Much as I enjoy pursuing my own projects (like "What is Torch Tiger?") I am not very disciplined at forcing myself to develop great and original character designs. The part of a project I really want to spend all my time on is the telling of the story - working on the staging, the expressions and the clarity and pace of the storytelling.
So as far as character designs go, I try to get through it so I can get to the stuff I like. That seems like a blasphemous statement, huh? The web is full of great websites where great artists post amazing and original character designs (look no further than my links!). Character Design is an art and a discipline and I totally admire people who are good at it. So I am not writing this post to share all my wonderful knowledge or anything, but heck, I'll do my best to share what I've learned from my limited experience in this arena (I had some wonderful teachers at CalArts that taught amazing classes on this topic and it's no reflection on them how little of it stuck with me).
So admittedly my standards for designing characters is low (I'm not proud of that, but it's true). Here are the criteria I try to meet when doing designs:
1. It should be something I understand in three dimensions and that I can draw from any angle. It should look appealing from every angle.
2. The design should lend itself to expressive expressions and poses. Those expressions and poses should be appealing as well.
3. It should look as original as possible; not like something you've seen before
But as far as that last one goes - everyone has a different idea of that. Obviously my designs aren't exactly ground-breaking or shockingly fresh and original. But I do try very hard to make sure I'm not repeating something I've seen before.
When I set out to design some vultures for my Torch Tiger, I had grand aspirations to do something really new and different. I didn't want to draw the cliche Disney cartoon vulture so I did a ton of research to try and find a new approach, to re-invent the cartoon vulture.
Guess what? I ended up drawing the stereotypical Disney cartoon vulture. The problem was that the more I looked at photos of real vultures the more they looked like the stock vulture design to me. They really do have those weird tufts of feathers at the base of their necks. When I tired to re-invent them and pull more from the awesome shapes that real vultures have they lost some of their appeal and expressiveness. Also the tone of the story got more disturbing every time I pushed them towards the more realistic, or evil or menacing. Cartoony vultures made it more palatable that they were talking about chowing down on a corpse.
Real vultures are so cool looking. They are ripe for re-invention but I will have to leave that up to a better artist.
So I tried to put my energy into designing three distinct personalities (more on that later).
Anyway, that leads to what I think is the most important law of character design, and one that doesn't get talked about enough, is:
5. The character design should describe the personality of the character.
I'm a big believer in the idea that when you look at a design, you should get an immediate sense of what that character's personality is. It's just like in live-action: directors cast actors who are going to bring a physical presence to the role and look like what the audience expects the character to look like.
There is a big difference between the types of actors cast as leading men and those who are "character actors" who tend to be cast in the supporting type of roles over and over.
People who are "character actors" (and not leading men or women) have a look and a persona that comes through that matches the type of character the director is trying to put over within the film (people like Steve Buscemi or Paul Giamatti) . That way the director doesn't have to spend a lot of time establishing what "type" of character this is, when you see the actor he's picked you get a sense of what type of character this is. It's a "shorthand" that helps the telling of the story.
If the designs of your characters don't work well in telling the audience what type of personality they are then you have to do extra work in informing the audience what they are like. The best characters are ones in which the design, voice acting and animation are all unified and work together in concert to create a great character. The sooner the audience can hook into exactly what type of character they're dealing with, the quicker they can engage with the story and focus on the events that are unfolding instead of trying to decipher what type of character they're seeing.
Look no further than "Up" to find a great example of this. Everything about their appearance is so spot-on specific to their personalities, and the animation and voice casting on both characters is phenomenal. They are extreme opposites of each other in every way which is always great for interest and entertainment.
I'm going to talk a bit more about "Up" here and I'll keep it very vague...this shouldn't be much of a spoiler but in case you want to avoid all knowledge of the film skip the next paragraph.
Also at one point in the movie "Up" there is an example of a character where the design and voice are in direct contrast to each other and it's very funny. The reason it works so well is because when you see the design you expect one thing and then you get another. But something like that won't work unless the design is carefully worked out so that when you see the character you're primed to hear one thing in particular...and then it's turned on it's head, which is funny. But if the design wasn't done in the right way, the joke would fall flat. Another aspect to consider is the point of the movie where this happens. It happens fairly late into the movie, which I think is part of why it works. Because all through the first half of the movie, the film has been crafted perfectly so that all of the voices fit the designs perfectly. So when you come to the point where you meet a character who has a voice that doesn't match his character...it's much funnier. If a movie opens with a character with a voice that doesn't match his design, it will never work as well, because the audience won't really know if that's common or uncommon for the world of the film. As far as they know at that point the film might take place in a world where everyone talks in a way that contrasts their design.
So when I was designing my "Torch Tiger" designs I knew I only had eight pages to put over my story and characters. It's the story of three desperately hungry vultures who conspire to knock off a passing cowboy so they can feast on his carcass.
I wanted to come up with three character designs that telegraphed their personalities right away. There's a regular, main "Straight" vulture, and then the other two are variations on him: a slow, dim-witted heavy vulture (who never talks) and a skinny, agitated and stressed-out vulture (who talks constantly). I tried to keep their personalities and characterizations simple because eight pages is really a short space to tell a story and it's really difficult (if not impossible) to get any kind of nuanced or complicated characterization within that amount of time.
Their would-be victim of a cowboy presents more of an interesting problem: it's kind of a macabre story so I wanted the designs to be goofy enough to keep the story light. Originally I was planning to make him look pretty seedy and disreputable, like a western bad guy, so that you would be on the "side" of the vultures and not be disgusted at the thought that they were "doing him in". If he looks like he's an unsavory character and maybe "deserves" to be done in then maybe the story is a bit more palatable and you can laugh at it instead of being disgusted with it.
Also I toyed with making him look really dumb because that can also help us not feel so sympathetic towards a character.
But somehow I ended up with a sort of neutral-looking guy. This is going to sound strange, but when I was drawing a more extreme version of this character it threw the story into a different framework. When I drew him as a typical kind of Western bad guy, your mind sees him and says "Oh, that's a criminal", and then you're waiting for that idea to come into play somewhere, for his "criminality" to mean something within the story. But I didn't have room for that within eight pages, so I couldn't really play that out. Plus, the more dangerous and menacing he looked, the more dark the tone of the story became. So I kind of ended up with a bland "everyman" cowboy, because that's what the story seemed to need. The story isn't really about the cowboy, it's about the vultures, so what I needed more than anything was just a target for their mischief that read immediately as a cowboy...nothing more, nothing less. Any "inflection" I put on him seemed to throw the story into more complicated territory and added confusion to the simplicity of the tale I was trying to tell.
So I thought I came up with a pretty original looking cowboy, and he was fun to draw and I could get pretty expressive with him. As I was cleaning up my comic, my eight-year-old daughter looked over my shoulder and said "Oh, you're drawing that cowboy from 'Home on the Range'". So I guess I wasn't being as original as I thought. I worked on "Home on the Range" and never considered the similarities until she pointed it out. By the time she mentioned it it was too late to change it. Next year I'll have to show her my designs earlier in the process.
So that's it from an overall philosophical standpoint. Hopefully someday I can talk about the more artistic and technical side of character design, like proportions and rhythm and stuff.
Well there you have it. Hope that helps somebody, somewhere. In summation: a design - be it for an environment, a character, or even a prop - should always serve the story first. A cool design that doesn't serve the story is just a cool design. It's like wearing shoes made out of gold. It may look cool but it's going to be uncomfortable and it's not going to fulfill it's primary function: to help you walk around from place to place.
All images copyright 2009 Mark Kennedy