< Previous | Contents | Next >
All media is output using a single data level, depending on your selection. There are three options:
— Automatic: The output data level of all clips is set automatically based on the codec you select to render to in the “Render to” drop-down menu.
— Video: All clips are rendered as normally scaled for video (10-bit 64–940).
— Full: All clips are rendered as full range (10-bit 4–1019).
For most projects, leaving this setting on “Automatic” will yield the appropriate results. However, if you’re rendering media for use by another image processing application (such as a compositing
application) that is capable of handling “full range” data, then full range output is preferable for media exchange as it provides the greatest data fidelity. For example, when outputting media for VFX work as a DPX image sequence, or as a ProRes 4444 encoded QuickTime file, choosing “Unscaled full range data” guarantees the maximum available image quality. However, it is essential that the application you use to process this media is set to read it as “full range” data, otherwise the images will not
look correct.
So, What’s the “Proper” Data Range for Output?
Strictly speaking, there is no absolutely “proper” data range to use when outputting image data. As long as the Levels setting of each clip in the Media Pool is set to reflect how each clip was created, your primary consideration is which data range is compatible with the media format or application you’re delivering to. If the media format you’re exporting to supports either normally scaled or full range, and the application that media will be imported into supports either normally scaled or full range, then it’s really your choice, as long as everyone involved with the project understands how the data range of the media is meant to be interpreted once they receive it.
Outputting to hardware is a bit trickier, in that you need to make sure that the external display or VTR you’re outputting to is set up to receive a signal using the data range you’ve chosen. If the device
is limited to only one data range, then you need to be sure that you’re outputting to it using that data range, or the levels of the image will appear to be incorrect, even though the image data being processed by Resolve is actually fine.
Introduction to DaVinci Resolve Color Management
How color is managed in DaVinci Resolve depends on the “Color Science” setting at the top of the Color Management panel of the Project Settings. There are four options: DaVinci YRGB, DaVinci YRGB Color Managed, DaVinci ACEScc, and DaVinci ACEScct. This section discusses the second setting, DaVinci YRGB Color Managed. ACEScc and ACEScct is discussed in the following section in this chapter.
Display Referred vs. Scene Referred Color Management
The default DaVinci YRGB color science setting, which is what DaVinci Resolve has always used, relies on what is called “Display Referred” color management. This means that Resolve has no information about how the source media used in the Timeline is supposed to look; you can only