the Scarlet Letters

Clarification:

It could be rightly said that all I did in my last project was create an animatic or a pop-through before shooting the real film.

However, I might as well have used that pop-thru with FrameThief's rotoscope function... Which would have allowed me to match up the shot that I'm posing with a pre-determined pose to an unprecedented level of accuracy.

On "Nightmare Before Christmas" they certainly used pop-throughs -- but the whole thing was still being shot on actual film, so it wasn't possible then to do what I'm describing.

It seems to me that being able to match up the frame that you're taking with a pre-shot frame to this level of accuracy... Well, it changes the whole game.

Here is a significant related post I made over on the SMA thread:

I hope not to reduce this discussion down to "straight-ahead vs. pose-to-pose -- which is better?"

In "The Animator's Survival Kit," Richard Williams talks about this supposed dichotomy quite early on, and asserts that the obvious answer is "both." That's my feeling too -- each should be used for best effect, at the appropriate time.

That said, it seems to me that there's a meaningful dialogue to be had about what the strengths of each approach are, so that you can make intelligent decisions about when to use which one.

It feels, relatively speaking, like there's been very little talk on SMA about planning out your animation -- so I want to put the weight of my focus there for the moment.

====================

Strengths of pose-to-pose that I'm noticing:

*\ "Let the pose read"
In "Stop Motion: Passion, Process and Performance," Barry Purves' single strongest piece of advice seems to be "let the pose read." I believe what he's saying is that (a) stopmo animators tend not to hold poses long enough, instead rushing on into the next action; and (b) we don't always choose poses with strong silhouettes.

It seems to me that developing your physical action by creating a series of poses that you transition between is an immediate cure to the bulk of these problems.

*\ Greater control of where the action is going
I've had shots where the puppet's action just sort of gets away from me. I realize that I've accidentally set an arm in motion... And I have to improvise a way to get it to a natural stopping point. When I'm working pose-to-pose, on the other hand, I always have a point A and point B... Random flailing about is less likely to find its way into my performance.

*\ Smaller stretches of animation
When I start a 5 second shot with vague sketches of what poses I want to hit, I feel like I have a full 5 second stretch to animate. However, when I've actually tried out my poses in advance, taken digital photos of them using the actual puppet I'll be working with, and I know how long each pose will last... Then it feels like I have 3 or 4 one-and-a-half-second stretches of animation to shoot, rather than one big one. And that's much less scary.

I haven't applied the pose-to-pose approach to a walk cycle yet -- but the thought of doing so makes walks seem much more doable to me. There are five keyframes in a single step forward, and nine to cycle back to the beginning. If I shoot photos of those poses in advance, and keep them as reference while I'm animating, it feels like hitting the poses will be far, far easier.

*\ Non-linear development of action
I am, at heart, more of a fabricator than an actor. What's appealing to me is coming up with the best poses I can, and then [i]assembling[/i] them into a performance that makes sense. Instead of finding pose A, then finding pose B, then finding pose C... I want to brainstorm poses A,B,C,D,E, and F -- and then figure out which options are worth keeping, and how they can make sense in a string. This is a strategy for developing action that can only be accomplished when you're in a pose-to-pose mindset.

====================

I realize I've neglected to talk about the particular strengths of straight-ahead animation... Forgive me that for the moment.

See, aside from a discussion of "when is it appropriate to use each approach," there's something more that I want to get at here... It feels like I'm groping to articulate something about the very essence of stop motion animation changing.

Uncle Ray didn't have framegrabbers. I've heard some folks talk about how he became a "human framegrabber." I've had moments of experiencing this myself... When I'm really in my "zone," and I feel like I can still see the last frame I shot, as if it were onion-skinned over the current frame.... And I feel like I can see the arcs in space that the nose, the hand, and the foot are going to swing through...

I've heard some folks talk about how a pitfall of animators working with framegrabbers is that we can get into the habit of constantly referring to the computer screen, rather than looking at the puppet itself. (Given the contortions I have to do around my camera and the set sometimes, just to get at the puppet, perhaps that's understandable!)

Frustration.

I can see how I could see how the conversation could now swing from "pose-to-pose vs. straight-ahead" to "framegrabber vs. no framegrabber"... And that's not it, either, really.

Barry has a phrase in his book about the "muscle memories" he's developed from posing puppets. The ways that your hands know how to interact with the armature, where to hold things so the fur won't boil, how to use your own body to change the body of the model...

I find myself wondering: By using a framegrabber and pose-to-pose, am I simply never going to develop those "muscle memories" that Barry describes? In a physical way, am I going to be a different kind of animator than Uncle Ray? Are there ever going to be animators like Purves and Harryhausen again if we aren't forced to adapt ourselves to the animation process in such radical ways?

