COMPLETE ANIMATION

WHAT IS AN ANIMATION PROJECT?

Stephan Vladimir Bugaj

Full Animation versus Visual Effects

In contemporary filmmaking, the line between an animation project and a visual effects project has become blurred. At one time distinctions could more reasonably be made. Visual effects could be described as animation done in support of live-action plates, whereas full animation could be described as an artificial, created world; or, put another way, visual effects is primarily about environments and noncharacter dynamics, whereas full animation is primarily about character dynamics. But consider, for example, the Star Wars prequels. Those films have animated characters in nearly all of the shots and essentially all of the environments and noncharacter dynamics are also created digitally. A film like James Cameron’s Avatar (2009) takes this idea even farther. These films are basically photorealistic animated films, with live-action plates created in support of the animation. However, a simple definition falls short of the meaning:

•   A full animation project can simply be defined as a project that has absolutely no live-action component whatsoever.

This distinction separates the output of Pixar, Dreamworks, and similar animation companies from what ILM, Weta, and similar visual effects facilities generally produce. But this is a formal distinction more than a functional one. Blue Sky’s film Robots (2005) is 100% CG, whereas Pixar’s WALL-E (2008) is 99% CG (there are a few live-action shots), but both are universally regarded as animated features. James Cameron’s Avatar (2009) is perhaps 90% CG and Michael Bay’s Transformers (2007) is 75% CG (approximately). Both are considered live-action (albeit VFX-heavy) features. But 75% CG versus 100% CG is not a very substantial difference in terms of CG footage on screen when it comes to doing the actual production work, considering that a full CG feature is often shorter than a visual effects-driven feature, such as the 98-minute length of WALL-E (2008) compared to the 144-minute runtime for Transformers (2007). Of course, there are differences between the two. In cases where the CG work amounts to less than 100% of the project, a live-action set still exists, but the post-production difference for the animators, TDs, editors, compositors, etc. isn’t so great.

As the distinction becomes blurred, an effort must be made to define an animation project’s functionality. One such attempt is:

•   An animation project is one in which stylization, not realism, is the foremost visual concern.

This distinction has two problems. One is that it would incorrectly categorize a photoreal animated film like Final Fantasy: The Spirits Within (2001) as a live-action or hybrid project, and it ignores the fact that some live-action films have stylization as their foremost visual concern (the Matrix films, for example). Another possible candidate for a workable definition is:

•   An animation project is one in which the animated footage is the primary focus, and any live-action footage (if any) is subordinate to it artistically.

WALL-E (2008) is a clear example of a film where this distinction works. Who Framed Roger Rabbit (1988) is a little less clear. And the Star Wars prequels are less clear still. However, a good case could be made that all of those are animated projects with live-action plates created in support of a synthetic world and the animated characters that populate it. However, the Star Wars prequels may not fit the definition—because it is unclear whether the live-action footage was truly subordinate or not. Refining the above definition to try to avoid defining exactly what the filmmakers’ intentions in this area really were leads to a definition that seems to make the most sense as a functional definition:

•   An animation project is one that uses an animation pipeline rather than a visual effects pipeline.

Even this distinction is becoming blurred as the needs of the two pipelines converge. But to make a distinction within the current state of the art, one needs to roughly define what an animation pipeline is and after that refine and elucidate on that definition. This is not a difference one can see just by watching the film, which is the point—unless one accepts stylization as the differentiator and, in turn, accepts some full-animation projects, such as Beowulf (2007) or Final Fantasy: The Spirits Within (2001), as being virtual live action. The primary difference is one that can’t necessarily be seen. To establish this pipeline-based differentiation, consider this definition of the distinction between live-action and animation pipelines:

•   An animation pipeline is one in which shots are defined in the animation system, and live-action footage (if any) is integrated into the animation system footage either using 2D compositing or as animated textures on a 3D scene object.

•   A visual effects pipeline is one in which shots are defined by the digitized live-action footage, and animation is integrated into said footage.

In a visual effects pipeline, the most important thing procedurally is the live-action footage. Each shot starts with the live-action framing and movement, and the pipeline is built on the necessity of conforming the visual effects to this source live-action footage. With this definition it becomes more about how the film was actually made, rather than what the intention was. If the starting point for all shots was the live-action footage, it is primarily a live-action project even if the results are highly stylized such as Sin City (2005) or 300 (2006). If the footage (and camera definitions) originates in the animation system, it is primarily an animation project. So since the Star Wars prequels could have been either one, the pipeline approach used defines whether it was an animated project or a live-action project.

An alternate distinction could be that animated films are made up primarily of animated footage, and live-action films are made up primarily of photographic plates. This distinction becomes muddied by films like Dinosaurs (1991), which has all-CG characters composited into live-action backgrounds, and Spy Kids 3D: Game Over (2003), which is all live-action characters composited into mostly CG backgrounds. The distinction can be refined further to consider whether the characters are primarily CG or live action—but then Star Wars Episodes I-III (1999, 2002, 2005) reveal the problem with that definition: Those films used extensive CG backgrounds and included large numbers of CG characters interacting with live-action characters.

Taking the “Where does the pipeline originate?” definition to its logical conclusion presents problems: Is a rotoscoped film like A Scanner Darkly (2006) live action? If camera moves originate in a previsualization system, does that make the film animated? The latter is clearly not the case. Even if the previs camera setups are fed into a camera control system, the finishing pipeline would still be live-action plate based. The former is a more interesting film theory question, but in terms of a practical distinction a film like A Scanner Darkly (2006) is a live-action film. For the purposes of pipeline distinction, this is because the on-set, photographic process was primary in plate acquisition. From a film theoretic perspective, it is also because live-action performance was primary over keyframe animated or puppeteered performances.

This distinction is not very helpful for film theorists or awards voters, but for visual effects and animation professionals it is one of the only distinctions left that makes much sense. And even that is a fading distinction. Advances in CG technology and the growing popularity of hyper-unrealistic spectacle films draw the two previously distinct disciplines closer and closer together every day. As organizations like VES and SIGGRAPH, and artists moving between companies, continue to spread ideas around visual effects, animation, and games facilities, even the pipeline distinctions will start to blur. Before too long, there will be no de facto visual or procedural difference whatsoever between a full animation project and a mostly CG project with live-action plates. Because of this narrowing gap, understanding animation projects has already become important for essentially anyone in this business. Tools, techniques, and pipelines that were previously restricted to the specialized domain of full animation have been moving into the visual effects world for some time now.

DIFFERENCE BETWEEN VISUAL EFFECTS AND ANIMATION

Rob Bredow

Visual effects and animation have a lot of similarities and some significant differences. Visual effects are only a part of a film and are used to augment the live action. Animation, however, is its own world. The animation studio has complete control over everything in the scene and creates the entire movie from scratch.

Visual effects are typically added to live-action photography in post-production and are integrated with the live action. Even shots that are 100% virtual in a visual effects show are typically locked to the design and execution of the surrounding live action. Additionally, some of the animation within visual effects is not directly viewed (roto-motion for matching an actor or object for interaction, casting shadows, etc.). Also, some animation within visual effects is necessary to match live-action motion to animation such as a stunt double who switches during the shot from a stuntperson to an animated double.

In animation, it’s all about creating a virtual world in which the story can be told. With no live-action elements to photograph, the art direction is not limited to any real-world elements. The camera can be dictated by the characters’ performance or virtual set blocking requirements since there is no live-action plate to match. In addition, nothing is ever gotten for free. Every prop, set, sky character, and lighting element needs to be designed and built in the style of the film and assembled correctly for every shot.

Production Pipelines

A good place to start breaking down the differences between live-action visual effects and animation is by examining the respective shot post-production pipelines.

Although Figure 9.1 is a simplification of film production environments, it provides a useful reference. With the exception of the layout department, the overlap between live-action visual effects and fully computer-generated animation is significant. In particular, as live-action visual effects become more prevalent throughout film productions and include regularly replacing sets and characters for an entire film, the distinction between the two types of projects is blurred even further.

Figure 9.1 A simplified production pipeline comparing animation and visual effects productions. (Image courtesy of Rob Bredow.)

This section compares the differences and similarities between the two classes of work, starting from the beginning production stages and working through final delivery.

Production

One of the early production differences between visual effects and animation is the approach to budgeting and bidding the project. In both cases, there is usually a set budget to be spent and always a desire to achieve as much volume of work and visual complexity as possible.

In the case of visual effects, this is usually accomplished during post-production as shots are identified, bid, and awarded to the visual effects facility. This process often involves multiple production facilities competing for the work, and costs are controlled by finding the most efficient facilities and minimizing the complexity requirements on a shot-by-shot basis.

In animation, the entire production is commonly awarded to a single production facility (which may or may not be owned by the production company) and an overall budget is set from the start for accomplishing all of the computer graphics animation for the film. The delivery specifications usually include the length of the film and the number of characters, sets, and other metrics that are used to broadly measure the complexity of the movie. From that point on, the production moves forward with careful monitoring of the complexity of the film both in terms of the characters, sets, and props and the complexity of the individual shots to ensure it fits into the original bucket. Overages are usually limited to the addition of major components or schedule delays.

The production of a CG feature can naturally be broken down into two phases:

1.  Pre-production: building of all the assets

2.  Production: creating the shots

Pre-Production

The building of all of the assets is a substantial part of the creation for both visual effects work and a CG feature film. The pre-production step includes designing, modeling, texturing, materials, and lighting as well as building the animation rigs for the characters and setting the style of the animation for the film. Depending on complexity, pre-production accounts for 20% to 35% of an animated film’s budget and can use approximately half of the production schedule.

The design is driven by the art department whose members primarily design the characters, environments, and props with 2D drawings and sketches for the quickest iterations. Hundreds of designs are attempted while defining the look of the picture and creating all of the assets for the film. Some approved designs will be simple sketches from a single angle, while others may require precise drafting of all of the spatial details that will be used to inform the later departments.

Often the process of designing key characters and environments is continued into the 3D modeling phase where the creative team will explore the 3D interpretation of the model. Early in the process, it’s common to encounter significant setbacks during the translation of the 2D drawings to the 3D world. The setbacks may require design changes to satisfy the creative requirements of the show.

Also, the production models of characters and environments have significant technical requirements:

1.  being modeled in a median position or T-pose so that the character can be rigged efficiently,

2.  certain topologies or layouts to meet rigging, texturing, and rendering needs,

3.  specific resolution in various areas of deformation to allow for posing, and

4.  the model may need to be generated at multiple levels of detail if hundreds of copies of the model will be seen in the same shot.

