arbitrary, goals. There is no true 'best practice' in game design.

Ben said...

I don't think the assertion was that there is one right way to do save points.

It's like hallways in buildings, to retain the architecture metaphor: Sometimes they need to be wide and tall, sometimes narrower and shorter. Sometimes the doors along them need to be wide and prominent, sometimes all alike and small. It all varies by what you're trying to do with the hall or save points.

After we've been making games for a hundred years, it may be clear that when you're making a survival horror game, you need to put save points X to Y minutes of tense game play apart in general in the same way that it seems obvious that the public area hallways of the main building of a university should be two stories wide and six people-widths across while the halls between offices in the back should really be 2.5 people across and just the one story.

Yeah?

Humingway said...

Yeah, this is really a great post. Henry Jenkins suggests a similar analogy in his article "Game Design as Narrative Architecture", but he sticks closely to the question of narrative, while you make great observations about the analogy in general.
So, are you making a game yet? :)

Charles said...

Hmm, I hear what you're saying Ben.

I suppose that I'm just reacting to the metaphor the way I am because it was popular in the game design community a couple of years ago and I still disagree with it.

Games, in my opinion, are a much broader concept than buildings. We pretty much know the elements of a building and are just arranging them in different ways. With games, we have no idea what the boundaries are. What's the analogue to a 'hallway', or a 'roof', or even a 'room'?

As far as game design is concerned, rules are a far, far more flexible material than anything that an architect has on hand.

Justin Keverne said...

Like any analogy it isn't perfect so looking for direct analogues to "hallways" and "rooms" probably isn't helpful.

The general point however that games are a craft makes sense. It's the point but forth by Earnest Adams and Andrew Rollings in their book on Game Design.

Form follows Function, there are no hard and fast rules for how save points or camera controls should be handled for all games, but for each specific game there is an elegant solution to those problems. Ignoring the craft of game design can mean that ideal solution is never found.

Simply put games need to be functional as well as artistic. If nobody can play the game it doesn't matter about it's deeper themes or concepts, nobody will ever experience them.

WorldMaker said...

On our school campus there are two buildings that stand out both as central class buildings and as interesting architecture... In perusing school materials for a project I glanced upon the official architectural style of the buildings and was introduced to "Neo-brutalism" in the context of architecture: an (extreme) loss of overall aesthetic appeal for more "urbane" notions such as greater exterior space on lower floors. One building hulks above a large social courtyard and the other a major throughfare; and both have ample nooks and crannies around the outsides for impromptu, private meeting/working spots.

To an extent I'm fascinated by this style's trade-offs of looks versus use. It's made me much more aware when I look at "unappealing" buildings that there may be more to the design than what is directly visible. New construction materials and engineering advances have happily given architects more visually pleasing ways to get toward some of the same functional ideals, but it's much easier to see the trade-offs involved when it's an ugly concrete block.

To bring this back to games, I think that metaphorically games have never had that "concrete block" phase in terms of the aesthetics of a game's play and it's functionality. There's no point where a game designer had to say "well, in order to get this brilliant central courtyard of wonder we'll should build everything out of concrete". The trade-offs in game design are rarely that binary, and as others mention above are huge spectra of options that in some ways we've barely scratched the surface. The architecture metaphor here reminds us that in games, at least, it often takes a lot more work to infer a designer's intent from the work, whereas architecture the recipes are more easily discoverable, particularly within a given scope of time or materials at hand.

Some people love the "brutality" of few save points in a game like Dead Rising, because it can offer an interesting challenge or yield interesting stories ("so there I was fighting my way toward the restroom when the chainsaw breaks..."). Obviously there are other ways to do both of those things, but they involve trade-offs in other areas: more writing staff, less player control...

Sometimes, perhaps, the more brutal games are the most interesting, at least to learn something from.

Iroquois Pliskin said...

@charles and ben: as the ben says, I'm not claiming that there's only one way to do things; like you say, it can be good in a survival horror to limit your access to save points or ammunition to keep the tension high. The best design choices are going to be relative to the type of experience you are trying to make.

That being said I tend to think there are some best practices when it comes to the camera and controls-- like, always put the camera on the right thumbstick unless you have a *really* good reason, drive with the right trigger, etc. This kind of stuff might become more standardized in the way ben describes, due to the physical limitations of the human hand.

