Previz: Why Run Through It Before Shooting

Before the official shoot begins, professional crews usually do one thing: they run through the entire scene in a low-cost way first. This stage is called previz. Its output is not the final image, but a rough, movable schematic version—using simple 3D models instead of actors and basic spaces instead of real sets—to verify “whether this scene can be shot this way.” It may sound unnecessary, but it is one of the most effective ways to control costs.

What exactly does previz verify?

Previz is not meant to answer “does the image look good?” but several more fundamental questions.

1. Do the spatial relationships work?

Where the characters stand, how much distance there is in front and behind them, where the camera can be placed to see whom—these are hard to judge in text, but become clear as soon as you test them in 3D space. Blocking that reads smoothly in a script often reveals problems once placed into a concrete space, such as “this position blocks the lead character” or “there is nowhere to put the camera.”

2. Do the shots connect smoothly?

A single shot may look fine, but the sequence may not work at all when the shots are connected. If a character is on the left side of the frame in one shot and suddenly jumps to the right side in the next, the audience instantly loses their sense of orientation. Previz plays multiple shots together in sequence, and these problems are exposed immediately.

3. Is the rhythm right?

The length of a scene, the pauses between lines, and how much time an action requires are all easy to misjudge on paper. Once they are made into a movable previz version, the sense of rhythm becomes directly perceptible.

4. Are there enough camera positions?

How many camera positions a dialogue scene needs, what each camera position covers, and how often to cut between them—previz can first arrange camera positions with simplified models to confirm whether the coverage is sufficient.

Using 3D scenes for previz

Traditional previz requires a dedicated team and software, so the cost is not low. Now it can also be completed directly inside a creation platform. NadouPro’s 3D director console provides exactly this capability: building 3D scenes, placing character models and cameras, and creating or exporting animation. It can be broken down into several stages:

1. 3D scene construction: organize scenes and 3D assets in the director console, and block out the space first. The goal is not fine detail, but correct structure—where the door is, where the wall is, and where the main activity area is.
2. Character model placement and posing: place character models in the 3D scene and adjust their poses. The characters do not need to look good, but they must express “where they stand, which direction they face, and what posture they are in.”
3. Camera placement and camera position setup: place cameras in the 3D scene and design camera positions. This is the most valuable part of previz—you can directly see “what is in the frame when shooting from this position,” instead of relying on imagination.
4. Multi-camera positions and animation timeline: organize multiple camera positions and create animated previz on the timeline. Link blocking, actions, and camera-position switches together to form a complete schematic version of the scene.

After this is done, you get a playable low-cost version. It is not polished, but it is enough to judge whether the scene’s blocking works.

Three practical benefits of previz

• Benefit 1: Expose problems early.

If you discover in 3D previz that “this camera position cannot capture the lead character,” the cost of changing it is almost zero; if you only discover it after formal generation or even shooting, the cost of changing it may be dozens of times higher. The earlier a problem is found, the lower the cost—this is the core value of previz.

• Benefit 2: Align the team’s understanding.

Text descriptions easily lead to different interpretations. When a director says “a shot with a strong sense of oppression,” the cinematographer may understand it as a low-angle shot, the art team as a cramped space, and the actor as a close-up. A previz version visualizes the understanding, so everyone sees the same thing and the discussion has a shared basis.

• Benefit 3: Provide clear reference for subsequent generation.

After previz is complete, the camera position, angle, and character position of each shot have already been determined. This information can go directly into the storyboard and become the basis for later material generation, instead of being described again in text. In effect, this means finishing the “uncertain creative decisions” in advance, so later execution becomes “implementing a known plan.”

When is previz worth doing?

Previz has a cost, so not every project needs it. It is clearly worthwhile in several cases:

• Complex multi-character blocking: in scenes with more than three people, spatial relationships can easily become confusing
• Action scenes: blocking, chases, and fights are hard to judge on paper in terms of rhythm and space
• Key shots: the most important shots in the entire film are worth verifying first
• Multi-person collaborative projects: when understanding needs to be aligned, previz is the most effective communication tool

Conversely, simple one-person dialogue scenes or lyrical shots with a fixed camera can usually go straight into storyboarding.

Common misconceptions

• Misconception 1: Treating previz as the finished film. Previz is a verification tool, and rough visuals are normal. If you pursue beautiful previz, costs will get out of control.
• Misconception 2: Making only single shots without looking at the whole. The value of previz lies in continuous playback; looking at a single frame will not reveal shot-connection problems.
• Misconception 3: Skipping previz and making materials directly. For complex blocking, this has a high probability of rework.
• Misconception 4: Throwing previz away once it is done. The camera-position information determined in previz should be written into the storyboard as the basis for later execution.

Finally

The essence of previz is to use a low-cost method to consume uncertainty in advance. It does not produce the final image, nor does it directly enter the finished film, but it determines the amount of rework in every later stage. In the production process, it is a typical step that “looks unnecessary, but actually saves money.” For AI film and video projects, the value of previz may be even greater than in traditional production. That is because the main cost of AI generation is not money, but the time spent trying again and again—and previz is exactly what makes it possible to complete “trial and error” at the cheapest point.

Contains AI-generated information that may be incorrect. Please verify independently.