For this reason, if the exploration of the 3D model has a significant creative aspect, it’s common to create a scratch 3D model that doesn’t meet the technical requirements for production but can be used to explore the style of the film. Once the scratch model has been approved, the creation of the final model is primarily concerned with the technical exercise of adjusting the model to meet the production requirements.

Once the model has been approved, the process of adding the textures and materials can begin in tandem with the rigging of the character for animation.

For textures and materials, there are many similarities between live-action visual effects and animation. The basic process of laying out textures to paint the objects and assigning the correct materials for the various parts of the models is the same. The differences are primarily dictated by the style of the animated film. While some animation will adhere to photorealistic principles in terms of the behavior of light, reflection, and other material characteristics, an animated film will often make intentional variations away from pure realism for creative impact. Translating the artistic vision from the 2D artwork to the 3D world in a way that achieves the artistic goals, which maintains consistency and lighting control for later production, is collaboration between the art department and the computer graphics supervisor. It is very common during the early phase of pre-production to view characters, sets, and props on turntables in various lighting conditions. As time progresses, it’s common to create test shots that combine environments, characters, and props under various lights for the sole purpose of testing the artistic and technical assumptions.

Simultaneously, the character riggers work with the lead animators to build the characters with their control systems, skin deformations, and facial rigs. Again, in this area many of the tools are very similar to visual effects production in terms of their primary functionality. And once again, the differences are primarily driven by the stylistic needs of the film. For highly expressive cartoony animation styles, great care is required to allow extreme ranges for the facial and body animations, which might include the ability to pull the corners of the mouth all the way up around the ears or bend an arm like a noodle to achieve a pose. These artistic choices are generally quite simple to draw in 2D animation, but in 3D animation they require careful attention to ensure that the animator is provided with the correct set of controls, that the model has the correct flexibility, and that the textures do not overly stretch or become otherwise compromised in the deformation process.

Once the preliminary rigs are in place, the character animation leads begin their testing, putting the characters through their paces. This accomplishes two goals:

1.  It tests the range of motion provided by the riggers for the characters to ensure it’s sufficient for the character’s role in the movie.

2.  It begins to explore the animation style for the character and get early feedback from the directors on the performance.

The collaboration between the character riggers and the animation testing continues throughout the pre-production process, while the entire cast of characters is built up to meet the standards of the production.

Production

Pre-production is rarely completed, but there is a slow transition as the first sequences are turned over into production and the members of the pre-production teams begin to assume their new production roles on shots. It’s common that new environments, props, and even characters are created as sequences are entering production so the two production phases may overlap very significantly.

Layout

The job of the layout department is to translate the storyboards into the 3D animation world. This consists of establishing the position of the camera along with its lens, the basic blocking of the characters, and the dressing of the sets. The layout artists perform much the same role as the camera operator on a live-action set but do not generally include lighting. (Wall-E, released in 2008, was a notable exception where Pixar first experimented with more sophisticated lighting in layout to help set the mood of the shots early in production.) When completed, the layout department delivers a 3D scene for each shot that includes the characters, the set, and the camera ready for animation.

The final camera work is approved in a stage called final layout after the animation is completed. It’s also common for there to be interaction between animation and the layout teams for shots with dynamic cameras because each department needs to react to the other’s work.

In most visual effects work, the closest analogy to the animation layout department is matchmove. In the matchmove department, the live-action photography is precisely matched with a virtual camera and set for every frame of the shot. This generates the camera and environment in which the visual effects work can take place. It is common practice in visual effects work to modify the live-action photography to accommodate the creative requirements of the scene—a task handled by matchmovers, plate-mungers, or the animation department depending on the production. Additionally, some visual effects facilities have embraced the idea of a hybrid layout/matchmove department for complicated CG environments for live-action films.

Animation

The differences in the animation department between visual effects work and animation depend almost entirely on the subject matter and the style of the animated film. Some animated films desire a high level of realism, and for those the same animation principles that guide live-action visual effects work are applicable to animation. Animated films can rely on the same technology options as well, including rigging techniques, deformation methods, use or nonuse of motion capture or video reference, and many other similarities.

However, most often the desire of the filmmakers is to establish a unique look and style for the animation of the movie. This often requires the technologies and animators to adopt new techniques to either emulate a particular look from a 2D animation technique or simply create something new that is well suited to the 3D animation world.

Animation teams for animated films are larger than the average visual effects film and often include dozens of experienced character animators. Animation quotas tend to be based on feet of footage per week, whereas visual effects animation quotas are usually tracked by shot count.

Cloth and Hair

The cloth and hair departments have more similarities than differences when comparing visual effects and animation. Both types of work require refined techniques for grooming, costuming, and manipulating the simulations to achieve the desired look for the picture.

Since animated films require every costume and hair simulation to be animated for every character in every shot of the film, complexity of cloth and hair can have a significant impact on the cost of the film. Simplification strategies include using shorter hairstyles, tighter fitting clothes that can be attached directly to the skin, and other techniques that avoid having to run simulations for every element in every shot. Modern simulation techniques, however, are better at solving clothing and hair issues and require less frequent artist intervention, which can make previously costly costumes and hairstyles more efficient.

Effects Animation

Effects animation is another discipline that is heavily influenced by style when comparing visual effects and animation. Animated films can require completely photorealistic effects animation to complement a detailed style that can be leveraged off traditional visual effects animation techniques. However, animated films can also provide their own set of nonphotorealistic choices for anything from the patterns created in breaking glass all the way to smoke billowing off a fire.

One of the biggest challenges in effects animation for an animated feature is the nonphysical dynamics requested of simulations. Oftentimes, stylized character animation will create movements that move extremely quickly from frame to frame and have a tendency to break systems that expect movement to be based loosely in reality. In these cases, the motion needs to be isolated to get the desired look from an otherwise physical simulation.

In general, a talented effects artist will use similar techniques to control a simulation or create an effect whether it’s for a live-action film or an animated feature.

Lighting and Compositing

For lighting and compositing tasks, there are some significant differences. In addition to the obvious requirements that visual effects work must approach photorealism and animation can be created in any style imaginable, there are other important distinctions.

When dealing with animation, the entire scene is lit using 3D tools and there is no requirement to match the lighting to a live-action plate. Because of this flexibility, most studios rely on an art department consisting of one to three painters during production to establish the lighting style of the sequences throughout the movie with loosely painted thumbnails for each important scene. These thumbnail references (referred to as lighting keys) are then referenced by the lead lighters for the sequence and help provide a consistent visual style for the movie.

In addition, it’s much easier to leverage from key lighting in an animated movie. Key lighting is the process of lighting several key shots in a sequence to establish both the creative look and the technical procedures used to achieve that look. The most experienced lighters are generally given the key lighting responsibilities and help establish the look and workflow for the other artists on the film. Once the key lighting is established, the rest of the lighting team can reference those lighting rigs when creating the rest of the shots. Lighting adjustments are made for each shot just as in live-action photography, but the key lighting rig gives a starting point and encourages consistency and efficiency.

In visual effects work, key lighting can still be used for sequences with a lot of CG elements, but since the requirement is the match against live-action photography, there are more shot-by-shot adjustments so the payoff for key lighting work is reduced.

Compositing for animation is generally a more straightforward technique than for visual effects work. Generally speaking, all elements are CG rendered and therefore come with clean mattes and possibly even extra control passes to adjust the lighting during the compositing phase. Unlike live-action photography where these elements are filmed with strict physical constraints and can never be perfectly registered, in animation the various elements and mattes can fit together perfectly. In addition, complicated techniques for pulling mattes from blue or green screens and other live-action techniques are simply not required.

However, compositing serves a very useful purpose in animation work. Some facilities prefer to use a composite-heavy workflow where many individual passes are rendered out per character and then the final look is dialed in during the compositing phase. This has the advantage of avoiding long rerendering times and providing fast feedback to the artist who balances the various lights and elements to create the final look. One disadvantage of this technique is that these compositing scripts generally require a lot of adjustment from one shot to another so shot-to-shot efficiencies may be reduced. In addition, some techniques that are easy to use in a compositing package can create looks that don’t follow any photographic rules, which can result in visual artifacts if not used carefully.

Among artists and supervisors, there are strong opinions about whether it’s better to dial in the final look in the 3D lighting package or during the 2D compositing phase, and both techniques can create top-quality imagery. Generally speaking, the longer the render times and the more limited the access to computer processing power, the more appealing the 2D-centric workflow. On the flip side, if the film has a style that can be achieved mostly in 3D lighting and there is sufficient rendering power to iterate on the lighting side, the improved artist efficiency and consistency between shots can be a very significant factor.

In addition to dialing in the look of the lighting, compositing is always required when special-purpose renderers are used to achieve various effects. For example, some looks require volumetric smoke elements to be rendered by stand-alone renderers that are tuned to create images efficiently. In those cases, the compositor will combine the smoke elements with the other rendered elements in the compositing package in much the same way they would approach a live-action visual effects composite. Once again, the advantage of the all-CG elements is in the perfect registration of the matte elements, which gives the compositor the ability to make a clean composite in less time.

A SURVEY AND HISTORY OF ANIMATION TECHNIQUES

Frank Gladstone

If, by definition, animation is instilling motion where none exists, then motion picture animation is just a bit over 100 years old. The first animated film to which a date can be fixed is J. Stewart Blackton’s simple but very clever “trickfilm,” Humorous Phases of Funny Faces (1906).

If, however, one’s definition of animation is instilling personality where none exists, the starting point might be Winsor McCay’s Gertie the Dinosaur (1914). That film was designed to be part of McCay’s vaudeville act¹ and featured Gertie, an animated dinosaur who expressed a range of human emotions when prompted by her real-life creator, McCay, who appeared in person, syncing his actions and comments with the character (foreshadowing the live-action/animation combinations that are so familiar today). Gertie was, by turns, shy, mischievous, scared, sad, and content—personality traits now referred to as character animation, creating, as pioneering Disney animators Frank Thomas and Ollie Johnston so beautifully put it, “the illusion of life.”

This serves to focus on the two very basic and interrelated definitions of animation, which are the core of the art form: animation as motion and, more importantly, animation as emotion.

All forms of animation rely on preparing a series of still images in a sequential order that, when displayed in a continuous manner at a predetermined speed, gives the viewer the sense that the images are moving of their own volition. Today, there are three basic approaches to achieving animation: traditional or 2D animation, stop-motion animation (also called stop-frame animation), and CG (computer-generated) or 3D animation. Within each of these disciplines are subdisciplines and crossovers.

