Lessons From An Intergalactic Puffball [Part 3]

Well, this will be the last portion of this long-winded series. For those of you who've read the whole thing, kudos and thanks for your interest. For those who haven't... I'll try to be more interesting in the future.

What Makes a Character


I can't help but marvel sometimes at the wide range of characters that have stood out over the years in this industry. Especially when you look at really simple ones like Kirby. Just a pink ball with a face, big floppy shoes and stubby little arms. We know part of the reason for this simplicity is the time period of his debut. When Kirby appeared in Kirby's Dream Land on the GameBoy in 1992, the developers at HAL Laboratories were somewhat more limited in terms of graphics capabilities as opposed to today's technology. Kirby's most interesting traits had to clearly stand out even at a small size.

Of course, as time progressed it was possible for artists to refine his appearance while retaining the essential elements that make him who he is.



I'm sure there were times in Kirby's younger days when the developers wished he could appear onscreen with greater detail. But I think being limited graphically had its advantages too. In a way, it kind of forced them to isolate those essential elements I mentioned earlier. Basically, they had a few pixels to work with, and they had to ask themselves: "What details are absolutely necessary for this sprite to look like 'Kirby'?" Without color, the glare in his eyes, or his rosy cheeks we can still look at both images above and recognize them as the same being. There was no room on the GameBoy for those finer details, but they weren't really needed at the time either.

What Makes a Face

This reminds me of some of the motives behind Picasso's work. Without getting too off subject, I'll just say that for a time, Picasso moved away from painting realistically; instead his paintings were driven by the motivation to generate recognition of a subject from the audience, using as little visual information as possible.



Picasso called this piece Man With Hat. So obviously the man had, among other things, a human being in mind when he made this. But I'm sure if we saw a person walking around the streets who looked like this we'd likely run in the opposite direction, or at least recommend them to the nearest medical establishment.

Regardless of the fact that this doesn't look like a realistic human, there are a few isolated elements here that make me think "This picture has a face in it." Two identical polygons with dots inside of them translate as eyes, a broad sloping line between them translates as a nose, and some vague curves on the right side translate as an ear. Even though Picasso didn't render every hair, pore, and molecule, he left enough for us to believe we were looking at a person.

Analytic Cubism not your cup of tea? Think of a big yellow smiley "face". Again, living things don't actually have faces like this--it's really just a yellow circle, two identical ovals and a line below, but we still call it a face. We've been trained one way or another to isolate the basic ingredients of a subject, in this case a face, even without the unnecessary details. All you really need is two identical shapes on the same horizontal plane, and a different shape below for a lot of people to think "That's a face", since we often place most importance on the eyes and mouth.

Character Details: What's Necessary vs. What's Possible

You might be scratching your head and wondering:

Why does any of this matter in video games?

Technology has advanced. If we want the player to see a face, we can just draw or model an actual face.

 A valid point. Artmaking technology is more powerful than ever, and with more detailed and refined art becoming the standard, anything less could be considered amateurish or sloppy looking. Is there still room for simple looking characters like Kirby in this medium?

In a sense I feel video game art is moving in the opposite direction that 19th century Western painting did. With the advent of photography and other factors, painters moved away from painstaking realism towards a host of other, more abstract forms of visual representation. With more polygons and more powerful systems, much of video game art is approaching that painstaking realism with its models and sprites, leaving behind abstract forms like a few colored squares, which were the fewest symbols necessary to convey a character.

I don't mind this shift, it's really cool to look at how our standards have changed over the years. I definitely don't mind the kind of detail we're capable of adding to our characters today--we should be grateful for the tools we have. But I think in terms of character design, it can be useful to look at it in terms of what details are absolutely necessary, and which ones will more quickly be forgotten by players. To put it simply, what would our characters today look like if they were reduced to their essential details? If they had been constrained to only a few pixels onscreen?



It can be fun to load up on details right from the start, but when we're trying to design memorable characters who can stand out from the crowd, exercises like this might be useful to start off with as well. We can ask questions in the initial stages of design: Are all of those buttons really necessary? Will people remember that crazy hairstyle? When reduced to a few pixels, does my character still stand out from others in its genre? When I think of Ezio from the Assassin's Creed series shown above, I remember the unique shape of his hood, and the blades near the wrist, and I think the designers did a great job at making sure details like that stood out. I probably couldn't tell you how many feathers were on his belt crest without looking, on the other hand. All characters have traits that really make them unique, but it can be important to emphasize those from the start, instead of getting lost in just how many frills that dress should have too early on.

Overall I'm not worried about simpler character designs being totally left to the wayside. The main Angry Birds character is a good example--just a round, bright red robin with its enormous eyebrows, it doesn't even have wings. We can learn from simple designs like this, Kirby, and the practitioners of Cubism, and consider what's really necessary for people to look at a visual and think "This looks like [your character's name]".

No comments:

Post a Comment