CRI and TLCI Explained: Color Accuracy of Lights
Author: Fiat Lux Guides Editorial Team Reviewer: Fiat Lux Guides Editorial Team Most recently reviewed First published Filed under Color and White Balance
CRI (Color Rendering Index) measures how accurately a light renders eight pastel color samples to the human eye, while TLCI (Television Lighting Consistency Index) simulates how a broadcast camera sensor and display reproduce 18 ColorChecker patches. For video and LED lighting, TLCI and extended CRI metrics like R9 give a clearer picture of skin-tone accuracy than CRI Ra alone.
Belongs to the guide: Color Temperature and White Balance Explained
Checked by Fiat Lux Guides’ editorial desk against the cited sources.
When you compare continuous LED lights for photography or video, two lights with the exact same Kelvin rating can still make skin tones look completely different. That happens because Kelvin values (see color and white-balance fundamentals) only describe the white point of a light—not how completely it reproduces all the colors in a scene.
To measure how faithfully a light reveals colors, manufacturers publish CRI and TLCI ratings. Understanding what each number actually tests—and where each metric falls short—helps you avoid muddy reds, green skin casts, and hours of frustrating color correction.
What is CRI (Color Rendering Index)?

CRI (Ra) is a 0-to-100 scale defined by the CIE (CIE 13.3-1995) that measures how accurately a light source renders eight pastel test colors to the human eye compared to a reference illuminant. A score of 100 means the light renders those eight samples identically to an incandescent blackbody radiator (below 5,000 K) or a standard daylight spectrum (5,000 K and above).
However, as DOE materials (Naomi Miller of PNNL, and Yoshi Ohno of NIST in his color quality overview) explain, standard CRI (Ra) has two well-known blind spots for modern LED lighting:
- It only averages eight muted pastel samples (
R1–R8). The DOE deck states that CRI “does not include saturated colors” and that an R9 value is used in addition to give a better estimate of red rendering. The extended samples R9 to R12 (saturated red, yellow, green, blue) are not part of the headlineRanumber, so a light can score a high Ra while rendering reds poorly. - It models human vision, not camera sensors. Two lights with spiky, discontinuous LED spectra can look acceptable to the human eye yet trigger unexpected hue shifts on a camera’s red, green, and blue Bayer filters.
For photography and video, R9 (saturated red) is especially important because human skin tones depend heavily on red wavelengths. Whenever a manufacturer publishes extended CRI scores, checking R9 alongside Ra tells you far more about how skin will look than Ra alone.
What is TLCI (Television Lighting Consistency Index)?
TLCI is a 0-to-100 color accuracy metric developed by the European Broadcasting Union (EBU Tech 3355) specifically to predict how a camera—not a human eye—sees a light. Instead of modeling human cone cells, TLCI simulates an entire broadcast camera chain:
- 18 colored test patches from a standard ColorChecker chart (excluding the six grayscale patches, which are included in the related TLMF-2013 matching metric).
- A standardized camera model with defined RGB spectral responsivity curves, a linear color matrix, and gamma correction.
- A standardized display model using ITU-R BT.709 / sRGB primaries, computing color differences (
ΔE*) in CIEDE2000.
Because TLCI tests saturated ColorChecker patches through a camera sensor model, it catches the exact spectral gaps that cause colorists headaches in post-production.
What do TLCI scores mean in practice?
EBU Tech 3355 calibrates its TLCI (Qa) scale so that Qa = 50 represents a typical daylight fluorescent tube—the historical watershed between lights that can be color-corrected acceptably and lights that cannot. In the EBU’s subjective testing with professional colorists and live broadcast engineers:
TLCI Score (Qa) |
Practical Meaning (paraphrasing the EBU film-style scale) |
|---|---|
| 85–100 | High color fidelity; little to no color correction needed for film or live broadcast |
| 75–85 | Good performance; minor colorist adjustments may be needed to match other lights |
| 50–75 | Moderate errors; requires noticeable color-grading work and may struggle when intercut |
| Below 50 | Poor camera color rendering; difficult or impossible to correct cleanly in post |
Keep in mind that EBU Tech 3355 also notes a caveat: the calculation is only truly credible when the distance from the luminaire’s chromaticity to that of its correlated color temperature is below 1 (per CIE Tech 15:2004). If a fixture has a strong green or magenta tint offset (covered in Kelvin vs tint), white balance and color accuracy suffer even before spectral gaps come into play.