Traditional Animation

Traditional animation is sometimes referred to as 2D animation or cel animation, named for the celluloid sheets or, cels, on which the individual animation elements were transferred before being photographed onto film. Generally, traditional animation is drawn, with characters, props, effects, and environments first committed to paper. Next, all objects that will appear to move in the final film are traced to cels, painted with opaque colors on the reverse side of the inked line², and assembled in the sequential order that was determined during the animation process. Each setup is a layered sandwich of cels over background art, often with foreground art as well, positioned beneath the animation camera and photographed one frame at a time. This setup is then replaced by the next one, and the next, and so on, until the entire scene is completed.

Cel animation can be a tedious process, because each step has to be planned, timed, and animated, with a drawing (and often several drawings) created for every frame of film. Because of this, from the earliest days, storyboards have been employed to help plan out the story and pacing of each sequence before actual animation occurs. (In fact, storyboards were first developed for animation.) From the storyboard, a story reel, in early days referred to as a Leica³ reel, is created by photographing the storyboard panels and editing them according to a rough estimate of scene length, often employing a temporary soundtrack as well. Story reels, sometimes called animatics, are designed to help time the sequences and test out the film’s cinematic nature as well as set the tone for the voice acting and animated performances to come. Story reels and the information they provide are the precursor to the previsualization techniques used today in visual effects and CG animation. Similar to previs (in fact, previsualization footage is frequently referred to as an animatic), these reels are the guide for the creation of layouts: backgrounds, foregrounds, camera, character and object placement, and the paths of action within each scene.

Eventually, from the story reels and the voice tracks, exposure sheets⁴ are developed. These are usually broken down by scene and give the length of the individual shot. The dialogue is carefully and extensively noted as it is translated into units of linguistic sound (i.e., phonemes), so that the animator can provide lip-sync for the characters. Other specific sound cues, camera movements, and action directions may be entered as well. From this information the animator knows how many frames he or she will have to work with and what will happen in the scene. As animation is completed, it is noted on the exposure sheets.

Animation is either drawn “straight ahead” or by using “key poses.” Straight ahead means that the animator begins at the start of an action and then proceeds directly to the next drawing and each succeeding one in sequential order. Key posing means the animator determines the main segments, or top points of an action, and draws these points as keys. He or she will also determine the number of drawings between each key and how close or far apart those drawings will be in relation to one another. Usually, the animator will provide a chart to show the number and spacing of the drawings between the keys. The most midpositioned drawing is called the breakdown and the ones between the breakdown and keys are called the in-betweens. Assistant animators will tie down and complete the animator’s rough keys, making sure the drawings are consistent with the character (i.e., on model) and that all details are accounted for. Breakdown artists and in-betweeners will fill in and finish the additional drawings. This operation is commonly referred to as cleanup.

Effects animation depicts things that are not characters: objects, props, natural phenomena, and the like. Atmospheric effects are often stylized, providing a kind of enhanced reality that is more in relationship to the tone and subtext of the story. Effects are often animated straight ahead.

Most feature quality animation is designed to work on ones or twos. This refers to how many frames are photographed of each sequential setup prior to being replaced with the next setup in the sequence. Because motion picture film is projected at 24 fps, ones require 24 separate drawings to achieve 1 second of movement per character or object, while twos, meaning each cel or frame is photographed twice, require 12 drawings per second. Of course, animation cels can be held if a character or an object is at rest, cutting down on the number of drawings that must be animated and prepared per second of film time.

Two terms are often misused in describing animation: full animation and limited animation. Full animation means that movements and emotions are thoroughly articulated and rendered. Full animation is almost always done on ones or twos (except for held drawings) and often has more than one character moving in the same shot. In full animation, a preponderance of complicated effects is usually at work and nearly all action happens on screen. Limited animation means that movement and emotions are stylized, and oftentimes only one element, or even a part of one element, is moving on screen. Effects are formulaic and much of the action happens off screen.

Limited animation began in the very early days and then took a back seat as people became used to higher theatrical quality. It was reinvented and refined when animation began to be produced specifically for television, accommodating that medium’s need for higher volume and lower production costs.

Additionally, while full animation takes little heed of the number of drawings, usually producing as many as deemed necessary to create the desired effect or mood, limited animation does not have that luxury. Instead, it must rely on clever design, isolated facial elements and body parts (so that some movement can happen while other elements hold still), emphasis on layouts and backgrounds, abbreviated and formulaic movements, repeated action (cycles), re-used animation, cel level manipulation, and camera tricks. All of these devices are employed, in any number of combinations, in order to cut down on artwork.

This does not mean that all full animation is good and all limited animation is bad. In fact, sometimes full animation can be overarticulated, or spongy, or be so naturalistic that it almost appears creepy. Conversely, in the hands of a good designer and with careful cinematics, limited animation can be quite compelling and often its uncomplicated approach enhances a story rather than detracting from it. In short, the old adage “less is more” also can apply to animation. This is nowhere more apparent than in the often subtle and beautiful limited animation techniques found in much of anime, the influential Japanese-inspired animation methodology that has become popular with audiences worldwide.

Confusion arises from these two terms when incorrect interpretation is used, especially when the term full animation is used to mean that the shot, scene, or sequence is made up entirely of animated elements, and limited animation is used to mean a shot, scene, or sequence which has some animated and some live-action elements. Although this may accurately describe the construction of a scene, it mixes definitions that have been used for decades to describe the style of animation.

For example, when Jessica Rabbit comes out to do her opening song in Who Framed Roger Rabbit (1988), all of the shots are made up of live-action and animated elements, but the actual animation of Jessica and the other cartoon characters is heavily articulated, full animation. Conversely, when Fred Flintstone is talking to Wilma in any episode of The Flintstones (1960-1966), every component of each shot is made up entirely of drawn elements but the animation style is decidedly limited.

Remember, in animation terms, full means more articulated and complex animation and limited means more stylized and less intricate animation. The terms do not refer specifically to the combining of live-action and animated elements.

Finally, one more term first applied to traditional animation more than 90 years ago that is still in use today is rotoscoping. Invented by the Fleischer Brothers around 1917, the rotoscope was originally designed to allow an animator to trace live-action footage, frame by frame, onto paper, as a method of animating characters. Thus, rotoscoping is the forerunner of today’s motion capture technology⁵. The rotoscope was also used as a reference guide for placing animated characters or objects into an otherwise live-action scene and the term is still used to describe the techniques needed for placement of animated elements into live environments (plates).

Stop-Motion

The second approach to animation is stop-motion. This technique is probably as old as traditional drawn animation⁶ and is most definitely the first attempt to give a 3D look to animated film. That is because stop-motion, by its very nature, is 3D. It is achieved by photographing actual objects one frame at a time and moving the objects in a sequential pattern, between frames, so that when projected at a continuous and predetermined speed, the objects appear to move.

Early on, these stop-motion objects were simply items from everyday life, but soon more complex characters evolved, allowing for personality animation. Movement of these characters was achieved by either manipulation of articulated puppets or by object replacement. Many times, animation was achieved by a combination of both methods.

Articulation is best described as an object, usually a figurine, that possesses some sort of internal structure that facilitates the object being placed in a fixed and stable position, in which it will remain until photographed, and then adjusted into a sequential fixed position for the next photograph and so on until the movement is completed.

Replacement is the term used to describe the process of replacing one object with another between frames in order to achieve movement. For example, a replacement model would have perhaps 18 or 24 separate sets of legs, each in a graduated walking position. Each set would be replaced in turn, between frames, to provide the walking effect. (Alternatively, an articulated model would have just one set of flexible legs that could be positioned in all of the various motions to create a walking effect.) Although extremely time consuming to create, replacement animation allows for easily repeatable and reusable actions and can be more efficient for some kinds of movement during the actual animation process.

A good example of mixed use of articulated and replacement techniques would be the Jack Skellington character from The Nightmare Before Christmas (1993). Jack’s body was completely articulated, using ball-and-socket and lever-style joints to achieve all of the complicated body movement and choreography the screenplay calls for. Jack’s lip-sync and expressions, however, were done with a wide range of replacement heads, notched to fit consistently onto the body and sculpted with mouths and other facial features in various positions so that the animators, by switching out a predetermined continuum of heads between frames, could make Jack both talk and emote appropriately.

The technique of stop-motion demands that all animation be done straight ahead. There is no practical way to position a stop-motion model or puppet for key frames and then go back to put in the in-between motions. In fact, before the days of video playback, there was no way to even preview what a stop-motion animator had shot. Animators simply had to continue their work straight ahead, assisted only by small appliances on the set—called surface gauges—that acted as simple reference marks to indicate where their models had been in the previous frame. Today, using frame-by-frame video playback, stop-motion animators can look at the frames they have completed as they set up each new position, which is a great help in keeping the animation consistent and error free.

Stop-motion environments are basically miniature practical sets, often built in forced perspective if added depth is needed. (Often, characters will be scaled in forced perspective as well.) Sets are lit in the same way a live-action set would be, with care taken to keep the light temperatures low, because the lights will be on not for minutes but hours while the animation is achieved. Additionally, sets usually have some way to tie down or fix the feet or base of the animated puppets, keeping them stable when in their various, often off-balanced positions. Another improvement resulting from today’s technology allows animators to use external rigs that are visible to the camera to balance or support characters. These rigs can then be removed or painted out of the shot using CG image repair technologies (Photoshop, After Effects, etc.).

Traditionally, moving camera shots in stop-motion required complicated rail systems and human-driven, very precise movement along that rail. Today, computer-controlled camera cranes (motion control) can make smoother and repeated setups possible. Currently, most stop-motion is photographed using lightweight digital still cameras and lenses that send the images directly to computers for storage, so camera moves have become even more flexible and practical.

Since the 1920s, stop-motion has also been used to add fantastic characters into live-action films. This was achieved by animating the characters so that their motions would be in sync when optically matched with previously photographed live-action plates. The most famous of these films is undoubtedly Willis O’Brien’s masterpiece, King Kong (1933).

O’Brien’s protégé, Ray Harryhausen, took his mentor’s techniques a step farther, both by giving his animated characters more subtle emotional depth and by perfecting a beamsplitting, rear screen/front screen projection technique, making it possible to join his complex stop-motion creatures more convincingly with live-action actors and backgrounds, resulting in such memorable work as the skeleton fight in Jason and the Argonauts (1963).

Today, these kinds of “monster in the real world” visuals are achieved exclusively with CG-animated imagery, because stop-motion’s relatively long production schedules and specifically its dream-like movements, which always contrasted somewhat with the live-action plates, are not able to satisfy modern tastes for reality.⁷

Computer Graphic Technology

The third common approach to animation is the use of computer graphic technology. This method of making animated images has more in common with its traditional 2D and stop-motion animation forebears than might seem obvious at first glance.

