Pre-production is the planning stage where the story, characters, environments and shot list are defined before modeling or animation starts. Its purpose is to reduce expensive changes later, because a shot revised after it has been rigged, animated and rendered can cost many times more than one adjusted on paper. Teams produce scripts, concept art, animatics and style frames to agree on tone and staging. For example, a short film team may discover in an animatic that a camera angle hides the key expression, and fixing it there costs minutes instead of days of re-rendering. Strong pre-production therefore gives animators a clear target and keeps production on schedule and budget.
A storyboard is a sequence of drawn panels that shows the key moments, camera framing and action of each scene. An animatic adds rough timing to these panels by editing them together with a temporary soundtrack, so the director can judge pacing and duration before full animation. Animators use this material to understand what each shot must accomplish, such as a character reacting to a surprise or a vehicle crossing the frame. Because the animatic is cheap to change, problems with comedic timing, unclear staging or excessive shot length are caught early. Many professionals treat the animatic as the blueprint that guides every later decision about performance and camera movement.
Timing refers to the number of frames an action takes, while spacing describes how the object's position changes across those frames. Together they control weight, speed and emotion. A ball dropped in twelve frames with even spacing looks mechanical, but the same drop with slow starts and fast landings suggests gravity and a real object. Animators often plan using frame counts at a set frame rate, so a two-second action at twenty-four frames per second needs forty-eight frames. Small changes in spacing can transform a character from calm to frantic. Mastering timing is therefore central to making motion read clearly and convincingly to an audience.
Anticipation is a preparatory movement that signals an action before it happens, such as a character crouching before a jump. Follow-through and overlapping action describe how parts of a body or object continue moving after the main motion stops, such as hair or a coat swinging after a turn. These principles reflect the physical reality that masses resist sudden changes and lag behind the body. Without them, motion looks stiff and unnatural. A practical example is a character who throws a punch: the wind-up, the strike, and the arm's gradual settling each contribute to a sense of effort and weight. Animators apply these ideas to nearly every performance, from subtle facial gestures to large action sequences.
Squash and stretch changes an object's shape while preserving its approximate volume, to suggest material flexibility, impact and speed. A bouncing ball flattens on landing and elongates during fast movement, which helps the viewer perceive force. Animators apply it with care because exaggerating the effect can make a rigid object look like rubber or distract from the performance. In 3D work, this is often achieved through shape keys, scale controls or deformation rigs rather than redrawing each frame. A realistic use is a cartoon character whose face compresses slightly when it hits a wall. The technique is most effective when the degree of distortion matches the story's tone and the object's intended material.
Character design balances visual appeal with the practical needs of animation. Designers consider silhouette, which should read clearly against any background, as well as proportions, expressive features and the range of poses the character must adopt. A clear silhouette helps the audience recognize the character even when it is shown in shadow or at a distance. The design must also support the rig, so joints, facial regions and clothing must be planned to deform without tearing or collapsing. For example, a character with very thin limbs may be easier to draw but harder to deform cleanly during a crouch. Good designs therefore serve both the story and the technical pipeline, making performance possible without excessive repair work.
Modeling for animation requires that a mesh bends, stretches and deforms without losing its form. Topology, the arrangement of polygons or edges, is therefore critical. Clean, mostly quadrilateral geometry with edge loops placed around joints and facial features allows smooth deformation, while messy triangles or poles in the wrong places can cause pinching and visual artifacts. Static models can be dense and irregular because they are never moved. A studio might require characters to be built with predictable edge flow so that rigs and blend shapes perform reliably across many shots. Environments and props also need thought, since objects that will be broken, opened or moved need appropriate pivot points and separated parts to support the intended motion.
A keyframe records the value of a property, such as position, rotation or a facial control, at a specific frame. Animators set keys at the extreme poses and important moments of an action, such as the start, peak and end of a head turn. Choosing key poses carefully is central to the workflow, because the key poses define the story beats and shape of the motion before the in-between frames are refined. A common approach is to block the action with a few strong keys, check the silhouette and timing, and then add detail. A realistic example is a character jumping, where keys might capture the crouch, the apex and the landing. Good key selection makes later refinement faster and more coherent.
Interpolation is the process by which software calculates values between two keyframes. Animators control this through curves, typically shown in a graph editor, which determine how quickly a value changes over time. Linear interpolation produces constant speed and often looks mechanical, while smooth or ease-in and ease-out curves create gradual acceleration and deceleration that feel more natural. Adjusting tangent handles lets animators shape each transition precisely. For example, a door opening may use ease-in at the start and ease-out at the end to suggest a person pushing and then letting it settle. Because curves control rhythm and energy, professionals often spend substantial time refining them to achieve clarity and expressive motion.
Forward kinematics, or FK, sets the rotation of each joint in a chain in sequence, from the root outward, so a shoulder, elbow and wrist are posed one by one. Inverse kinematics, or IK, places an end effector such as a hand or foot at a target position, and the system solves the joint angles automatically. FK offers direct control and suits sweeping arcs or natural swinging motion. IK is useful when the end of a limb must stay in contact with a surface, such as a foot planted on the ground during a walk. Many rigs allow animators to switch between the two, which lets them choose the more efficient method for each part of a performance.