====================

AH-HA! Here's a bit of what I want to express... Have any of you experienced "Animation Hypnosis?" (I'm just making that word up.)

You get so into what you're doing, it feels like time slows down, and you're experiencing life on the time-scale of your puppet. You walk away from the stage, and it seems like everyone is [i]moving so fast![/i] It's hard to talk -- trying to turn thoughts into words quickly enough so that the folks you're talking to don't get frustrated with you and just walk away.

Well, frame-grabbers, pose-to-pose, non-linear action development and the such are designed to help you AVOID that experience to some extent. They make life easier, so that you can pop into the animation world long enough to shoot a frame, and then get yourself back out into normal time-space again.

THAT'S a change in what it means to be an animator that I find really fascinating!

And yet another noteworthy bit of text that I wrote over at SMA:

"I CAN imagine what you are describing morphing into rotoscoped animation."

Ah! I'm glad you bring up rotoscoping, Ron -- that's exactly where I'm going... But in an unusual way.

Let's call this method pop-through rotoscoping.

==========
THE PERILS OF ROTOSCOPING

From what I've read that's written by cel animators, rotoscoping is fairly reviled. Disney's character Snow White was rotoscoped... Whereas the seven dwarfs weren't -- they're full of "squash and stretch." A lot of folks feel that despite being technically more naturalistic, Snow White doesn't quite look right in the world she inhabits.

As general rule, animation -- cel, stopmo, CG -- is supposed to be stylized, and doesn't look right when it's not. You just can't translate live action directly into animation and have it look... "natural." (How this applies to motion capture is a rich discussion topic for another day!)

What I'm exploring, however, is NOT rotoscoping puppets against live action -- but rather against pre-shot poses of the very same puppet that you're going to be animating with.

(Truth be told, I haven't actually tried rotoscoping with my pop-through footage... I've only used the pop-through as a reference film, off to the side of my framegrabber... But the term "pop-through rotoscoping" best captures the spirit of the endeavor.)

By animating according to the poses laid down in the pop-through, you don't lose stylization. The keyframes -- your extremes -- get to be as extreme as you want. So, this approach to stopmo may ultimately meet technical definitions of rotoscoping, but in effect it's a lot more like key-framing in cel animation.

==========
THE FUN OF IMPROVISATION - WHERE DOES IT FIT IN THE PROCESS?

I'm really interested in what both you and Nick say about getting a "bad squishy feeling in your stomach." Please know that I'm not casting doubt on your methods -- I'm a huge fan of your works! ...What I think I'm hearing is that much of the fun and pleasure comes from the improvisational aspects of animating.

I play piano. I prefer to play improvisationally, rather than reading from sheet music. I think there's potentially a metaphor in there for how different animators find pleasure in what they do. I can appreciate both Jazz and Classical (my own music is neither, really)... But as the person producing live music, there's a distinct difference in terms of where the "play" in playing the music occurs.

When you talk about how planning out the animation all in advance might make animating feel like "grunt work," I ask myself: "so how do 2D cel animators find fun in what they're doing?"

My intuition is that when you create increasingly structured animation, the fun shifts into the details.

A 2D animator who can competently produce a generic walk, can then build upon that to create a unique walk sequence -- squeezing more and more wonderful little moments into their film. When I think about what I like in really good cel animation, it's most often the tiny stuff -- for instance, having a character's eyes go really wide just for a quarter of a second when they hear something that surprises them -- before they find composure and respond with command of the situation. I like seeing a character's thoughts play out across their face, fast and furious.

It seems like master cel animators must be having fun improvising at some level, even when they're working out the inbetweens that connect keyframes... Rhetorically: Could there maybe be a similar sort of fun to be had when you're doing pop-through rotoscoping in stopmo?

==========
"WHAT WOULD ANTHONY SCOTT DO?"

Right now, the "hold-pose-hold" style that I'm working on is most influenced by Robot Chicken... Not because I want to emulate the show's content, but because I'm fascinated by how articulated the poses and transitions are...

Eventually, though, I'd like to have the skills to do the sort of thing that Anthony Scott can pull off. I learned from seeing one of his clips that a walk test should not merely be proof that a puppet can put one foot in front of the other -- a walk should tell you volumes about the personality of the character that you're animating. I just don't know how else to get to the level of detail he demonstrates if I don't embrace a sort of hierarchical approach to animating.

Hierarchical... By which I mean, I build the performance out of parts... Roughing in the key poses that I'm going to hit -- and then progressively layering on top of that the subtleties that make the performance something special.

(And yes, as LIO pointed out, this cel-inspired approach seems to shine through Justin Rasch's clips -- which have also very much been on my mind.)

"You walk awayfrom the stage, and it seems like everyone is moving so fast!"

ha! i never knew you felt that way sometimes. :D