The Drive to Realism — Part 2: CG Takes the Wheel

[1]

Felix the Cat introduced film structure to cartoons and linked story and jokes to movement. © Felix the Cat Prods. [1]

Read Drive to Realism — Part 1 [2]

I remember the moment well: sitting in a small darkened theater about 25 years ago, with a group of fellow NFB animators and some folks from California, all of us staring intently at a monitor on which there appeared an image of an orange. It was a nice looking orange with bright, clear color and a pleasing shape — a good photograph, I thought, of California’s finest. The problem was that the Californians weren’t farmers, they were computer people. And the image we were looking at wasn’t a photograph. And there was, in fact, no orange.

This was my first encounter with CG technology (which had just cracked the code for texture mapping) and as mundane as this presentation sounds today, at the time it was shattering. I can still recall distinctly the sensation of trying to wrap my mind around the fact that no actual orange had participated in the making of this image — that it wasn’t a record of reality but a re-creation.

What was this thing that could produce such a convincing orange imposter? And what did it mean for a realism obsessed field like animation? Only one thing was sure: this was going to change everything. But exactly how, it was too soon to tell. Even today, we don’t really understand the true nature of this beast. And how can we harness what we don’t understand?

So what can we do about this?

First we have to recognize that new technologies are launched on the back of what came before, linked by the characteristics shared between the old and the new.

With one technology layered directly on the other, the early results tend to be pretty awkward: cars which look like horse drawn carriages with a motor where the horse used to be; movies which are aren’t much more than recorded stage plays. Form and function go out of synch as old ways of thinking and new ways of doing collide.

[3]

Pioneer John Lasseter helped to define CG’s aesthetic. Courtesy of Disney. [3]

CG, to make things even more complicated, draws from at least three older technologies: 2D animation, model animation and live action. Like model animation and live action, CG can work in the third dimension. And like 2D, CG is flexible and creates worlds instead of recording them. So far so good, but so what?

Well, the key here is not just what a new technology is, but how we think about it.

Wherever the links allow, we tend to carry over models for both what to do and how to do it from the old technology to the new. So live action and model animation, for example, influence lighting and camera and how characters are constructed.

But the most striking thing is what is happening with movement and performance. Here there is one overriding influence and it’s not the obvious one. You’d think that puppet animation would have had the advantage in the shift to CG but in fact, the winner of this showdown is clearly classical animation.

This goes back to the beginnings of CG as an industry. After an early attempt to get CG animation rolling by training computer-literate techies how to animate, classical animators — led by such pioneers as John Lasseter — began to jump in and define CG’s aesthetic, using their own background as a template.

The end result is that current CG character animation looks, more than anything, like 2D animation with an extra dimension. This standard even shows up in the debate about how to educate CG animators, with many people holding that you have to “paper-train” animators in classical techniques before you put them on the computers.

What’s wrong with that? Well it’s not that the current approach is wrong. But left unchecked, it will stunt CG’s growth.

Hanging on to the old ways too long creates a self limiting view. It’s like looking at a nail that’s been hammered into the wall. Since all you see is a metal disc, you may start to believe that that is all there is to a nail and therefore fail to see it’s most useful and unique characteristics.

So we need to be aware that, when it comes to CG, we are really only looking at the head of the nail. In other words, this is just the beginning.

In order for the new technology to reach its full potential, it has to start exerting its own properties. So instead of looking for what makes the new technology the same as the old one, we have to start looking for what makes it different.

[4]

Photorealism gets creepy. An example is Shrek’s Princess Fiona. Courtesy of DreamWorks Pictures. [4]

Freed from the past, a new technology begins to develop its own syntax, which then connects it to new kinds of content and forms of storytelling. We know that a new technology is really coming into its own is when it starts to do things that couldn’t be done in the old technology

Think of when live action realized it could move the camera and thus ceased to be merely filmed theater. Or when plastic stopped pretending to be wood and started being only plastic (and then fake wood regained its cool with post modern irony).

Early commercial animation, like Mutt and Jeff, was often based on comic strips. As you watched the film, the underlying structure of the strip so obvious you could pretty much see where the layout of each panel would be. Even more importantly, the animation made little or no difference to the story — it was just window dressing. Then along came Felix the Cat, and suddenly there was real film structure. And there were stories and jokes that depended on movement and magic.

So how is CG different from its predecessors? Of course, CG’s most unique quality is its ability to simulate realism (more accurately, photorealism) to a degree unimaginable by even the greatest masters of the old technologies. At first glance, this seems like a dream come true — the finale to a great quest, but is it really that simple?

In fact, once we start to examine it, we discover that this one attribute brings up many issues — some more problematic than others.

Right off the top, it’s odd and thrilling, but also unsettling to suddenly find that one’s obsession has been matched with a new technology that holds the promise of fulfillment.

