BT.2020 ITU-R BT.2020 (UHDTV Color Space)

What BT.2020 actually specifies

BT.2020 was published by the ITU on 23 August 2012 as Recommendation ITU-R BT.2020, defining the system colorimetry, image formats, frame rates, bit depths, chroma subsamplings, and transfer function for ultra-high-definition television with standard dynamic range and wide color gamut. The recommendation covers picture resolutions of 3840x2160 ("4K") and 7680x4320 ("8K"), progressive frame rates, sample bit depths of 10 or 12 bits per channel, RGB and luma-chroma signal representations with 4:4:4, 4:2:2, and 4:2:0 chroma sampling, color primaries, and an SDR opto-electronic transfer function. It is the SDR/WCG counterpart to Rec. 2100, which extends BT.2020 with HDR transfer functions.

The defining feature is the chromaticity triangle. BT.2020's primaries on the CIE 1931 xy chromaticity diagram are red (0.708, 0.292), green (0.170, 0.797), and blue (0.131, 0.046), with a D65 white point at (0.3127, 0.3290) and a correlated color temperature of approximately 6500 K. These primaries form a triangle much larger than Rec. 709 or DCI-P3. Sample encoding is either 10 or 12 bits per sample, with 4:4:4 full-resolution sampling supported alongside 4:2:2 and 4:2:0 chroma subsampling for luma-chroma signals.

Why the gamut is large and hard to hit

BT.2020 covers 75.8% of the CIE 1931 chromaticity area — roughly twice the Rec. 709 / sRGB coverage of 35.9% and substantially wider than DCI-P3 at 53.6%. Adobe RGB sits at 52.1% for comparison. The gain is concentrated in the green and red regions, where the BT.2020 triangle reaches the spectral locus.

That position on the locus is the catch. The BT.2020 primaries correspond to monochromatic light at 630 nm (red), 532 nm (green), and 467 nm (blue). To fully reproduce them, a display effectively needs three laser-pure or equivalent narrow-band emitters; conventional LED-backlit LCDs and even most quantum-dot panels emit broader spectra and cannot reach the corner of the BT.2020 triangle. No commercially affordable consumer display reaches 100% BT.2020 coverage in xy chromaticity. Quantum-dot LCDs with optimized color filters have demonstrated greater than 90% coverage in lab settings, while typical 2025-2026 high-end OLED and Mini-LED TVs measure roughly 75-80% BT.2020 — the same panels generally hit around 99% DCI-P3. Because the DCI-P3 sub-gamut sits well inside BT.2020, a high DCI-P3 score does not imply high BT.2020 coverage.

The native BT.2020 OETF is the same conventional power-law gamma curve used by Rec. 709 and is intended for SDR. BT.2020 itself does not define an HDR transfer function; HDR transfer functions are layered on top of BT.2020 chromaticity by Rec. 2100.

How BT.2020 shows up in HDR specs and TV reviews

Rec. 2100 (BT.2100) extends BT.2020 for HDR by reusing the BT.2020 color primaries and adding two HDR transfer functions: PQ (Perceptual Quantization, SMPTE ST 2084) and HLG (Hybrid Log-Gamma). HDR10, HDR10+, and Dolby Vision all use PQ + BT.2020 chromaticity; HLG broadcast HDR uses the HLG curve + BT.2020 chromaticity. So BT.2020 is the chromaticity container shared by every modern HDR format.

For HDR10 specifically, BT.2020 is mandatory. The HDR10 specification published in August 2015 requires the BT.2020 color space, the SMPTE ST 2084 (PQ) EOTF, 10 bits per channel, 4:2:0 chroma subsampling, and SMPTE ST 2086 mastering-display metadata plus MaxCLL/MaxFALL values. Every UHD Blu-ray and every HDR10 streaming title therefore signals BT.2020 primaries in its bitstream.

On the measurement side, reviewers report "Rec. 2020 coverage" as the percentage of the BT.2020 triangle on the CIE 1931 xy diagram that a display can address with its native primaries, alongside DCI-P3 coverage. A wide-color-gamut TV today generally posts about 95-99% DCI-P3 and roughly 70-80% Rec. 2020, with the xy-area metric being more conservative than "volume" coverage that includes luminance.

Common confusions

BT.2020 is the container, not the mastering gamut. Although HDR10, HDR10+, and Dolby Vision streams signal BT.2020 chromaticity, virtually all current HDR content is color-graded inside a P3-D65 sub-gamut and then carried in the BT.2020 container. Studios and Dolby explicitly recommend a Rec. 2020 color-space transform followed by a P3-D65 gamut limiter, because no consumer mastering monitor or consumer TV can faithfully reproduce true BT.2020 primaries. The BT.2020 envelope is forward-compatibility headroom, not the working space.

75-80% BT.2020 is a flagship number, not a failure. 100% BT.2020 is not currently achievable on consumer hardware — the primaries demand monochromatic light. A display reading around 75-80% BT.2020 is a flagship-tier wide-color-gamut panel, and 90%+ has only been demonstrated with advanced quantum-dot-on-color-filter prototypes. Buyers should not penalize a TV for missing "100% Rec. 2020," because no consumer TV achieves it and almost no content actually requires it.

BT.2020 does not describe the HDR signal curve. BT.2020's own OETF is the conventional gamma curve inherited from Rec. 709 and is intended for SDR. PQ (SMPTE ST 2084) and HLG are the HDR transfer functions, and they are defined by Rec. 2100 layered on top of BT.2020 chromaticity. A label like "HDR with BT.2020" is shorthand for "BT.2020 primaries + Rec. 2100 PQ or HLG curve" — the BT.2020 spec itself does not encode the HDR signal.

"4K BT.2020" tags the container, not the saturation. BT.2020 bundles resolution, frame rate, bit depth, chroma sampling, and a wide-gamut color volume into a single recommendation, but a stream tagged "BT.2020" merely declares the container. Most 4K HDR content actively occupies only the P3-D65 sub-gamut inside that container, so the tag tells you the chromaticity declared in the bitstream — not how saturated the actual graded pixels are.

BT.2020 and Rec. 2020 are the same document. Both "BT.2020" (the ITU-R series prefix BT for broadcasting service) and "Rec. 2020" (short for Recommendation 2020) refer to the same standard; manufacturer marketing tends toward "Rec. 2020," broadcast and spec literature toward "BT.2020." They are not separate standards.