First, character animation in CG can be compared to the process of moving a virtual stop-motion puppet in that the 3D model has a basic, though usually much more heavily articulated, virtual armature, controlled by the animator who moves the character into key positions—generally referred to as key frames. Similar to traditional 2D work, the animator can make decisions about the number, speed, and placement of in-between positions, which can then be generated by the computer.

In fact, most of the activities involved in CG animation production can be directly related to similar activities on the more traditional animation production pipeline. For instance, once story and production designs are determined—much the same process for all types of animated films—CG proceeds with the modeling of characters and rigging of their armatures with appropriate movable parts and controls. These activities are directly correlated to developing model sheets and drawing formulas for 2D animation and construction of puppets for stop-motion.

As stated earlier, story reels, animatics, and layout development are the precursors to previsualization in CG. (Previs is sometimes referred to as rough layout.) Virtual sets are constructed and lit in CG, using the same artistic considerations and patterns that are common to stop-motion practical sets or layout design in 2D.

CG effects animation, especially of natural phenomena and atmosphere, often involves the ability to program particle systems, which drive the effects and, like traditional approaches, are most often stylized and animated in a computer-generated version of straight-ahead animation.

In traditional 2D work, cels are inked and painted (done digitally nowadays) and backgrounds are completed with great attention being paid to the lighting, color, and atmosphere of the sequence. In CG, characters and environments are surfaced or textured and lit with the same considerations, though CG offers a much broader and more flexible variety of surfaces and lighting applications. Finally, when all elements are accounted for, they are composited into a harmonious image—analogous to cel setups that are positioned and photographed in 2D—but the digital age offers an unlimited number of levels to be composited, whereas the older technology was only able to afford six or seven layers to work with before the density of the overlaid cels began to “milk out” the image being photographed.

Camera motion within CG’s virtual sets is unlimited, in contrast to the restricted movement afforded by traditional technologies. The early attempts at depth in cel animation involved either a diorama-like arrangement or a multilevel photography apparatus called a multiplane camera. These kinds of shots were very time consuming and expensive to produce. The fact that CG technology allows as many digital levels as necessary and that each level has the ability to be moved and adjusted independently now makes multiplane effects possible in any scene. Combined with the virtual moving camera, depth has become a much more usable tool in CG animation than it ever was with the traditional technologies.

Finally, all elements in CG animation’s virtual world have to be rendered into the images that their digital information represents. This process is actually begun at the very start of animation production, in rough form, and continues throughout the production process in increasingly more polished versions until the final render and output—sometimes referred to as film-out. The need to differentiate, render, and record the production images as they progress throughout the production process drives the technical aspects of the animation pipeline. So, for all intents and purposes, pipeline has come to mean both the production process and the hardware and software needed to cope with the demands of that process. Most often, it is necessary to determine the technical requirements of specific productions before actual animation begins to ensure that the pipeline is robust enough to deal with the myriad demands of each stage of the process and can cope with the huge amounts of digital information that will be required to render images throughout the production—up to and including the final images.

To this point, the discussion has been focused on CG animation created within and for a virtual 3D world. There is also the creation of CG animation from live-action reference. Motion capture, like its rotoscope predecessor, involves recording the actual movements of humans or animals. Information about the coordinates of these digitally recorded movements is noted in the computer and that information is then used to manipulate the rigs of animated figures.

Finally, it should also be noted that there are increasing improvements and uses for 2D CG technology; that is, 2D stylized animated images using computer tools. There are many ways to do this, from animating in classic 3D fashion and then “toonshading” the images to create a flat or painterly look, to using software applications such as Flash that allow traditionally drawn images that are scanned into the system to be combined and manipulated in a kind of computer version of limited animation.

In the final analysis, by whatever means animated images have been created—stop-motion, traditional 2D, CG technology, or some combination of techniques—the important thing has always been the ability to use these tools to achieve a believable performance, creating not only movement but also honest and identifiable emotion out of the whole cloth of an artist’s imagination.

CONSIDERATIONS FOR A FULL CG-ANIMATED FEATURE PIPEHNE

Stephan Vladimir Bugaj

The most substantial consideration in developing a pipeline for feature animation is the realization that an entire world must be created from whole cloth and usually a complex, visually stylized one at that. This implies certain artistic, technical, and production management requirements. All elements of the world must be internally consistent, without any photographic plates to provide context (and when dealing with a stylized world various new problems arise such as how to stylize physically based simulations into cartoon physics). Since the world must be created from nothing, feature animation requires a pipeline that addresses every step of image generation from production design, to camera layout, to effects, to final composite.

A feature animation studio needs:

•   sufficient artist resources to do the work,

•   large amounts of computer power and storage for the many technical assets that will be developed, and

•   a production management system that can keep track of many assets across many departments, often working in parallel, and over the long time to delivery that is necessary for producing top-quality feature animation (typically 2 to 5 years for an A-list animated feature).

For those familiar with other animation pipelines, it is important to consider the differences between hand-drawn pipelines and CG animation pipelines. Whether the drawing is done on paper or a digital tablet, hand-drawn animation deals almost entirely with image assets (image layers that compose into frames in the film). Drawn animation requires a system for managing collections of image files, organized into something along these lines: style guides and reference keys (including pencil tests), background plates by sequence; if used, reusable element libraries (walk cycles, facial gestures and phonemes, visual effects elements, etc.) and photographic plates to be rotoscoped. Images in this pipeline basically move from storyboard artists, to layout artists, to background pencilers, to background inkers, to keyframe pencilers, to in-betweeners, to inkers, and finally on to the final assembly (either photography on an animation stand or digital compositing of the plates).

A CG animation pipeline, on the other hand, is more technically complex and is broken down into more departments. These departments are often doing parallel work rather than working in a strictly linear pipeline, and the ability to break down assets into elements relevant to separate, parallel tasks is very important, as is reusability. Reusable components, such as character articulation rigs, shaders and texturers, procedural models, and so on, are created by technical artists and reused throughout the film. CG feature animation pipelines seem similar to a visual effects pipeline in that there are CG modelers, shading/texturing artists, character animators, lighters, effects animators, simulation technical artists, and shot finaling departments such as rendering and compositing. However, because there are no photographic source plates, scanning/input and matchmove are not the starting point of an animation pipeline; rather camera layout in the 3D system and blocking animation are the starting points. In animation the CG artists are responsible for all of the acting and cinematography, as opposed to starting from material already acquired on set.

Because animation involves the creation of the entire film, the production pipeline is based around sequences, starting with camera layout and ending with final render, and reel deadlines drive the sequence delivery dates (reels being literally all the sequences that are contiguous on a standard reel of 35mm film). In all filmmaking, a shot is action that takes place during one location and at one point in time.

Sequences are collections of shots that are conceptually related by their role in the story. Everyone is at least roughly familiar with the idea of sequences from DVD chapters, but in animation they are explicitly used to organize production. Assignments are based on sequence delivery deadlines, and sequence deliveries build into a reel lock in which a full reel of the film is fed from final render (meaning the approved CG render prior to color grading) into post-production and ultimately edited in to the reel, color graded, and approved for film-out.

Pre-production (particularly storyboard creation and animatics editing) are also heavily sequence based, and that means production design tends to also deliver based on sequences (with the exception of designs for primary characters—a variance that character modeling, rigging, and shading also share because the primary characters appear in too many sequences to be sequence-driven deliverables). Environments (sets/locations) and master lighting rigs are developed for sequences, except for certain key multi-sequence environments, and this work is then inherited by all shots in a given sequence.

Task completion is tracked at the asset level, which blocks all shot-level completions until the assets in a given shot are finaled. Layered on top, shot-level task completion is tracked for the shot work itself, and that blocks sequence completion. A sequence is blocked against completion until all shots in that sequence, and by extension all assets in those shots, are completed. Sequences and reels are reviewed in-context to ensure continuity, both during in-production checkpoints and after finaling. At any point, fixes may be opened on assets or shots that block delivery of the sequence (and therefore its reel) to post-production, and these also must be tracked in the production management system.

CG Feature Animation Pipeline

Storyboard Animatics and Previs

The entire film is boarded and edited into animatics reels. Because changing animation is so expensive, board artists and editors work closely with the director, head of story, and screenwriter to develop the animated film to a nearly finished edit before production begins in earnest. Previs is a full CG extension of the animatics process, where boards are replaced by CG models, and is sometimes used in animation in the same way that it’s used in visual effects: to mock up the environment, camera, and action in 3D prior to deciding on final shot designs.

Production Design and Look Development

Because there are no plates, the creation of production designs as a starting point for creating the visual world is utterly essential in animation. There is no opportunity to find interesting people, places, and things in the real world to film. Look development is the process of taking sample sequence(s) and pushing them through the entire pipeline (all the way to film-out) in order to find the right way to translate the production designs into a CG look.

Layout and Set Dressing

In animation, layout does the camera portion of cinematography. Layout artists determine framing and camera moves using virtual cameras that have frame aspect ratios, different focal lengths, aperture/depth of field (DOF), and even virtual dolly, crane, and other jib controls. Layout artists also block in rough environments and character animation, using whatever models are available at the time. Set dressers take the rough layout stand-ins and dress in the real finished environment models for each sequence.

Modeling, Texturing, and Articulation (Rigging)

Environments and character models are digitally sculpted and textured, and characters and other animated models are articulated. In large feature animation pipelines, this may constitute as many as eight departments (one each of modeling, shading, texture painting, and rigging for characters and environments).

Character Animation, Character Simulation, and Crowds

Voice recordings are made, and characters are keyframe animated based on the reading (or motion capture data is acquired and finessed). This is the acting phase of animated filmmaking. Clothing, hair, and any procedural animation elements that are driven by the animation key frames are also done at this time. Shot simulation department(s) may finesse the final simulations, but the setup departments for cloth, hair, and procedurals provide the animators with stand-ins so they can see rough simulations in real time. Crowd animation is usually achieved by a separate department, consisting of animators who create libraries of crowd movements, and technical artists who use procedural systems to control the location, grouping, and movement of the crowds.

Lighting and Compositing

The virtual scene is lit, completing the cinematographic look with light and shadow. Although this stage of the pipeline is easy to describe, it is often a very precise and time-consuming process, because in an animation pipeline there is no natural lighting to modify and enhance. Additionally, in animation, decomposition of shots into layers of elements, and subsequent compositing, is usually done in the lighting department.

Effects Animation and Simulation

Effects artists in animation do nearly everything that effects artists do for live action, using all of the same simulation, effects animation, and compositing techniques, and yet this is only one stage of the CG animation pipeline. And because animation has no real world plates, even easily photographed elements such as a burning match, a car tire kicking up dust, or a person drinking a glass of water must be laboriously constructed.

