# Chroma subsampling explained: 4:4:4 vs 4:2:2 vs 4:2:0

> What is chroma subsampling? Learn why 4:2:0 is fine for movies, why 4:2:2 rules cameras, and why 4:2:0 causes blurry text on external Mac displays.

Published 2026-10-16 · HDR & picture · https://www.hazenstudio.com/maclc/blog/chroma-subsampling/

## In short

- Chroma subsampling compresses video by reducing color resolution while keeping full brightness detail.
- Our eyes see fine detail mostly through brightness; we resolve color at a much lower resolution.
- 4:2:0 halves both horizontal and vertical color resolution, cutting video data rates by 50% without visible loss in films.
- On a computer monitor, 4:2:0 makes thin text look fuzzy and colored at the edges, because text needs sharp color edges at single-pixel scale.

Chroma subsampling keeps every pixel's brightness but stores color at a lower resolution, discarding up to 75% of the color data. Because we see fine detail mostly through brightness, films look just as sharp: almost every video you stream or buy is stored this way.

Computer screens are another story. If the connection to an external monitor falls back to 4:2:0, thin text gets blurry, colored fringes.

## What is chroma subsampling and how does the human eye perceive color?

Our visual system resolves fine detail mostly through brightness. Look at a photo of a city skyline: the sharp lines of windows and antennas come from changes in brightness. Blur the colors a little and the picture still looks sharp; blur the brightness and it doesn't.

Engineers built on this when color television arrived in the 1950s. Rather than transmitting three full-bandwidth color channels, broadcast systems learned to isolate brightness and compress the color data.

The following table summarizes the mathematical properties of the three primary subsampling schemes standardized in [Recommendation ITU-R BT.2100](https://www.itu.int/rec/R-REC-BT.2100):

| Sampling Format | Horizontal Chroma | Vertical Chroma | Samples per 2 × 2 Quad | Bandwidth vs 4:4:4 | Primary Applications | Text Quality on Desktop |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| **4:4:4** | 100% (full) | 100% (full) | 12 samples (4Y, 4Cb, 4Cr) | 100% (uncompressed) | Computer OS, text rendering, high-end VFX | Sharp, clean edges |
| **4:2:2** | 50% (half) | 100% (full) | 8 samples (4Y, 2Cb, 2Cr) | 66.7% (33% savings) | Camera recording, Apple ProRes, broadcast | Acceptable, minor softness |
| **4:2:0** | 50% (half) | 50% (half) | 6 samples (4Y, 1Cb, 1Cr) | 50.0% (50% savings) | UHD Blu-ray, Netflix, YouTube, HEVC, AV1 | Blurry, color fringing on fine lines |

## What is Y'CbCr and how is it different from RGB?

Computer graphics cards and display panels operate natively in the RGB (Red, Green, Blue) color model. In uncompressed RGB, every single pixel on screen contains three independent subpixel values. While RGB is straightforward, it is inefficient for video storage because the red, green, and blue signals are highly correlated; bright white, for example, requires maximum values across all three channels simultaneously.

To perform chroma subsampling, video systems convert RGB into a component color model called **Y'CbCr**:

1. **Y' (Luma):** The brightness component, calculated as a weighted sum of red, green, and blue after gamma pre-correction. Luma carries all the fine high-frequency edge detail, shadows, and textures.
2. **Cb (Chroma Blue):** The blue-difference component (B' − Y'), representing how much blue differs from luma.
3. **Cr (Chroma Red):** The red-difference component (R' − Y'), representing how much red differs from luma.

Green information does not need its own separate channel because it can be reconstructed mathematically at playback by subtracting Cb and Cr from Y'.

Once video is separated into Y', Cb, and Cr planes, compression engineers can leave the Y' luma plane completely untouched at full resolution while aggressively downsampling the Cb and Cr chroma planes. To understand how quantization bit depth layers on top of this, see [8-bit vs 10-bit color](/maclc/blog/8-bit-vs-10-bit-color/).

## How do 4:4:4, 4:2:2, and 4:2:0 actually work?

Chroma subsampling notation uses a three-number ratio (J:a:b), typically evaluated across a reference grid that is two pixels tall and four pixels wide (J = 4):

- **J (first number, usually 4):** The horizontal pixel width of the sampling block.
- **a (second number):** The number of chroma samples taken in the top row of J pixels.
- **b (third number):** The number of chroma samples taken in the bottom row of J pixels.

### 4:4:4 (No subsampling)
In 4:4:4 video, every pixel has its own unique luma, blue-difference, and red-difference sample. In a 2 × 2 quad of four pixels, there are 4 luma samples, 4 Cb samples, and 4 Cr samples (12 samples total). There is zero loss of color detail. Every single pixel can display an independent, sharp color distinct from its immediate neighbors.

### 4:2:2 (Horizontal subsampling)
In 4:2:2 video, chroma resolution is halved horizontally while remaining untouched vertically. In a 2 × 2 pixel quad, there are 4 luma samples, but only 2 Cb samples and 2 Cr samples (8 samples total). Two horizontally adjacent pixels share the exact same color information, cutting bandwidth by 33.3%.

This format is the industry standard for professional acquisition codecs such as Apple ProRes 422. Preserving full vertical chroma resolution gives video editors the clean horizontal boundaries needed to pull clean green-screen keys and perform color grading without edge artifacts.

### 4:2:0 (Horizontal and vertical subsampling)
In 4:2:0 video, chroma resolution is halved both horizontally and vertically. In a 2 × 2 quad of four pixels, there are 4 luma samples, but only 1 Cb sample and 1 Cr sample (6 samples total).

