Every shot in a CG production is built from assets created primarily by the Modeling and Setup departments. These assets can include characters, props, environments, vehicles, and many other elements.
The initial asset list is often identified during storyboarding and refined during layout. However, asset requirements can continue to evolve throughout production, even after rendering.
Because every department relies on knowing exactly what is present in a scene, maintaining a clear and accurate casting list is essential. The level of detail you track will depend on your production needs, available resources, and pipeline maturity. The following sections describe three common levels of asset casting and explain when each is most useful.
Level 1: List of Assets Present in the Shot
The most basic approach is to maintain a simple list of all assets that appear in a shot.
This information is particularly valuable for production managers because it helps them:
- Identify which shots are affected when an asset changes
- Estimate the overall importance and usage of an asset throughout the production
- Assess the impact of asset-related delays or revisions
Asset casting is also useful for Pipeline TDs. With this information, they can build automated scene assembly tools that import all required assets for a shot. Pipeline tools can also assist production teams in generating and maintaining these lists to reduce manual efforts.
Example: Shot 01 contains Agent327, SuperEvil, Gun, Cars, Street
Level 2: List and Quantity of Assets
The next level of detail is to record both the asset type and the number of times it appears in the shot.
While this information may have limited value for production management, it significantly improves scene-building workflows by allowing tools to instantiate the correct number of assets automatically. It also gives artists a better understanding of scene complexity before opening the shot.
Example: Shot 01 contains Agent327 (1), SuperEvil (1), Gun (1), Cars (3), Street (1)
Level 3: List of Asset Instances
The most detailed approach is to track every individual asset instance in a shot.
Because maintaining this level of detail requires additional effort, it is best suited to productions that have the time, resources, and pipeline support to manage it properly.
Tracking individual instances provides several advantages:
1. Track Instance-Specific Work
Individual asset instances often require modifications that are unique to a particular shot. For example, a character may need a damaged model, a custom rig variation, or shot-specific setup adjustments.
By tracking instances separately, production teams can identify which assets generated additional work, monitor the status of shot-specific modifications, and improve visibility into production costs and dependencies.
2. Enable Instance-Based Processing
Storing assets as individual instances allows pipeline tools to generate, cache, and import data at the instance level.
This can significantly reduce processing time because only modified instances need to be updated, rather than rebuilding the entire scene.
Example: Shot 01 contains Agent327-1-Wounded, SuperEvil-1, Gun-1, Car-1-Blue, Car-2-Red, Car-3-Broken, Street-1
While instance-level tracking provides substantial benefits for automation, optimization, and communication, it also increases production complexity. Before adopting this approach, make sure the additional maintenance effort is justified by the value it brings to your pipeline!
Final Thoughts
This article focuses on casting at the shot level, but the same principles can be applied at higher levels of production like sequences, episodes, or entire projects. This is especially useful in episodic productions where asset usage needs to be tracked across multiple episodes.
In many cases, a basic asset list is sufficient. But productions that require more advanced automation, dependency tracking, or performance optimization benefit from recording quantities or individual instances.
Although asset casting seems straightforward, maintaining accurate casting information is a time-consuming task.
In any case, it remains one of the most important datasets shared across departments because it directly affects planning, communication, and pipeline automation.
For a related topic, consider reading our article on asset management and dependencies. Understanding how assets interact throughout production is key to building a pipeline that scales efficiently and reduces costly rework.