Shot Finaling, Rendering, and Cleanup

Shot finaling departments are responsible for cleanup of the scene construction, layering, and compositing, as well as optimization to meet render time goals, and post-render paint/roto cleanup if necessary.

Post-Production

Most animation studios do most, if not all, of their post-production in house. Editing has been going on throughout the production process. In post-production, a final edit is created and then the film goes through color grading, sound design, sound editing, final sound mix, and finally film-out (which includes one or both of literally filming out duplication masters on an Arri laser or similar device, or packaging them up for digital cinema delivery).

To create a feature animation, a studio needs artists who can do all of those things. Notice that this includes things like cinematography (both camera and lighting), editing, and acting (character animators) that aren’t normally considered essential by shops accustomed to work other than CG feature animation. This means a different staffing and artist development strategy, and a comprehensive filmmaking approach to managing the studio.

Production Management

Production management is another key part of feature animation. Most CG animation shops spend a lot of time on general asset development before a single shot is ever finaled. Character rigs, shaders, garments and wigs, effects and procedural animation rigs, and so on, are all laboriously tested—not just technically, but in look development tests as well. This means that a good percentage of the 2 to 5 years spent creating an animated feature involves everything being in an unfinished state. This situation requires both the right attitude on the part of the producers and a robust production tracking system necessary to keep track of the many assets, and the many intricate interdependencies among assets, and between sequences, shots, and assets. A feature animation production tracking system needs to:

•   Track different departmental checkpoints for each kind of asset and shot. A shot or asset needs to clear multiple departments, each of which may have many task checkpoints. These are likely to be interleaved as departments hand work back and forth in order to support iterative refinement.

•   Track interdependencies between technical assets such as geometry, articulation rigs, and shaders, which may be built up using multiple levels of reusable assets. An example would be a sequence shading materials palette that depends on a show-wide materials palette, which in turn depends on a show-wide shading template library, which references a studio-wide shading template library that’s built upon a shading language function library.

•   Track dependencies between shots and assets, sequences and shots, and reels and sequences.

•   Have a system of notes and fixes through which production management can, based on notes given by the director or other lead staff, block clearance of (or even “un-final”) a shot or asset relative to a given departmental checkpoint—and therefore all assets, shots, sequences, and reels that depend on it.

Technical Considerations

Technical considerations are also important. CG animation requires different software from cel animation, and although the base packages are often very similar to what is used in the visual effects world, the exact configurations and usage thereof are determined by the fact that a single shop is going to do production design, animatics, previs, creation of every element in every shot, and much of the post-production. Feature CG teams also tend to be larger, placing an added technical infrastructure and budgetary burden on the studio as compared to a live-action film of a similar scope.

Technical considerations for CG feature animation include the following:

•   The need for end-to-end software pipeline integration that allows assets and shots to be made up of assets comprised of many components, starting with images and clips created by production design and story, and carrying through into production where many associated 3D assets will be created, which then get rendered out to images that feed into postproduction. Add tracking software to keep track of it all.

•   A fast, high-capacity network. CG animation moves around a lot of large assets, and parallel iterative refinement means this is done quite frequently.

•   A large render farm. Because every element seen in the frame must be generated, animation films are often rendering enormous scenes. Even though the scenes are usually broken up into smaller parts to reduce the load on individual render farm machines, a lot of powerful machines are needed to compute all of the scene data. Farms with several thousands of nodes are not unheard of.

•   A large image storage farm. Since there is no photo imagery for reference, all the work of previous departments (and multiple iterations thereof) acts as reference materials for the departments farther downstream. Therefore, huge numbers of image assets are kept online.

•   A large model and shot storage farm. To allow for version control and other safe work practices, multiple revisions of every file for every shot and asset made must be kept. Since that’s everything in the film, that adds up quickly.

•   A sophisticated backup system. All of this data need to be backed up, and in a way that does not interfere with ongoing work. The backup system must not only be of sufficient capacity, it also needs to be highly optimized.

While many of these considerations, technical and otherwise, also apply to highly effects-heavy live-action films such as the Pirates of the Caribbean series (2003, 2006, 2007) or Lord of the Rings series (2001, 2002, 2003), this is because those films operate at a scale where they come close to the staffing, management, and technical requirements of an animated feature. However, even in those cases, the existence of photographic plates means that certain elements of production on the CG side are omitted, such as camera staging and editing, while others are reduced, such as acting and lighting. When developing a full CG feature pipeline, the main thing to consider is that no matter how experienced a studio may be with some substantial portion of the filmmaking pipeline, with full animation there is nothing to fall back on and it is necessary to become capable in every aspect of filmmaking.

MANAGING AN ANIMATED FILM

Don Hahn

Film Management and Personal Style

The experience of managing an animated film can vary depending on the technique used for the film. CG production and motion capture usually present more technology challenges than do stop-motion and 2D animation. But the approach to management is essentially the same no matter what technique or medium is used.

To begin this section on the management of an animated film, a discussion of personal style is in order. Each producer has a personal bag of tricks that he or she brings to a project. Some are known as creative producers; others come from a business or finance background. This personal background will have as much to do with the management approach on a film as almost any other factor.

The producer on an animated film is a coach, psychotherapist, and cheerleader all wrapped into one. Producers have an important creative role in helping the directors tell the story by building a strong team of collaborators on each film. There is also an equal responsibility to the studio to deliver the film on time and on budget.

Budgets and schedules aside for a moment, the producer’s number one priority is to assemble and maintain a world-class team and create a movie of lasting quality. What do producers need to do to successfully manage an animated film? The answer is deceptively simple: They hire the best people that can be found and then do exactly what those people tell them to do!

Building Brain Trusts

A priority for any project manager is to make sure there is a brain trust of experts in place as a support network. Strong brain trusts must be assembled in three areas:

1. Administration: consisting of a human resources head, a lawyer, a production finance person, and a technology lead.

2. Production: The top managers of the film who evaluate the project, set goals, and then help the team attain them.

3. Creative: A group of trusted story artists who will tell the truth about the movie and where it needs to go. A brain trust of writers and story people can help navigate story problems as they happen. At Disney/Pixar the story brain trust includes directors from other projects who can come in with completely fresh eyes and give notes on a project. This is a secret weapon that far too many studios ignore. The producer/manager needs to create that atmosphere of creative debate in a way that will “plus” the movie with frequent notes and debate.

These three groups of people—administration, production, and creative—are the safety net. No single individual has the expertise that these people will bring to the table on a regular basis to solve the problems that come up during the course of the production.

Business Affairs

The business affairs person on the film consults on employee contracts, liaises with labor relations, and negotiates contracts for key talent including producers, directors, composers, and songwriters. Voice talent deals need special attention: Is the talent expected to make personal appearances for the film? How many recording sessions are guaranteed in the contract? What, if any, bonuses do they get on the back end of the film? The guidance and counsel provided by a business affairs associate bring essential order to the filmmaking process.

Production Finance

The schedule and budget are living and breathing documents. It is important to remember that animation is not a traditional assembly-line activity; it is more like sports. A team can prepare, train, and recruit the best players and set a game plan, but it has no control over the variables of the game. The team does have control over how quickly it can react to changes in the game, but it can be prepared and conditioned to adjust to the new conditions and still play at top level.

The weekly monitoring of each department will paint a picture of how the team is playing. Some departments are never a problem and others, usually departments that are prone to creative changes, can be chaotic. The success of these departments in the crunch goes back to the original casting of the players and their experience with the stress and fluidity of production.

There is a mythology in production that for a film to be properly managed, it has to hit quotas, budget, and schedule as originally planned. The truth about successful production management is all about adapting to change. Again, the key here is preparedness. Expect and train for change. Expect chaos and moments of indecision, and then train the management team how to react to the change; accessing, listening, planning, and refocusing attention are the goals.

Human Resources

The term human resources (HR) has expanded to include many diverse areas of traditional personnel department functions. The HR staff establishes compensation guidelines for the production and can provide valuable third-person counseling to individual crew members outside the pressures of production.

A key part of human resources management is performance evaluation. If regular performance reviews are in place, the company will have the chance to coach employees and guide them to better performance. In the case of high-performing crew members, it is an opportunity to provide reinforcement, and even bonuses or salary adjustments as a reward.

Animated film production is a marathon and it is normal to have some staff turnover. Regular performance evaluation keeps an open channel of communication with the employees and offers the producer and production management staff a way to coach and counsel people throughout the long production period while constantly keeping the crew performing at the highest level.

Recruitment and training often fall to the HR department. Productions need to staff up at a moment’s notice and rely on HR to provide an up-to-date contact list of potential crew members. Production should guide HR as to specific future needs and provide a clear artist loading schedule on a film. At the same time HR needs an active connection to the industry, labor unions, and schools to help supply the demand for strong people. Most forward-thinking studios have strong long-term relationships with potential talent and encourage artists to upload their portfolios to the studio website.

If the film is a union project, HR also needs a strong connection with labor unions. Strong and frequent communication with the union rep or shop steward will speed communication when a labor problem arises.

Last, HR often takes on the overall task of training and development. Training on an animated film needs to be focused and production specific. A producer or production manager should let the HR department know about necessary training goals and problems. It is better for the production to put forward the training priorities that will directly benefit the movie than to let HR administer the program.

Building the Core Creative Team

Assume a concept for a film exists and now it is time to build a creative team to execute that concept. Take time in the process now to evaluate the scope and scale of the concept and take stock: Can the story be executed for the money budgeted? What new technologies will be needed to execute the film? What key talent can be identified now to start the production?

The director is the first crucial hire. He or she has three responsibilities:

1.  to articulate the vision and the story clearly to the crew,

2.  to give candid critiques, and

3.  to rally the crew and build morale around that vision.

The storytelling aspects of directing are obvious: Work with a writer and the story team to bring the film to life on reels. For a producer, the key ingredient is to foster collaboration between the director and all of the direct artistic leads such as art directors, editors, and animators. It is the director’s vision—but the team has to be on board with that vision. It doesn’t mean that the story must be negotiated with the team, but the team does need to be heard and the content needs to be debated. If the story isn’t debated early and with candor, it will get debated in the reviews when the movie opens. It is better to have those creative discussions early, using the process to build a better, stronger, more coherent team with which to work.

It is not surprising that morale building is included here as one of the director’s duties. The core team, and eventually the entire production crew, has to sustain the director’s vision over a period of years. And that team must believe where the director is taking them. There will always be frustration, debate, and disagreement on any film, but if the team can agree on supporting the director’s vision of the film, the process and result will be stronger.

