There have been some great comments to the last few posts that I wanted to explore further:
- Is video reference a creative dead end, akin to rotoscoping, and even a form of creative plagiarism?
- What about lipstick cam footage? (That wasn’t from the comments, but I was thinking about it today.)
- How does an animator using reference ‘get outside of themselves’ to keep each performance fresh and unique?
- Should animators take acting classes so their reference will be better?
Keith Lango gave a strong argument against self-generated video reference, and I thank him for raising some red flags. It absolutely can become a crutch, and it can lead to technically strong animation that lacks any zing.
If you’re shooting reference to copy from, rather than to learn from, then you won’t advance as an animator. Just copying will help you produce a particular shot with more nuance and better biomechanics, but you risk becoming an overfed mo-cap machine. In fact, one of the things that makes most mo-cap look unsatisfactory is that there’s no discrimination to what’s captured. A mo-cap system captures as much data as it can, without judgment. The key, distinctive elements of a behavior are given the same weight as random, irrelevant movements (while some key information is completely missed).
As an animator using any kind of reference, we have to have the intelligence and artistic judgment to edit out the ‘noise’ and exaggerate the good stuff. Ah, yes, exaggeration. One of Frank and Ollie’s 12 principles, and one which many people consider the very soul of animation. Mike Nguyen talks about exactly that here with a beautiful pencil test to illustrate.
We want to go far beyond any form of rotoscoping or roto-lite. As I’ve said, I’m a crap actor. The last thing I want is animation that’s a little bit better than what I can do as an actual actor. I want animation that’s a million times more interesting than any performance I could ever do.
So, should we avoid shooting our own reference if we aren’t great actors? I don’t think so. Reference is a chance to go beyond the ‘rules of thumb’ that have been handed down. It’s a chance to avoid using the lore we’ve learned from other animators, and observe for ourselves how something really works. For example, I recently read these two rules attributed to Ollie Johnston (among his ‘12 rules for expressions‘):
10. Eyes in close-up should move 3 frames ahead of accent.
11. In a blink, eyes should close 3-4 frames ahead of accent.
Those might be two good general rules, but really, do you want to follow that by rote every time you do a close up? With every character? During any kind of emotion? I don’t. It’s by looking very closely at video reference of real people, myself included, that allows me to consider other possibilities.
Speaking of eyes in particular, I think this is a place where video reference has it all over on using a mirror as an aid. Haven’t we all been encouraged to use a mirror when doing facial acting or dialog? The problem with mirrors are twofold. First, you’re always looking right at yourself with a mirror. It’s very unnatural, since you will almost never have your character looking directly into the camera.
Second, most of us act things out in slow motion when acting into a mirror. Otherwise things go by too quickly. But what we do in slow motion is completely different than what we do at full speed. Not that you should never use a mirror — just know that, like video reference, it’s not the end all and be all.
The key with video reference is to not just use it to copy from, but to break down a performance, analyze what’s good about it, and then incorporate and exaggerate that.
Walks are another good place for video reference. How many hundreds of pages of animation books have been devoted to the rules for walks? Unfortunately, a lot of those rules are contradictory, overly complex, or don’t really work that well except for extremely cartoony characters. For example, I see students struggling to learn the rules for, say, what the hips do during a walk. Hey, guess what, hip movement varies dramatically from person to person. Put on a white belt and some dark pants, wrap a line of masking tape around your knees, and film yourself. Do the same with some friends. See what really happens to the hips, the feet, the knees the ups and downs, instead of relying on what someone wrote in a book.
I know from reading Keith Lango’s blog that he has pretty similar views on video reference. I’ve gone into this level of detail not to argue for video reference, but to give context to my last post — it’s a potentially useful tool, and there are some ways to make it more useful. I also realize I haven’t used any video reference in months. My guess is that I use it in maybe 5% of my shots. It’s just another tool in my bag of tricks. Sometimes an invaluable tool, sometimes just a different way to do things.
A form of video reference we haven’t talked about is using lipstick cam video of voice actors. For those who don’t know, on most feature productions a small ‘lipstick camera’ is positioned in the recording booth to capture the voice actor’s facial performance as they deliver their lines. This isn’t always done, but when it is, I find it extremely useful. Why ignore all that potential information about what the actual face that’s producing the dialog is really doing?
Again, the goal is NOT to copy the actor. Few voice actors try to give useful physical performances. So it’s very limited as acting reference. Where I find it a gold mine is in watching the actual head accents, the little eye squints and half blinks and full blinks (which tend to be very consistent from reading to reading), and in noting the variety of mouth shapes. It always amazes me when I watch a half a dozen similar reads of the same line by the same actor, and see how distinctly different mouth shapes are used to create the same sound.
Now, how does an animator use reference to ‘get outside’ him or herself, to keep each performance fresh and unique. Hmmm, that’s the $64 question. We can ask the same question whether one is using video reference or not – how do you avoid falling into cliche?
I suspect anyone who gets dependent upon video reference will actually tend to have a harder time getting outside themselves. Personally, imagining a variety of performances and personality types that I can animate is much easier than actually acting out a variety of performances or personalities. I suppose just letting the camera run while you riff on the dialog and situation, trying everything you can think of, might produce an occasional creative breakthrough. But for me, if I’m going to try to get outside myself, I’m going to avoid shooting video reference of myself. Instead, I’ll try to think of interesting personalities I’ve known, or good actors I can study.
Which brings us to the final question. Is it useful for animators to take acting classes or improv classes? Yes, if that’s your cup of tea. And no, if it’s not (which is pretty much what James Baxter said in the latest animation podcast).
I think it’s more useful for any animator to read about, dissect, discuss, analyze, and study good (and bad) acting performances. And to do the same as you observe real life. If you enjoy performing, then I suppose actual classes might be eye opening. But I think something more akin to Don Graham’s Action Analysis classes from the early days of Disney would give more bang for the buck. A group of us started up informal weekly meetings at DreamWorks during Over the Hedge to do just that, and it can be both fun and useful.