All four pixels in the 2 × 2 quad share a single averaged color value. This slashes the color data by 75% and reduces the total raw uncompressed bitstream volume by exactly 50% compared to 4:4:4.

## Why is 4:2:0 the universal standard for streaming and Blu-ray?

Nearly every consumer digital video format in existence uses 4:2:0 chroma subsampling. This includes:

- Physical Ultra HD Blu-ray and standard 1080p Blu-ray discs
- Commercial streaming services (Netflix, Apple TV+, Amazon Prime, Disney+, Max)
- Web video platforms (YouTube, Vimeo, Twitch)
- Broadcast digital television (ATSC 3.0, DVB-T2)
- Smartphone video recording (including iPhone Dolby Vision)

The reason 4:2:0 is universal is bandwidth efficiency. Halving the raw data rate of every frame before video compression even begins allows codecs like HEVC (H.265) and AV1 to dedicate their bitrate budgets to preserving sharp luminance details and avoiding macroblocking.

In real-world cinema, color boundaries in nature are soft and continuous. Skin tones blend smoothly into clothing, leaves blend into branches, and skies blend into clouds. Because these color shifts coincide with luma changes, human eyes cannot perceive the 50% reduction in color resolution when sitting at a normal couch distance. For an analysis of how data rates shape video quality, read [what is bitrate](/maclc/blog/what-is-bitrate/).

MacLC, the video player we make, plays 4:2:0 files from streaming and discs as well as 4:2:2 and 4:4:4 footage from cameras, such as ProRes, decoded in hardware where the Mac has the circuits for it.

## Why does 4:2:0 make small text fringe and blur on a Mac monitor?

While 4:2:0 subsampling is invisible in movies, it is disastrous for a computer operating system.

Computer user interfaces are constructed from synthetic, high-contrast, single-pixel geometry:

- Menu bar text is often only 1 or 2 pixels wide.
- Text edges are smoothed one pixel at a time, often against a contrasting background.
- Black text on a white background, or white text on dark backgrounds, creates extreme instantaneous contrast transitions across a single pixel boundary.

When a Mac sends 4:2:0 to a monitor, every block of 2 × 2 pixels shares one color. A one-pixel colored edge, like red text or a colored link on white, gets its color averaged with the neighbors.

The result is **text fringing**: colored text looks smudged, thin lines lose their crispness, and reading for long gets tiring.

## How do you fix chroma subsampling and text fringing on macOS?

If text on your external monitor looks soft or carries colored halos, your Mac is likely outputting a Y'CbCr 4:2:0 signal instead of full RGB 4:4:4.

It happens mostly over HDMI, when the link lacks bandwidth. HDMI 2.0 tops out at 18 Gbps, while 4K at 60 Hz with 10-bit color in full RGB needs a little more than that, so the devices fall back to 4:2:2 or 4:2:0 rather than show nothing.

To restore crisp RGB 4:4:4 text on macOS:

1. **Switch to USB-C / DisplayPort:** DisplayPort 1.4 and Thunderbolt connections offer significantly higher bandwidth than older HDMI ports, carrying full 4:4:4 RGB 10-bit at 4K 60Hz and 120Hz without compression.
2. **Use HDMI 2.1 end to end:** if your Mac's HDMI port supports HDMI 2.1 (check its tech specs), use a certified Ultra High Speed cable and a monitor input that supports it.
3. **Turn off HDR on limited monitors:** with an entry-level HDR monitor on HDMI 2.0, switching **High Dynamic Range** off in **System Settings › Displays** frees bandwidth for full RGB at 8 bits.
4. **Lower the refresh rate:** 4K at 30 Hz needs half the bandwidth of 60 Hz; fine for documents, not for video.

For details on codec architectures that process these signals, explore [H.264 vs H.265 vs AV1](/maclc/blog/h264-vs-h265-vs-av1/).

## What to do next

Chroma subsampling makes 4K streaming feasible over consumer home broadband. In films and television, 4:2:0 cuts bandwidth in half without degrading visible picture quality.

On the desktop, you want full RGB for sharp text: connect external screens over DisplayPort, Thunderbolt or HDMI 2.1. For video, 4:2:0 is the norm, and any good player handles it; MacLC is [free to download](/maclc/donate/).

## Questions people ask

### Can you see the difference between 4:2:0 and 4:4:4 in movies?

No. In live-action film and animation, natural color transitions align with brightness changes. Because the human eye is far less sensitive to fine color detail, 4:2:0 video looks virtually identical to uncompressed 4:4:4 video at normal viewing distances.

### Why does text look blurry or have red fringes on my Mac external monitor?

When an HDMI link lacks the bandwidth for full RGB at the chosen resolution, refresh rate and bit depth, the Mac and the monitor can settle on Y'CbCr 4:2:2 or 4:2:0. Then pixels share color samples and colored text edges smear.

### What chroma subsampling does Netflix, YouTube, and 4K Blu-ray use?

Virtually all commercial consumer video delivery—including 4K Ultra HD Blu-ray discs, Netflix, Apple TV+, Disney+, and YouTube—is encoded strictly in 4:2:0 chroma subsampling to save transmission bandwidth.

### Why do professional cameras record in 4:2:2 instead of 4:2:0?

Cameras record in 4:2:2 (such as Apple ProRes 422) because visual effects workflows like green-screen chroma keying and heavy color grading require clean horizontal color edges to separate subjects from backgrounds cleanly.

---
MacLC is a free, open-source video player for Apple Silicon Macs: https://www.hazenstudio.com/maclc/ · download: https://www.hazenstudio.com/maclc/donate/