Chroma subsampling

What it means

Video is usually stored as one brightness channel and two color channels rather than as red, green and blue. Human vision is much more sensitive to brightness detail than to color detail, so codecs can store the color channels at reduced resolution without most viewers noticing.

The ratios describe how much color is kept. 4:4:4 keeps full color resolution. 4:2:2 halves the horizontal color resolution. 4:2:0, used by almost all consumer cameras, phones and streaming platforms, halves it both horizontally and vertically, storing a quarter of the color samples.

For normal viewing, 4:2:0 looks perfectly fine, which is why it became the standard for delivery.

How it works in practice

Subsampling matters most in post-production. Green screen keying, heavy color grading and fine graphics with colored edges all benefit from more color information. That is why professional cameras and intermediate codecs like ProRes offer 4:2:2 or 4:4:4.

For action footage, 4:2:0 combined with 10-bit color is usually a good balance. If you shoot a log profile and plan strong grades, 10-bit matters more than the subsampling ratio. Most action cameras record 4:2:0 in every mode, so there is often no choice to make anyway.

Screen recordings and graphics with thin colored text are where 4:2:0 shows its weakness most clearly.

What to watch out for

Converting a 4:2:0 file to 4:2:2 or 4:4:4 does not restore lost color detail. It only makes the file bigger. Choose higher subsampling at recording time, if the camera offers it, and keep everything else at the original quality.

Strong saturation boosts and sharp red or blue edges can reveal subsampling as blocky color fringes, especially after platform recompression. If bright red jackets or neon signs look jagged in your exports, ease off saturation a little and export at a solid Bitrate. Small text in contrasting colors suffers the same way, so keep titles bold and simple.