I also agree with Charles that the analogy with architecture starts to break down after a while (like justin says), because there are such a wide variety of games and genres that are possible. game-worlds themselves aren't constrained by physics, and because of this there is a really broad range of design possibilities.

@worldmaker: good point. your analogy brings up a deeper issue about the trade-offs between challenge and "usability" in game design. You're right that sometimes save point scarcity can inject a really compelling degree of challenge into the game and lead to all these memorable scenarios. Personally speaking I'm in favor of just letting the player save whenever they want and finding other ways to bring challenge in some other way (you know, punish the player for failure in a way that doesn't involve forcing them to play a long section of the game.)

Nelsormensch said...

A little late to the party and I don't have time to ruminate on this properly, but I did want to complement an excellent post. I think it's very important to remember that games are crafted things and the architecture metaphor emphasizes this excellently.

Interaction with other media tends to be quite binary- either you can consume it or you cannot because it has been somehow damaged. But short of an author deliberately leaving chapters out of a novel, reading a book always works.

Successful interaction with a game (and presumably with areas of a building) is on more of a spectrum. Certain areas can be deficient, but other things work fine. And I think this is where a lot of the frustration with poor games comes from, and how games will the poorest of storylines and characterization can still be considered successful. The craft of a game can be solid, even if its creative components are deplorable and vice versa.

Unlike film or novels, we can be frustrated with a game's crafted aspects while still enjoying its creative ones. Of course, the best games do both of these things well. But remembering that we cannot sacrifice one when chasing the other is vital.

Scott said...

The metaphor holds up in another way: results of both are essentially crafted for a specific consumer. I've seen brilliantly designed and aesthetically beautiful buildings...that are clunky as hell, that get too hot, or too cold, or have bad views, or the bathroom is too far, the doorways open the wrong way, etc. etc. In short, they just don't work for their intended audience - the people who have to live and work in them. Same thing in game design. It's too easy to get caught up in what WE want, and to forget to check our mechanics or designs against the actual patterns of use. It can be as frustrating to navigate a poorly designed game (where does THAT mean? what am I supposed to do?) as to live in a poorly designed building. When we, as "craftspeople" get too clever in this way we totally blow it. If we want to make conceptual or visual art that's different; that's game design versus sculpture.

Iroquois Pliskin said...

@nelsormensch and scott: hey thanks for commenting; I always feel like after I leave a comment to a post people stop chiming in, and I like to see the conversation keep going.

You points about craft vs. art thing is on point. Part of me thinks that the *art* of game design just is this ability to create something that is enjoyable to interact with. I've been thinking about this architecture analogy for a while and for a time I was tempted to say that the function of games is to give pleasure (the same way the function of buildings is shelter), and though I'm not wedded to that idea I think sometimes inclined to say that the more "artistic" or expressive elements are pretty secondary.

Like if you look at Nintendo, probably the greates game designers of all time, their games are rarely more than charming on a narrative level (I mean this as a compliment) but they are almost always impeccably crafted. Same goes for Capcom: Resident Evil 4 is just a brilliantly made game, one of my favorites, but its' not "artsy" in its ambitions.

Scott said...

Sounds like you have fodder for another post. The game art = pleasure giving power is a good starting point for games as a whole, but I also like the terminology that nelsormensch used when talking specifically about the game design elements: the creative components and the craft components. Combining these two sides together in a design is always the trickiest part; doing it in a skillful and interesting way is where some of the art in game design lies.

George said...

Dang...

As the son of two (TWO!) architects, I have to say your comparison is right on the money. I myself had intended to post an article on this very topic at my own site, but for now I guess I can just link to yours...

My father and I have both worshipped architects like Frank Lloyd Wright and Mies van der Rohe for as long as I can remember. Even before I knew I wanted to be a game designer - when I thought I'd be a writer - we had several lengthy conversations about the similarities between architecture and nearly all other art forms.

I'm of the opinion that a great building is one from which nothing can be removed without weakening the overall design. A great story has no unnecessary elements, and everything that happens has a significance in the overall plotline. Games, likewise, are at their best when everything has a clearly defined reason for being there, and nothing has been "tacked on" without purpose.

I, personally, would be inclined to think of game design as a sort of problem-solving. This is the philosophy that many architects bring to their craft; the client, the site, and the budget have their own restrictions, and the architect's job is to create an elegant design that satisfies all of these requirements with a minimum of waste.