The drive to realism in 2D has been with us for decades, but the old tools had those natural built in limitations which contained the obsession, disciplined it. Animators had to be content within the boundaries which kept the drive in a lower gear. Meanwhile, their excess energy could be channeled into figuring out inventive ways to create an illusion of a third dimension, of reality.

[5]

CG raises the dilemma of photorealism: something recorded vs. something invented. Above are two Chris Landreth works, Bingo (left) and Ryan (courtesy of Copper Heart Ent. and the National Film Board of Canada). [5]

But then, the animators get handed this new tool, one that dangles the promise of a manufactured photorealism which one day will be indistinguishable from the real thing. As it turns out, the promise has dangled there for years, the technology developing much more slowly than predicted but still making regular enticing advances: talk about a tease…

Of course, animation has always been an obsessive art form, but before CG we had control of the tools: how far you could take the illusion was between you and your pencil. No one was stuck waiting while the engineers worked toward perfecting the pencil, an excruciating process with new features added yearly. No one longed for the day the ultimate pencil — the one that allowed you to really draw to your full potential — would finally arrived on the market.

[6]

Chris Landreth might have the solution. Photo credit: Shira Avni. [6]

I’ve always thought that — while the pencil is clearly a tool — a faithful, already perfect servant — the computer is more like taking on an idiot savant partner. You have to work with the computer, which is great at some things and dreadful at others.

Adding to the frustration, CG isn’t fully formed by a long shot and its development is out of our hands and maybe we would be smart to wash our hands of the whole affair… But then there is its potential to fulfill our most obsessive desires — an altogether dangerous combination. This doesn’t mean we can’t get on top of this technology but awareness is essential because when we are enthralled, we are vulnerable.

This enthrallment, particularly the feeling of being so close to the goal after all these years without quite being there, may be a bigger factor than you’d think. It might explain the compulsion to go photoreal before the tools are really up to the job: the miscalculated use in Toy Story of full frontal realism in the human characters even though they contrasted badly with the total believability of the toys; the creepy near-realism of the human characters in Shrek are just two examples of less than clear judgment at work.

And yet, at the same time, the goal is still so far away: eight years after Toy Story, explosions are still far easier to achieve than getting a hand to grab a shirt. This is enough to make anyone crazy.

But it doesn’t end here: there are still more complicating factors.

By taking the illusion so much further, CG throws a whole new spotlight on the dilemma of animated realism. The strength of any medium lies in its ability to reveal truth. An essential element of that ability grows from the underlying relationship between the medium’s inherent properties and its content. But how can we achieve that relationship with a technology whose raison d’etre is so closely tied to the imitation of another medium?

As long as CG is used to achieve photorealism, it will have to deal with that built in dichotomy — that there is a gap between what it appears to be (something recorded) and what it is (something invented). What we have to be aware of here is that the gap itself communicates a deep message that has to be accounted for in our storytelling.

There is also the lag in achieving photorealism between human characters and just about everything else. Part of that lag is undoubtedly technological but another part is in the thinking.

The original source of the blockage comes from that continued attachment to classical approaches. But the longer it goes on, the more complicated it becomes. Client expectation, practices that were set in place when the technology was more limited and just sheer complacency now all play a role. Meanwhile, the films suffer from an inconsistency in design and movement that makes the communication gap even wider.

This only confirms the importance of moving to the next stage before the first one becomes an orthodoxy.

But don’t economics prove that CG should just carry on with the approach established in Finding Nemo [7]: high realism in the settings, non humans as the stars and a cartoony approach for the humans? A conservative route, yes, but a popular one.

But that presents yet another brand new dilemma. If we don’t address the issues CG raises about photorealism, particularly around the issue of human characters, there is a real risk that we will lose control over the whole thing. Never has an animation technology been so enticing or so threatening to the live-action industry. One thing we can bet on is that they will want to control and exploit this technology. If we don’t set the agenda, they will.

And have we really come all this way just so that live-action producers can extend the careers of aging action heroes, create new characters that seem totally real but require no star treatment and make films that can carry on production rain or shine? I don’t think so.

What a can of worms we’ve opened here. But don’t despair, there are solutions. And yes, they do involve Chris Landreth [8] — I haven’t forgotten, I promise.

Only thing is you’re going to have to wait till next time to find out what they are. Yes, I know it was supposed to be a two-part series and officially it still is, so the last installment will be the first part of the new series. Given that this is, after all, only the most important issue facing animation today, I think we need to give it just as much time as it needs to be properly explored.

Meantime, meditate on the special traits of CG that we’ve uncovered so far and try to see which other ones you can uncover. See you next time.

Ellen Besen [9] studied animation at Sheridan in the early 1970s. Since then she has directed award-winning films both independently and for the NFB, worked as a film programmer and journalist, taught storytelling and animation filmmaking at Sheridan and given story workshops at many institutions and festivals, including the Ottawa International Animation Festival. She is the director of The Zachary Schwartz Institute for Animation Filmmaking, an online school that specializes in storytelling and writing for animation.