Writing and Visual Development

Simply put, the script is a road map for a movie and the point of departure for any film project. In animation, it is a fluid document that changes as ideas grow and develop. Animation is a visual medium, so the writers work closely with storyboard artists, visual development artists, and the director to create characters, dialogue, and situations that work well visually. It is a highly collaborative team effort that makes a story work. Animation writers need to be in the room and working around the table with visual artists to launch the movie properly. The script and visual development art are important management tools to evaluate the artistic scope and scale of the movie and aid in early planning.

Storyboarding and Reels

The storyboarding process developed by the Walt Disney Studios in the 1930s was a creative breakthrough and remains the biggest management tool around for animation. As the original script is boarded, story artists supply a blueprint for the film that can be used not only to judge story progress but also to refine the budget and schedule for the film.

The move from script to storyboard is always a bumpy one. Animation is a visual medium and the story has to mature in a visual way. The boards are an incredibly inexpensive way to workshop the story before production starts. The storyboards allow creative heads to refine and remake the film in a visual way that will translate better to the final product. It is the least expensive and most effective way to introduce quality into a film.

By investing a few more weeks or months of story time into refining the storyboards, two things happen: A better story can be told in visual terms, and everyone will gain a better indication of the exact scope and scale of the film, which results in more accurate budgeting and scheduling.

Research, Training, and Development

The management team can work wonders in the research phase of a film. Once the central creative team is together, it is crucial to give them immersion training in the topic and world of the film. For some this means a field trip to the location of the film: The Lion King (1994) staffers traveled to Africa. The Cars (2006) crew took a trip on Route 66.

Producers can also bring in lecturers, experts, and entertainers to inspire and inform the staff and typically a reference library of films and books is organized for the crew. These early steps to immerse the crew in the film’s subject will pay dividends time and again.

There is a team-building aspect to this phase as well. As the team travels, visits museums, or works in long lecture sessions together, the team members get a chance to hear each other’s thoughts and test them against their own. It can be a much-needed early bonding experience.

Sometimes production management abdicates staff training to the human resources department. That should never happen. Dictate to the HR crew exactly what the needs are and let them help raise the level of training that will specifically benefit the movie.

Working with a Studio

The management team on a film has to “manage up” to the studio as well as down to the below-the-line crew.

Relationship building is the key here. No executive likes to be an outsider to the creation of a film he or she is charged with overseeing. There are times when it may be best to filter the kinds of information that flows to the studio, but the best course of action is to keep the studio very close, particularly in the early development of the project.

The early crucial decisions of story, scope, scale, artistic leadership, voice casting, and budget are all shared decisions. That is not to say that the studio is directing the film, but they are certainly commissioning the film and as such need to be a party to the creation of the project.

Each studio is somewhat different, but studios usually contain a very small core team of players who are collaborators. The studio head needs to be involved in key creative and financial decisions.

The studio executives in charge of production, finance, marketing, consumer products, and business affairs are the next line of important relationships. These relationships are important alliances for the management on an animated film. If the relationships have been built properly with these people, and if they are kept up on the status of the film, these people become a safety net when things go wrong. Once again any producer of an animated film worth his salt knows to look at these people—finance, marketing, business affairs, production, consumer products—as part of his or her brain trust. They are not bosses as much as they are partners, and by the very nature of having commissioned a film, they have a vested interest in having it succeed.

Completion Bonds

Most independently financed films, including many that are distributed by the major studios, require a completion bond.⁸ A completion bond or completion guaranty is a written contract that guarantees a motion picture will be finished and delivered on schedule and within budget.

The production usually secures a completion bond for the benefit of the bank and film financiers. In general, it assures them that the producers will complete and deliver the film in keeping with the screenplay, budget, and production schedule that the bank or financiers approved.

If the production of the film is abandoned, the completion guarantor will fully repay the bank and financiers. As more and more independent animated films appear, the completion bond is an important component of the film that provides assurance to the investors. It is not to be overlooked.

Managing Expectations

Managing expectations during the making of the film can keep the crew focused during a long and often grueling production period. For example, most early screenings of a film are horrible. On Who Framed Roger Rabbit (1988), an early audience preview screening was held where more than half of the audience walked out. Other films that failed horribly in their early versions include Toy Story (1995), Beauty and the Beast (1991), Aladdin (1992), and even Bambi (1942).

The crew and the studio need to understand that animation is one long intense process and early misfires are common. It is a need for concern, but not a need for panic. Bring in the creative brain trust and be candid and brutal about the film’s faults. Then dive back in.

Managing the buzz from a bad screening is crucial. As a manager, it is important to deflect panic and get people on the crew focused on the tasks ahead to make the film better. The movie should not be shown to anyone outside the core team of filmmakers led creatively by the director until there have been multiple screenings to improve the film.

Managing expectations is the key. Yes, the film will have to be shared with the crew and the studio, but the setup to the screening is as important as anything in these early days of the project. Remind people that this is “a rough sketch” or “a very rough pass on this sequence.” Anything to prepare the audience for the state of the film will help temper unneeded notes and critiques.

On the other hand, always solicit notes from the crew after a screening. It does two things: It produces a handful of genuinely useful notes from fresh eyes that haven’t seen the film before, and at the same time a sense of trust and team is built that is so important as the film goes into the production crunch period.

Facilities and Environment

The physical environment can help or hinder the production. Two examples: Some of the brightest and best animated films have been made in open-plan warehouse environments. The advantage of open plan is that it fosters immediate communication and a sense of team. One of the problems with the old animation building on the Disney lot is that it had separate wings with each artist having a private office. This hindered communication and it actually became much more formal. Memos and meetings took over where casual conversation once worked. The layout of the facility made it very hard to get the crew together in a common space where meetings or casual communications could take place.

Staffing Up for Production

Clearly the work of building a crew into a team is the most important job in managing a project. Alfred Hitchcock said that most of his work was done by the time he had cast his movie; his words can easily apply to casting the team for a production. Hire the brightest and smartest people, and then allow them to do what they do best. The success of a production is based purely on the team of people cast to create the film.

A number of approaches can be used to cast the team. Keep in mind that it is not a party—everyone doesn’t have to get along all the time, but there does need to be respect among the team players. Producers who cast their movies based on low labor cost usually get what they paid for—and in the end sometimes end up paying a great deal more than they should have to get an acceptable product. Remember: Cast the film with the best talent that can be afforded. Look for these crucial skills: the individual’s expertise, management skill, candor, and the ability to push the team to a higher level.

The team is only as good as the weakest member. Hire only “A” players who are not afraid to be surrounded by other “A” players and the team will have the potential to work at a very high level.

Production Management 101

Managing a project starts with thoughtful analysis of the scope and scale of the project, followed by a detailed breakdown of the project into smaller manageable tasks. The next step is to plan and commit people and money to work on those tasks, and finally to monitor the completion of each task on a regular basis. In short:

•   Analyze the job.

•   Break it down into component departments and weeks.

•   Commit people and money to do it.

•   Monitor the progress until finished.

At the core of all this is the trust and belief that the crew can maintain the schedule. Two criteria should be balanced:

1.  The project has to be delivered on time and budget.

2.  The project must live up to the highest creative potential possible.

There are no shortcuts to streamline production or they would have been used many times over.

Chaos and problems are the currency of an animated film production. The first step is to recognize that problems are a constant. The next step is to identify the problems and fix them quickly and meaningfully. Work with the artistic team to spot problems early.

A regular problem that shows up early is the lack of inventory. Each department relies on a regular flow of work to keep their people busy and to supply the next department with work. When early story changes happen, or departments fail early in the flow of work, the pipeline flow dries up and the crew sits without work. It is an incredibly common problem and also very fixable. It is better to send shots down through the pipeline to warm up the team and get them working. Some managers wait until everything is perfect before they will okay work for production. But it is far better to send off a few dozen shots, even if they have to be redone, just to test out the systems and see some early results.

Early shots are never pretty, but they serve to test the pipeline and the crew and get both ready for the real work crunch coming.

Once work does flow to the crew, do not stop it or let it up. Inventory feeds the studio and it cannot dry up. Regular approvals to send work into production are necessary at regular reliable intervals to feed the production monster.

A manager’s job is to serve the creative process and provide the team with a trustworthy plan to complete the project. If the team trusts the plan, and if they trust the manager’s ability to constantly revisit that plan in an effective way, then compliance with the production plan, budget, and schedule is a given. If the crew doesn’t have faith in the manager’s planning abilities, or his or her ability to recognize faults and repair them, the manager will always be chasing their trust—and it is a very hard thing to regain once it is lost. Never, ever, lose the trust of the crew. Admit mistakes, ask for help, but always solicit a mutual trust and respect between artist and management.

Reliable data tracking is the other basic prerequisite to strong management. Production finance will usually provide productivity reports and financial cost-to-date reports to help manage the problem. Accurate up-to-date information can help the manager anticipate problems and stop them before they get worse.

Using Subcontractors

It is likely that the production of a film will include some work to be done by subcontractors outside of the studio environment. For the most part, the deal with subcontractors is a standard negotiation. It is prudent to get three bids. See samples of their work, and get references from others who have hired them. Part of the deal with a subcontractor will be to set goals and benchmarks for delivery. These expectations should be clearly set out in advance in the contract and then managed during the production process.

Work done through a subcontractor should be assigned with careful documentation along with expectations for completion. In some cases if the amount of work is significant, it merits a remote manager to monitor the progress of the production. Receipt of work from the subcontractor should be carefully documented and weekly cost reports should be built in to the weekly production reports.

The Production Brain Trust

Just as there is a brain trust for the administrative team and a story brain trust for the creative team, there is also a brain trust for the production team. This is the engine room of the film. The key players here are the:

•   Associate Producer,

•   Production Manager,

•   Department Head, and

•   Production Department Manager (and team).

The Associate Producer has three things to worry about: people, time, and money. Associate Producers are always solving problems that sound like they come from some nightmarish math test. For example: If an animator can complete 4 seconds of film per week and the film is 90 minutes long and has to be animated in 52 weeks, how many animators will be needed to complete the film on time?

The mathematical answer is 26, but the real answer depends on how many characters there are per second; how experienced the animators are; how many holidays, sick days, and vacation days fall in that year; how much overtime the budget allows; how many versions are required before the scene gets approved; and how many computer crashes will take place during that 52 weeks. The art of producing and managing an animated movie has to allow for false starts, changes, and animation that doesn’t work the first time around.

Being a Production Manager (PM) for an animated film is like being the mayor of a small city full of filmmakers. The PM works closely with the producers to set goals for each week and manage the daily flow of work. Each department will need director time, sweatbox critique sessions need to be scheduled, and regular production meetings need to be scheduled in order to keep information flowing between departments.

