Monday, September 19, 2011

On My Mind: Design: Art or Science

It's been a while since this topic was mentioned and I forgot where or why it popped up, but I felt compelled to write something about it. I recall being asked a few times whether design is an art or a science, and I'm sure you've all seen the endless debates on which one is correct or more important. So I might as well throw in my two cents here:

Why does it have to be a binary choice? Design should be both art and science. Anyone who tells you otherwise isn't doing their job properly.

(Note: I'm going to put a few pictures to get my point across: I'm not saying these are bad games, in fact, I actually liked these games a lot...)

Design is Artistic: any designer who tells otherwise are terrible designers who would only make cookie cutter, boiler plate ideas that caters to no one but a spreadsheet.

Sometimes, a game can still get by with just the bare minimum, hitting all the bullet points that people expect to have in a video game: Oh, it has online multiplayer; Oh it has co-op; Oh it has a ranking system, etc... But then is this really a game, or merely another manufactured clone of something else you would have rather played? It's especially worse when you realized this game you're playing has all the parts lifted directly from another game. Sure, imitation is the sincerest form of flattery, but you have to ask whether the people on the other end really cared about the product they're making.

I liken a game designed on paper as something without a soul: Sure, it may be technically competent and without fault, but it comes off bland and manufactured. Even as the game conveys emotions through it's characters or setting, it feels so contrived that you can already tell that the designers looked at "the demographic of the audience" and planned their game around it.

You may or may not like where EA as a company is right now, but a long time ago, they had put out an ad with the headline "Can a Computer Make You Cry?"

I'm not sure if they had made that question their vision back then, but I think it's a pretty noble goal even today. If a game isn't there to create player emotion, then you might as well be just doing your everyday menial tasks (and grinding out your level caps). Other "entertainment medium" elicit emotions like joy, fear, sadness and anger; if games are about entertainment, then they should strive to do the same, which means staying away from designing strictly from a mechanical standpoint.


Design is Scientific: any designer who tells you otherwise are terrible designers who wouldn't understand player feedback when someone tell them the game "isn't fun to play".

Until very recently, the more artistic endeavours from the likes of Suda51 (Killer 7, Shadows of the Damned, pictured bottom) and Tetsuya Mizuguchi (Every Extend Extra Extreme, pictured below) were a rarity in gaming. Their games often have flaws, but still receive high praises for their creative look compared other games outside. However, typically, the games always suffer from terrible sales, driving publishers away.

The quick and dirty answer that everyone can understand is that these games are going to be harder to market: People want comfort food, they want games they are familiar with, and huge deviations can scare people away. In some sense, it's why "Scary Movie" is a summer blockbuster: people like the familiarity, even if it's unoriginal, forced and manufactured.

However, another cause I like to point out is getting proper, 'scientific' feedback. A lot of times, you'll see some of these games commit the most obvious mistakes, as if no one actually played through them. Sure, the creative process is sometimes about 'gut feeling', putting in what feels natural and makes sense, but other times, doing proper 'scientific research' can yield much more favourable, more refined experience. It's the difference between "I think this weapon should have 0.5 seconds to reload" vs "Our test shows that 0.5 seconds is too long and players are not choosing the weapon".

Take Limbo (pictured above) for example: while the game is about trying and dying, and the player's journey through the experience, there were a handful of places where the only way players will get by if they've died multiple times to learn what not to do. Sure, it may be the experience that the developers intended to give, but it's sure to alienate players who are frustrated with such mechanic (and goes against the idea of games as a mastery of skill, not mastery of guessing what the developer wants you to do). There were points in the game where you see obvious attempts at placing hints to show players what can happen, and I wonder if those sections were even more difficult before proper testing and feedback.

In fact, if you were to analyze Suda51's last few games, you'll realize that his games have gotten better and better at smoothing out the rough edges yet still retaining some of his absurd ideas. You put Killer7 and Shadows of the Damned next to each other, and you can still tell it's from the same creators, but Shadows clearly have had more refinement in terms of managing and meeting players expectation. In a third person shooter, people expect fluid, direct control (Shadows), not point and click wonky interface (Killer7). Some would argue that Suda's games have gotten more bland as of result, but as a tradeoff for a more playable game that reaches a wider audience? I think it's a decent tradeoff.


Personally, I know that I tend to favour the scientific side of doing things. If I can get all the info about how people are playing my game, why wouldn't I take it to improve up on it? Yet I understand that design should not be entirely based around player feedback, and sometimes it's worth it to go against what is expected to make a more interesting idea work. I hope other designers would take a look at where they fit in that spectrum, and be mindful of how they look at the other side when they create stuff.


  1. First thought of Halo. Created nothing really new, but polished existing features.

    There's definitely an art in creating new ideas, and a different art in polishing and presenting them. Mathematical art, perhaps? You're never going to get rid of the hackery and riding on previous success, but hopefully more game designers will learn and improve.

  2. Building on prior success, however, is much different than hackery. Take for example, Mario Kart 64 and Diddy Kong Racing. Early on when DKR was shown, people were labelling it as a clone of Mario Kart, but it did offer something that was different. But without looking at what worked and what didn't, who knows what it could have been.

  3. Zynga is also a good example of polishing (and ripping off) existing ideas.

    Good word choice. We should demand: More building on previous success and less riding on it!