MaxFALL

Definition & Measurement

MaxFALL is an abbreviation for Maximum Frame Average Light Level, a static HDR metadata value expressing the highest average brightness across any single frame in an entire video stream or file. Unlike peak brightness metrics, MaxFALL represents a frame-level average rather than a single-pixel measurement.

MaxFALL is measured and expressed in nits (candelas per square meter, cd/m²), a standardized unit of luminance that quantifies brightness. A single MaxFALL value applies to an entire video stream; it does not change frame-to-frame.

Calculation Method

MaxFALL is calculated by a specific algorithm: for each frame in the video, the brightness of all pixels is averaged, taking only the maximum color component (red, green, or blue) of each pixel. The highest average across all frames in the content becomes the MaxFALL value for the entire stream.

This calculation method is formally defined in CEA/CTA-861.3, the standard that specifies HDR static metadata encoding. The algorithm's normative specification appears in CEA-861.3 Annex A, though the complete calculation formula is not published in freely accessible technical documents.

MaxFALL vs. MaxCLL

MaxFALL differs fundamentally from MaxCLL (Maximum Content Light Level), which measures the brightness of the single brightest pixel in the entire content. Because MaxFALL averages brightness across all pixels in a frame before selecting the highest frame, MaxFALL values are typically significantly lower than MaxCLL values.

A documented example illustrates this relationship: a mastering environment (Sony BVM-X300, a 1,000-nit reference display) produced content with MaxCLL = 1,000 nits (the peak pixel brightness) while MaxFALL = 180 nits (the average of the brightest frame). Both values are required in HDR10 metadata and work together to describe the content's brightness profile.

Role in HDR10 and Standards

MaxFALL and MaxCLL are static HDR metadata values commonly paired with SMPTE ST 2086 (Mastering Display Color Volume) in HDR10 content. The formal data encoding for MaxFALL and MaxCLL is specified in CEA/CTA-861.3, which defines how these values are transmitted in InfoFrames over HDMI and other digital video connections.

While HDR10 is technically capable of peak brightness up to 10,000 nits, common HDR10 content is mastered with peak brightness between 1,000 and 4,000 nits. MaxFALL values in real-world graded content vary depending on the mastering display's brightness capabilities and the content's creative intent; one documented example measured MaxFALL at 22 nits for a specific graded clip, though no comprehensive survey of MaxFALL across different content categories has been published.

Display Application & Tone Mapping

Displays use MaxFALL and MaxCLL metadata to perform tone mapping, a process that compresses or reshapes the HDR signal to fit the display's actual brightness capabilities. If a film was mastered with peak brightness at 4,000 nits but the viewer's display only reaches 800 nits, these two metadata values guide the display's tone-mapping algorithm. They shape the brightness curve so the brightest highlights fit the display's range without clipping to pure white.

MaxFALL is particularly relevant to the management of sustained high-brightness content on emissive displays such as OLED panels, which can degrade if pixels are run at maximum intensity continuously. By informing the display's tone mapping, MaxFALL helps displays manage average frame brightness and prevent prolonged operation at the edge of their safe operating envelope.

Static vs. Dynamic Metadata

MaxFALL is static metadata, meaning a single value represents the entire video stream rather than adapting frame-to-frame. This differs from dynamic HDR metadata formats such as HDR10+, which can adjust metadata on a scene-by-scene or even frame-by-frame basis. Static metadata simplifies transmission and implementation at the cost of less scene-specific adaptation.