Film crews are organized by department, and each department has a department head who is a senior artist or technician who manages the artistic or technical goals of a particular department. Most importantly, the department head keeps watch over the quality level of the work—be it animation, modeling, or visual effects.

The department runs like a small business. A PDM (some studios call them Assistant Production Managers or Production Co-Coordinators) runs the department’s business. PDMs report to the production manager and work as a team, collaborating with other PDMs. They all manage the production each week. Then there are armies of tireless production secretaries, production assistants, and administrative staff who do everything from typing scripts and memos to arranging for catering during the production crunch, answering phones, booking travel, scheduling massages for tired artists, making coffee, running errands, and generally helping to move the production ahead.

Encouraging Iteration: Managing Change

Another aspect of fostering the creative environment is the art of critique. It is the secret weapon, the advantage that animation has over other movie techniques. Creating an atmosphere where opinion is not only tolerated but also welcomed and fostered is the goal. At every level, foster an openness to discuss and express feelings on story, animation, design, and nearly every aspect of the film.

The director and producer have to be particularly tolerant and welcoming to this culture or it won’t happen. The creative leads on the show have to be willing to take the time to listen to dozens of ideas, some of them awful and some useful, in order to get the film to a higher level. This group critique culture builds a tremendous amount of ownership between the filmmakers and the product if done right. Everyone’s pride grows when they feel that they have been heard or, at least, included in the process. The director will still have to go back to his or her office and sort through the notes and the final product is, at its best, a single vision, but the culture of honest debate and creative critique is crucial for films to reach their highest level.

This is an intimidating process for production management who want to hit deadlines. Soliciting changes is completely counterintuitive to achieving this. But the job is to deliver the highest quality film on budget and on schedule. If there is no cultural permission to be critical, then there is no incentive for the artists to contribute at their fullest level and they treat the project as nothing more than work for hire. Always driving to inflexibly hit numbers shows that there was no contingency for change and the message that schedule trumps quality is sent out. Neither of these things is good.

Technology

A producer needs a close partnership with a technology lead. This is a person who is constantly trying out new software packages and testing new commercially available tools. Each film is different, so if a project needs a particular tool that doesn’t exist, a software engineer can create a custom tool for that production.

For the movie Monsters, Inc. (2001) one of the main characters, Sully, needed to be completely covered in fur from head to toe. At the time, no off-the-shelf software was available to do this. So the software crew wrote tools to grow and groom fur on a computer-generated character. They also created other tools and procedures to manipulate that fur in just about any way that an artist might need. It was a huge breakthrough and the film’s success proved the value of the software and technology team’s efforts.

The software team doesn’t develop their tools in a vacuum. The close and equal collaboration of technical and artistic filmmakers produces the best circumstances for quality and yields the best software tools to fit the needs of the story. Always remember that everybody tells the story—everybody—even the software developer. The technology lead also plays a crucial part in keeping the equipment current and running and scheduling upgrades and maintenance.

Post

The editor, who has nurtured the film every step of the way, still has important work to do during the post-production processes when most of the film crew is long gone. Editors have been on the movie from the start, building the earliest story reels and working through dozens of iterations of the film during production. Now the editor presides over final sweatbox sessions and starts to collaborate with music and sound editors to get all of the sound and picture elements to the final sound mixing stage on schedule and with the highest quality.

It is worth pointing out that an editor’s most important job on the film is, simply, editing. When a crew spends years on a project, they sometimes lose perspective. At this final step of a film’s production, a good editor can look at the film with a fresh eye and cut scenes or whole sequences that are too lengthy or that no longer fit the structure of the film.

Managing the Event

A more global view should be taken of managing an animated film to include the event of the film. This means an awareness of consumer products, games, premieres, and events surrounding the release of the film. Consumer products department(s) will need key art and character art for toys and puzzles. The marketing department will need images from the film. Publicity will need access to the key talent of the film. Screen credits and final legal issues need to be resolved. DVD bonus material will need to be shot. Record albums are produced, finished, and shipped; theme park tie-ins and promotional partner relationships are launched.

It comes down to one simple issue: Part of the job is to sell the sellers. A good management team is always feeding the ancillary business groups that will create the toys and products that support and market the film. It is mutually beneficial to have the filmmakers create some of the art for these products and to at least have a discussion with the marketers about how the film will be presented to the public. At this point, the sellers will not only need art, but they can also benefit from the team’s enthusiasm for and understanding of the story. Thus, they should be given clear information on what makes this film spectacular and unique.

Audience Previews

There were 11 previews of The Lion King (1994); each time the story or the pacing of the film was changed. There is nothing magical about this process and it is a huge advantage that animation has over live-action films: the opportunity to animate new scenes and make changes to existing animation weeks before the film opens. Again, managing change is the key. In the early planning of the project, plan for these previews and embrace them as an important tool to the filmmaking process. The preview is not for marketing. They will learn from it, yes, but it is really done so that the film can get down to size and to be raised to its final level of quality. Learn to use previews.

Eventually, persistence of vision leads to the moment of truth: opening night, when the lights go down and the finished movie is shared with the general public for the first time. An animated film is the ultimate paradox. On one hand, it takes years of work and millions of hours of passion and labor to produce. On the other hand, the audience needs to completely forget all of the technology, craft, and hard work that are poured into a film and simply sit back and enjoy the illusion.

THE PRODUCTION PROCESS: AN ANIMATOR’s PERSPECTIVE

Lyndon Barrois

Working on CG-Animated Content in Live-Action Features

This section provides a general introduction to the world of combining character animation—the process of giving life and personality to inanimate objects to convey a believable performance—with live action.

Planning the Process

Many aspects are involved in the mixing of character-animated media with live-action plate photography. Certain guidelines are adhered to for proper execution. Even though it is an aesthetic medium, where in many cases the process and decisions are made off the cuff, there are still rules that apply to completing a project. Breaking down a script is by far the most important aspect of the field (and is covered in Overview and Breaking Down a Script in Chapter 2). A breakdown is simply the process of taking the script and dissecting it page by page, line by line. It focuses on the stage direction, dialogue, and character situations. It pays strict attention to the number of characters, scenes, and most importantly shots in which the characters appear or are involved in the story. When these things are determined, the characters can then go into a design phase. In a 3D spatial world, a character’s design is a crucial aspect of its performance on a 2D surface plane. It has to be designed in a way that is fitting to the story and to its nature.

Case Study 1: Avatar (2009), Iron Man (2008), Matrix Revolutions (2003)

The main characters in these films require an actual human character to be placed into a large, robot-like mechanism that has to be driven by that person to fight in a war. The production is allowed a very liberal budget (every filmmaker’s dream). The sequence has to be compelling and convincing. The desire is for the audience to feel the experience along with the actor. An experienced visual effects team may use an array of complex devices, including motion-based rigs, motion-controlled camera photography, and B and C witness cams. Using the script and breakdown as cues, the sequences are designed by using many tools and work flows: storyboards, previs, character design, character builds, motion tests, plate photography, and slap-compositing.

Once the sequence has been boarded and the characters rigged and built, motion tests begin. The specific use of this data should be twofold: first, to figure out how a behemoth human-driven artillery robot moves and performs; and second, to take the digital animation curves and feed that data into a three-piston motion-based rig on set to have the animation actually move the human driver according to the required performance of a specific shot.

To accomplish this, a rig is constructed for the actors to actually ride in the animation so as to sell the performance of the actor in the scene to an audience. The practically built driver’s seat for the actors is set on a three-piston gimbal that rotates in the X- and Y-axis to replicate east/west movement and tilts in the Z-axis to translate falling. To give the illusion of traveling in Z space, the A camera is attached to a motion-controlled camera rig that trucks toward and away from the seat and also pans and tilts for additional compelling angles. For compositing purposes on set and later through the pipeline, the pistons and the entire backdrop are covered in chroma-key greenscreen material and paint.

The raw video footage is then slap-composited onto the animation via QuickTime and on-set editing software for proof of concept. When all looks convincing, the shots are taken back for cleanup animation and tracking and then sent through the production pipeline for effects animation, final lighting, and compositing. When the mix is complete with gunfire, explosions, tracer fire, dying enemy combatants, and thundering debris, mayhem and magic are captured on film!

Case Study 2: Alvin & the Chipmunks (2007), G-Force (2009)

The main characters in these films are either chipmunks or guinea pigs. Therefore, the first question that must be answered is whether the rodent will be a realistic character or a stylized version. The characters’ design is determined based on what they have to do in the story; that is, do they just do the normal things that these animals do or are they doing things out of the ordinary realm of their nature? These scripts called for the animal to do the latter and, therefore, they had to be designed in a way that can achieve the desired performance.

A team of character designers is assembled to accomplish this crucial first task. The breakdown, storyboards, and/or 3D previsualization are generated for staging and blocking of the sequences and shots, and the character is designed based on its physical capabilities and what it has to do in those shots.

A common misconception is that in CG a character can do anything—which in essence is true. But the larger question is “Is it pleasing to the eye?” Strict attention is paid to its anatomy and skeletal makeup, as well as scale and proportion to its environment and real actors. Maquettes⁹ are usually generated in a series of action poses to convey a character’s performance and to allow the director and visual effects team to view the character from all angles. Then 3D models are built in the computer to the exact specifications of the maquettes. However, these models are not built in the action poses, but rather a default T-pose, which is a standing with arms spread pose, to determine its physical makeup. In this phase the characters are scrutinized from every angle via turnarounds (where the character is placed on a turntable in an animation file and rotated 360 degrees).

Next, the process of rigging, or assigning animatable controls to the character’s body parts, face, and extremities, begins. From there the character goes into the hands of character animators, who do a series of motion tests to determine how effectively a character moves in space. Questions of how does it walk, run, jump, push, pull, etc., are all taken into consideration. Hours of footage of the real animal are studied to determine these aspects. (In the case of nonexisting creatures such as dinosaurs, real-life animal references are chosen for their relative nature to their predecessors—bears, ostriches, rhinoceroses, giraffes, alligators, etc., may be studied). Another useful reference tool is actually studying the footage of the character’s voice-over actor, for annunciation, facial expressions, and mannerisms, if the character is required to talk.

In many cases, certain actions of a character aren’t specifically spelled out in a script; therefore the character should always be designed to perform any task that it should realistically be able to do. A common misconception is that cheats are a necessary tactic when integrating CG characters into real-world environments. A character’s physical integrity must be maintained from all angles, not just from the main camera angle. This is especially crucial when sending animation through the pipeline. Violated poses, those that involve interpenetration of body and limbs, hyperextending of extremities, and improper connections with ground planes and space relationships, are always sent back to be corrected—even when a performance is deemed final.