A rig is the internal control system that allows an animator to pose and move a model in a consistent, manageable way. It typically consists of a skeleton of joints, the skinning or weighting that links the mesh to those joints, and a set of controls such as handles, sliders and constraints. The skeleton defines the hierarchy of motion, while skin weights determine how strongly each joint influences nearby vertices. Controls provide an intuitive interface, so the animator can move a hand or raise an eyebrow without manipulating dozens of underlying parts. A well-designed rig hides complexity and gives the animator reliable, expressive tools, which is why rigging is treated as a specialized discipline within production.
Facial rigs allow animators to control expression through a combination of joints, blend shapes and driven controls. Blend shapes, also called morph targets, store alternative versions of the face, such as a smile or a brow raise, and blend between them according to a weight value. Joint-based systems are useful for jaw and eyelid movement, while corrective shapes fix problems that occur when multiple controls act together. A realistic example is a character whose eyes narrow while the cheeks lift, producing a convincing smile. Because the face carries most of a character's emotional communication, studios invest heavily in these systems, often refining them across many productions to achieve nuanced and reliable performances.
Constraints link one object's transform to another, for example making a hand follow a prop or keeping an eye aimed at a target. Automated controls, such as space-switching or procedural secondary motion, can produce consistent results across many shots while reducing manual keying. These tools save time and improve consistency, especially in crowd or background work. However, they can also introduce problems if the constraint order or hierarchy is misunderstood, causing unexpected flips, popping or lag. Experienced riggers document how each system behaves and provide overrides so that animators can intervene when a shot requires precise control. The aim is to automate routine behavior while preserving the creative freedom needed for expressive performance.
Lighting directs the viewer's attention, establishes time of day and emotional tone, and separates forms from their backgrounds. Common approaches include a key light that defines the main shape, a fill light that softens shadows and a rim or back light that outlines the subject. Color temperature also communicates mood, with warm tones often suggesting comfort and cool tones suggesting distance or tension. In animation, lighting must also support movement, because moving characters can pass through shadows and highlights that reveal their volume. A practical example is a nighttime chase scene in which a cool key and warm practical lights create contrast that helps the audience follow the action. Lighting is therefore a storytelling tool as much as a technical step.
Rendering converts a 3D scene into final images by simulating light, materials and camera behavior. Offline rendering uses path tracing or similar methods to produce highly accurate images, but each frame may take minutes or hours to compute. Real-time rendering uses rasterization and optimized techniques to display images instantly, enabling interactive feedback and previews. Studios choose between them according to the required fidelity, budget and schedule. For example, a feature film often relies on offline rendering for final frames, while a game cutscene or virtual production stage may depend on real-time engines. Understanding these trade-offs helps teams plan their pipelines and decide where to accept lower quality for speed.
Materials define how a surface responds to light, controlling properties such as color, roughness, reflectivity, transparency and subsurface scattering. Shading is the computation that applies these properties to the surface as the scene is rendered. Together, they establish whether an object looks metallic, skin-like, glossy or matte. In stylized animation, artists may deliberately simplify shading to produce flat or painterly results, while realistic work seeks physically plausible responses. A practical consideration is consistency, since characters, props and environments must appear to share the same visual language. Teams often create look development standards and test materials under the lighting used in the final shots.
A production pipeline is the ordered sequence of stages through which a shot or asset passes, typically including design, modeling, texturing, rigging, animation, simulation, lighting, rendering and compositing. Defined handoffs specify what each department delivers, in which file format and at what quality, so the next team can work without repeating effort. Pipelines are supported by asset management systems, version control and naming standards. Without them, a rigging update may silently break animation already completed, or a lighting team may receive files that cannot be loaded. Clear pipelines therefore protect schedules and reduce costly rework, especially on large productions with many contributors across time zones.
Simulation uses physics-based calculations to generate motion for elements that are difficult to animate by hand, such as cloth, hair, water or smoke. In a typical workflow, the animator blocks the character's core movement, and the simulation artist then adds secondary dynamics driven by that movement. Simulations are usually cached, meaning their results are stored so that lighting and rendering teams can use them without recalculating. Artists tune parameters such as stiffness, damping and collision settings to achieve believable results. A practical example is a character's coat that should swing during a run but must not pass through the legs. Careful integration keeps simulated elements consistent with the performance and avoids unexpected visual conflicts.
Review sessions give directors, supervisors and peers the opportunity to assess a shot against its goals, usually by viewing playblasts or rendered previews. Effective feedback is specific, tied to the story and framed around observable problems, such as a pose that obscures the expression or a beat that arrives too late. Animators often receive notes in a structured format and track revisions so that each round addresses prior concerns. Many studios set a fixed number of review stages, from rough blocking through polish, to control the cost of iteration. A realistic example is a supervisor noting that a character's head turn needs a stronger anticipation, which the animator then adjusts and presents again. The process keeps teams aligned while preserving creative focus.
Quality control verifies that a shot is technically sound and visually consistent before it moves to the next stage. Common checks include confirming that geometry does not intersect unintentionally, that joints deform cleanly, that timing matches the animatic and that the camera framing supports the action. Technical checks also cover frame rate, file integrity, correct asset versions and render settings. Teams frequently use automated scripts to flag issues such as missing textures or unbaked simulations, supplemented by human inspection. A small error, like a foot sliding during a walk cycle, can be noticeable to viewers, so these reviews protect the final image. Quality control is therefore a repeatable safeguard built into the pipeline rather than a last-minute check.