When animating on Cars, is there anything that you have wanted to animate but haven't been able to due to the limitations of using cars instead of people?The answer is yes (and no), but more importantly, this got me thinking about animation limited characters, such as cars, robots, fish, etc. I realized that this is a subject dear to my heart, hence this article. And yes, I'll expand on my answer to the question below.
Given the choice between animating a fully-articulated human character with dialog and silent, abstract character with limited articulation and vestigial features, I'll usually choose the latter. I like the freedom that you get from simplified characters; the more stylized or abstract the character is, the less the audience expects it to move in a naturalistic way, and (I think) the more likely the audience is to empathize with the character. I'm not going to get into the whole uncanny valley issue, but suffice to say that a realistic CG human has to work a lot harder than a cartoon blob to win the audience's affection.
Another thing I like about limited characters is the speed at which you can work. Because you have so few controls to work with, you have fewer keys to set, so you can spend more time experimenting. For example, on Cars the characters obviously lack arms, legs, spines, necks, etc. Most of the time I'm just treating the body of the car like a bouncing ball (or bouncing box in this case) and relying on timing and line of action to convey attitude. This means I can do most of my blocking with about 8 controls, which really speeds up my workflow! Sure, I'll also throw in the occasional tire gesture, but that doesn't add a lot more complexity to my shot, and I can keep working fast. My average weekly footage animating cars is almost double what it would be with more complex characters. I find I shoot a lot less reference when I'm planning for a limited character, because reference isn't as useful. I move in ways that a car can't, and vice versa. I can record facial expression reference, and general timing reference from my head movements, but not a lot more. I find that thumbnails are more useful, and I'll also just experiment right in the computer, because I can lay in poses so quickly and try different combinations in a short period of time. With a human character I'm shooting a lot more reference, and I'm likely blocking pose-to-pose, meticulously sculpting the poses for rhythm and balance. With cars I often block in a layered fashion, because the timing is more important to me than the poses (and how much can you really "pose" a car anyway?). This gives me more instantaneous feedback about the timing and texture of my shot.
Finally, I like animating limited characters because they pose unique challenges. It appeals to the part of my brain that likes to solve puzzles. How can I make a robot with no facial controls look sad? How can I make a car moonwalk? How can I tell a complex emotional story with no dialog, no facial expressions, and a legless, elbowless trash-compactor? I think of it like haiku; I have a very strict set of rules I have to adhere to, but within them I have lots of freedom. Limitations are essential to creative thinking and problem solving. As long as I have a clear idea of what the character's personality, mood and intent are, I've never run into a situation where I can't communicate what I want. Of course I don't always get it on the first try!
To expand on Andy's question, I often run into situations where I want to put across a particular gesture or attitude with a car, and I just can't make it read. The solution is often to not try to force human movement into a car (or robot or fish) but to try to find something unique about the limited physiology of your character that can convey the same emotion in a new way. Or in the case of an actual living creature, like a fish, trying to find a natural behavior of that creature that could also suggest a human behavior. Here are some clips to illustrate these points:
Cars - Lightning McQueen and Sally. These shots were animated by Rich Quade and Dave DeVan. Note how McQueen's front wheels are animated to suggest feet (usually they suggest hands) and how the gesture is made more "car-like" by rolling the tire on the ground. Also note the use of Sally's taillights in the three-point turn to communicate thought.
Finding Nemo - Coral. This is a deceptively simple shot of Coral saying "what?" by Shawn Krause. Notice how fish-like it feels. We've all seen fish in an aquarium do this kind of quick turn, and the animator has used this to suggest a take. Also notice that he didn't try to add in a humanistic gesture with the fins.
WALL-E - WALL-E and Eve. I animated all of this myself. There's a lot to see here, but here are some less obvious notes: WALL-E's neck compresses and expands to suggest changes in posture - when the neck compresses down, it's as if his shoulders were coming up. His arm joints suggest shoulders when they are at the top of his cube buddy, and they suggest elbows when they are down low. At around :39 WALL-E's head and body rotate around in opposite directions. Why? Because he can! At :42 I shaped Eve's eyes to look like a standard U.S. electrical socket when she illuminates the light bulb. At :56 WALL-E examines the Rubik's Cube by rotating his hand around 360 degrees. Again, because he can.
Bonus clip: Pocoyo - Pato the duck is a favorite character of mine, and he's extremely limited. Rather than trying repurpose natural duck behavior and physiology, the animators instead have chosen to invent a new vocabulary of motion for him, and use his simple body in unique ways. See how he is able to stack blocks without the use of arms:
In conclusion, I think limited characters have more fun! So if you're trying to flesh out your demo reel, why not try bringing life to a really limited character? It's a good way to develop your storytelling and problem-solving muscles, and it's a fun challenge. Happy Animating, and thanks for the question, Andy!