Production

Shooting Footage

Once a script is broken down and all of the considerations discussed above have been made, the shooting phase begins. When a character is to be added into a scene, on-set supervision is a must. Painstaking attention to detail must be maintained to sell a compelling and believable live-action performance. This involves interaction between the character(s) and the actors in terms of conversation, physical contact, sets, and props. For conversation, connecting eye lines is the key step to selling a performance. An audience has to believe that the actor is interacting with his or her animated costar. Timing and cadence of dialogue, touching, and disturbance of props are all taken into consideration.

If the character is CG, full-scale 2D cutouts or 3D prosthetics (or stuffies) of the character are generally used as a physical reference on the set. When these aren’t available to a production, a rod with a marker on it is used for the actor to focus his or her eye line when delivering lines or emotional responses. The shooting takes place in the usual manner with A, B, and C cameras, or witness cameras for coverage and proper tracking data. The tracking data (i.e., lens, distance of camera to character, measurements of the physical environment being shot, etc.) are all crucial to the integration process. This determines that the character is integrated into the proper camera space with the actor or environment. If these aspects are recorded improperly, the integration suffers and results in a fake look. Eye lines are off, the character is out of proportion, the character slides around the plate (footage), the lighting and compositing are improper, and the integration falls apart. This can also lead to gross budget overruns in an effort to “fix it in post.”

In many cases, clean plates¹⁰ are filmed for the purposes of aiding the compositing process. If an animated character covers a spot through body movement or dialogue, the displaced footage needs to be replaced or covered. This also holds true when props are disturbed by a character’s interaction with them and whether those props have to employ CG replicas or not.

For interaction with a CG or 2D character, the same tracking principles are employed. In the case of stop-motion characters, however, stricter attention to detail is a must and plate preparation has the extra component of employing motion-control methods of shooting. By its nature, stop-motion photography is a live-action process using the time-lapse or pixelation¹¹ method. The footage is generally shot in a series of passes. If it involves a live actor or two, that person or persons are shot with all of the performance considerations in mind, and another clean pass is shot to time for integration of the stop-motion animator to work within that same framework. Usually that animation will take place on a greenscreen stage to the exact specifications of the live-action shoot. That performance will then be composited into the footage with the actor. If the stop-motion performance doesn’t involve the actor but rather a solo animated performance and a camera move is involved, that camera is a motion control rig, timed and staged specifically for the length of the shot. The animator steps in with the puppet or puppets, gets the desired performance, and backs away. That shot is then recorded again in-camera either in real time or as single-frame passes for clean compositing.

Tracking Footage Properly

Tracking is the process of locking the relationship of character, environment, and props frame by frame in a 2D plane and conveying that in 3D space. It most often involves very complex camera moves. An experienced eye can immediately spot when a track is off—as can the audience members, though they may not be able to verbalize it. Therefore, it is always crucial to check the track first before proceeding with any blocking or performance work. If tracks are off, they need to be corrected before any meaningful work can proceed. In a complex show with hundreds of shots, truncated schedules, or both, these things sometimes unintentionally slide through the cracks. Therefore, having good communication and a good rapport with the tracking team is crucial. The size of a character or object should never be altered to fake a track. Once that footage is properly digitized and placed into the computer with all of the data gathered on set, the 3D environment is generated to exact scale and specifications. These include ground planes, wall to ceiling spaces, furniture, etc. For convincing character integration, proper tracking is a must to ensure correct lighting, shadows, and reflections with regard to a character and its environment.

Working with Riggers

One of the most important aspects of character development is the relationship with the character rigging team. The character rigging team’s involvement stretches from the aspects of the actual character animation, to technical effects animation relating to the character, to software and associated technical issues that may arise as that character flows through the integration pipeline. Muscle simulations, cloth, fur, and water interaction are some elements normally created by the effects animation team, but they also work in combination with the rigging department. If a character requires special tools to squash or stretch, distinct facial blend shapes for lip syncing dialogue, cloth interaction due to its costume, hair and fur behavior, and water interaction, these requirements are partly or wholly shepherded by the rigging team (depending on a particular studio’s pipeline structure).

It is crucial for riggers and animators to work hand in hand from the very beginning to form an understanding of how the characters move. For instance, designing creatures with six-jointed legs or 16 tentacles is no small task and has no room for assumptions and guesswork. It takes a full team effort and special rigging requests are almost always made during production that need to be addressed in a quick manner. Therefore, as with the matchmovers, riggers are an animator’s best friends.

Working with an Animation Team

Every character animator is different, and a comprehensive animation team brings an array of skill sets to every show. Whether it’s a small or large team, camaraderie and cohesion go a long way during increasingly tedious production schedules. Since all animators operate at different levels, depending on experience grade or work methods, it is important to cast shots and scenes accordingly within a show. For instance, some animators are better at subtle acting and emotional performance, whereas others may excel at executing broad action. During the course of production, every animator eventually must become familiar with every character in a show because that person may be called on in an instant to deliver a performance for a character he or she hasn’t been previously assigned. The goal is always to have everyone eventually work on an even keel. But until then, a well-balanced casting effort ensures performance continuity across an entire film.

Character and Environment Interaction

Very strict considerations need to be made for a character’s interaction in his/her/its environment and with props. First, the overall timing of the shot(s) needs to be examined. Is it a 5-, 8-, or 10-second action or sequence? Does it happen in one long take or a series of shots? What angles are most suitable for staging the action of the character—wide, medium, or tight close-ups? These questions are usually answered in the storyboard or previs phase, but more often than not, liberties are taken on set as the project and story evolve during the shooting phase.

Also taken into consideration is the characters’ effect on the objects or props they interact with. In the case of a chipmunk on a couch, there may be little or no compression or impression on the couch from the animal’s weight—depending on the object’s surface. But if the action calls for that chipmunk to leap from the couch to a table, then considerations would be made for the couch pillow to react to the force of the leap as well as the object on which the chipmunk lands. For example, when it lands does it move papers, trinkets, or keys? Does it knock over a vase and flowers? These props would be rigged practically on set, based on the timing of the move, or replaced digitally.

Suppose the chipmunk then leaps onto a bed. Then the cloth interaction, whether on a large or small scale, is an important detail of the action depending on how the cloth has to behave due to the character’s effect on it. These are very budget-conscious decisions. The same is true if there is a dinosaur, a gorilla, or behemoth killer robots rampaging through a forest, desert, or city. Admittedly, the stakes are higher at a much larger scale, but the same considerations as in the examples above are at play. How does the rampaging character affect the trees, sand, or concrete and glass edifices? How does that character move through the scene based on its physical size in relation to real-world physics? What needs to be timed and affected practically by the use of pyrotechnics, wind machines, etc.? Which scenes need digital props and enhancements? Again, these are script breakdown and budgetary determinations. But all are also defined by character integration into the live-action footage.

Animating to Dialogue and Sound Cues

Another important aspect in the integration process is animating characters to dialogue or voice tracks. In most cases, the voice actor is recorded during the principal photography phase, and those voice tracks are handed over to the animators. This is also a useful step in the character development stage, because it helps define a character’s personality and movements based on cadence, pronunciation, and inflection. An actor’s delivery is different in every take and those selected takes are what defines the character’s personality.

In some cases, even when a track is laid down and animation begins, ADR¹² sessions are required in which an actor will actually look at the performance of the animation in progress for a better performance match. And then there are the more extreme cases in which a voice or voices are recast while the animation is in progress, in which case that animation may or may not be affected. In CG character animation, the execution of dialogue points back to the character’s design—this time focusing on facial expression. Therefore the process used for animation phonemes, or sculpted syllabic mouth shapes, will be determined very early on. Is it “real” or is it stylized (cartoony)? Does it involve one rig or a series of rigged facial morphing targets or blend shapes? In the instance of sound cues, it all comes down to the ancient question of which came first, the chicken or the egg—does the animation drive the cue or vice versa? If a sound effects design pass is in progress, animators will often be given the unedited scratch tracks for timing cues. However, in most cases the animation does in fact determine where the sound cues will edit, and in those instances, the sound designers are given either director-approved finals or close to final animation footage.

Importance of Seeing Characters Projected and with Textures

Another very important step in combining 3D characters into a 2D image plane is seeing how the character performs with textures or lighting cues in reference to its environment in a scene. Many things in a character’s performance that are judged in quick-shade mode or low-resolution hardware textures are more or less apparent when a character is seen in its proper fully rendered lighting. Therefore the prelighting phase, or defining the character’s look in terms of color and surface texture, is a vital part of the production process. Some things may be pushed too extreme; others may not be exaggerated enough. These aspects are made evident by proper lighting. And since that phase is so memory intensive in terms of rendering time and disk space, the sooner it takes place the better.

This also holds true when it comes to the dailies process. Seeing the character projected on screen in sequence on a regular basis is an immeasurably important tool in the live-action integration process. It helps to answer all of the issues involving weight, proportion, and timing, as well as lighting and compositing, that are less evident even on a 24-inch desk monitor.

Conclusion

As stated at the beginning of this section, the steps laid out here serve only as a general overview to integrating character-animated visual effects within live-action projects. It is a constantly evolving process and therefore the practitioners of the medium are always in a state of learning or developing the advancements and inventions. Artists and technicians are working tirelessly toward improving the efficiency of the craft in all aspects and areas to enable the imaginations of filmmakers to broaden the entertainment experience and bring a better product to increasingly demanding audiences.

¹ McCay was also an important newspaper cartoonist for W.R. Hearst’s New York JournalAmerican.

² Called ink and paint.

³ Leica was the brand of lenses used on the animation cameras that photographed storyboards in the earliest Disney days.

⁴ Also called x-sheets or dope sheets.

⁵ Also referred to as MoCap or performance capture.

⁶ Stop-motion may even be older than 2D, but there are no confirmed dates to prove that.

⁷ Stop-motion models, being shot one still frame at a time, have no blurred elements, as do moving objects when filmed in real time, thereby giving the animated characters a sort of staccato effect to their movements when compared to the live-action plates in which they are positioned.

⁸ If the film is produced and fully financed by any of the major Hollywood studios, it is most likely self-guaranteed.

⁹ Very detailed physical sculptures.

¹⁰ Camera takes done without the actors.

¹¹ Pixelation is the single-framed photographic manipulation of live subjects via timed exposures or hand-maneuvered processes.

¹² ADR: automated dialogue replacement, which is basically a process in which a performer replaces his or her previously recorded dialogue by respeaking the lines in sync to picture or by recording new dialogue in sync to picture.