A game's design document, budget, timeframe and team size all impose their own restrictions, and a designer must work to meet these goals efficiently. Even in the physical construction of a level, a designer must make architectural choices - minimizing BSP usage, optimizing the placement of AI pathnodes, trimming the number of meshes onscreen to keep the framerate up, etc.

Like most architects, I consider efficiency to be one of the highest aspirations an artist can strive for. Art without craft is just an idea, lost and without proven merit.

This is the first time I've commented here, but it will surely not be the last. Heady conversation FTW, as they say.

Iroquois Pliskin said...

@scott: Yeah, maybe I should do a post on this but I'm kind of unsure what to say. There are some arts that are just straightforwardly about giving pleasure: like being a chef. Since games have expressive, capabilities that cookery lacks, I would like to be able to talk about the expressive side (you know, narrative and so on) in a way that was tightly integrated with talking about pleasure. If y'all have any ideas for that it'd be sweet.

@george: hey thanks for the kind words. I have this same thing were I prefer an efficiently designed game to an ambitious mess. (Like Metroid Zero mission: that game is so well designed: nothing is unnecessary.) I don't know much about games design but I'm led to believe it's about those things you talk about: coming up with efficient solutions to the problems that arise.

I read the Tracy Kidder book Soul of a New Machine last year and one thing that stuck with me is how much computer design is about close attention to detail and problem solving.

nobody said...

The more I read your posts the more I hate you. Let's be clear, hate is not too strong a word.

I read this the day you posted it and at that time I agreed with every eloquent phrasing and insightful reference. I waited for my thoughts to congeal into words. When I returned I found such thoughtful comments that my response was challenged as I considered their perspectives.

I agree with you. Games are more craft than art. But architecture is sublime and spot on. Charles' point about the fluidity of the medium when compared to architecture is valid. Yet as I considered his ideas I fumbled with my controller and felt that my preferences for playing SoulCalibur 4 on the PS controller had some significance. The bricks and mortar of game design may well be the chip set, software kit, drive type and speed (Blu-Ray vs DVD vs hard drive), online service capabilities and cost, target demographic and, when all is said and done, the game itself. There are certain "established rules" for game types, RPGs, platformers, first person shooters.

George's post says it better than I do, so many technical aspects that restrict "creative" interpretation. The possibilities aren't quite as broad as Charles' would have us believe. At least, that's my completely ignorant perspective. I agree with the idea of Charles' post, but not the reality of it.

Jonathan Blow turns heads with his game Braid not because it's universal or even arty, but because as a designer he is focussed on the craft of making a good game. His bombast is well known, and now it is justified.

Wordsmythe said...

I know I'm a couple million miles behind on this. I blame my day job working with law and finance.

Regarding "best practices when it comes to camera and controls": This sounds a lot like genre theory in other mediums. That is, an artist working within a genre is expected to stick to the norms of that genre, with the occasional departure from the genre's norms being an intentional move with added meaning. (I.e., it's done with "good reason.")

Something that comes to my mind when people reference architecture is the old engineer joke that architects make pretty designs, but the engineers make sure the thing stands up. Sometimes the joke explicitly references making sure the roof doesn't leak. I realize that good architects have a pretty solid understanding of physics and the various techniques and materials available, just as they also have to manage costs and customer interaction. The point is that architects usually work together with someone who specializes in the practical application. I'm not aware of any major designers sharing the spotlight and responsibility in this way, but then I'm also not much of an industry insider.

David said...

Understanding that you want to make a point about games by comparing them to architecture, I'd offer a few points about architecture that might extend the metaphor, or maybe just break it!

Architecture has a long history of separating itself from building. So, building is about keep the heat and cold in and the rain out. Architecture is about something else. That something else is always culture. Sometimes its culture as technology, sometimes culture as history, sometimes culture as expression.

When it comes to games, I think that game designers are still struggling to wear both hats--structural engineer and design artist--the architect.

Let's face it, Wright's buildings leaked. Meis was a genius, but odds are he couldn't produce a single page of construction documents. Game design needs more visionaries, and more talented hard nosed construction crews!

-- David (buzzcut.com)

resume for mba application said...

It took us three years to build the NeXT computer. If we'd given customers what they said they wanted, we'd have built a computer they'd have been happy with a year after we spoke to them - not something they'd want now.