CRI vs TLCI: Which matters more for your setup?

Use TLCI first for video and digital photography, and check CRI (especially R9) when TLCI is not listed. Because digital cameras—whether mirrorless cameras, cinema cameras, or webcams—filter light through RGB color arrays rather than human cones, TLCI aligns much more closely with what ends up in your footage.
That said, neither CRI nor TLCI is a complete guarantee on its own:
- Camera sensors vary: TLCI uses one standardized broadcast camera curve, whereas your specific camera may have slightly different red or green filter sensitivity.
- Newer metrics exist: As described in Ohno’s DOE/NIST presentation, CIE 224 (
Rf) and IES TM-30 expand testing to 99 color samples plus a gamut index (Rg) to show whether colors are desaturated or oversaturated. Meanwhile, fixtures like the Aputure Nova P300c also publish SSI (Spectral Similarity Index)—comparing the light’s actual spectrum directly to D55 daylight or tungsten—to predict performance across different cinema cameras.
Can high CRI or TLCI replace proper white balance?
No—high CRI and TLCI ensure a light contains accurate color information, but you still need to set the right white balance. Even a 98 CRI / 98 TLCI studio light will look orange or blue if your camera’s white balance is set to the wrong Kelvin value, or if you mix it with window light and room lamps of a different color temperature.
For best results, match your lights using our Kelvin reference chart, dial in a custom white balance with a gray card, and check our editorial policy for how we verify manufacturer and standards-body specifications.
Common questions
What is a good CRI and TLCI score for photography and video?
For video and color-critical photography, look for both CRI and TLCI scores of 90 or higher, with professional cinema LEDs commonly rated at 95 or above. In EBU Tech 3355, TLCI scores near the top of the 0–100 scale indicate minimal color-correction work for a camera colorist, whereas scores around 50 or below signal serious color errors.
Why can a light have a high CRI but still look wrong on camera?
Two reasons: first, standard CRI (Ra) averages only eight muted pastel swatches and ignores deep red (R9), which is vital for human skin tones. Second, CRI models human vision rather than digital camera RGB color filters, so a spiky LED spectrum can fool the human eye while causing color shifts on a sensor.
What is R9 in CRI ratings?
R9 is the ninth extended test color sample in the CIE CRI system, representing a saturated red. Because standard CRI (Ra) only averages samples R1 through R8, a light can score 90+ Ra while having a very low or even negative R9 value, which makes skin tones, lips, and warm reds look dull or muddy.
What is the difference between TLCI and CRI?
CRI (CIE 13.3-1995) evaluates how 8 pastel samples look to a human observer compared to a reference illuminant. TLCI (EBU Tech 3355) evaluates 18 ColorChecker patches through a standardized broadcast camera spectral response, linear matrix, and gamma curve to predict how accurately a camera will record colors.
What are IES TM-30 and SSI?
IES TM-30 (and CIE 224 Rf) evaluates color fidelity across 99 test samples along with a gamut index (Rg) to show oversaturation or undersaturation. SSI (Spectral Similarity Index, developed by the Academy) compares a light's full spectrum directly against a reference like D55 daylight or tungsten, which helps predict how different camera sensors will respond.
What we checked
- EBU Tech 3355: Method for the Assessment of the Colorimetric Properties of Luminaires (TLCI-2012), checked 2026-11-14 (tech.ebu.ch)
- U.S. DOE / PNNL (Naomi Miller): There's something about RED, TM-30 in practice (LightFair 2017), checked 2026-11-14 (energy.gov)
- U.S. DOE / NIST (Yoshi Ohno): Color Quality Metrics and Spectral Design of SSL Sources, checked 2026-11-14 (energy.gov)
- Aputure Nova P300c Kit Official Specifications, checked 2026-11-14 (aputure.com)
You might also like
- Color Temperature and White Balance Explained
Learn Kelvin color temperature, camera white balance presets, green/magenta tint, CRI vs TLCI, and how to prevent mixed-lighting color casts.
- Custom White Balance with a Gray Card
How to set custom white balance in-camera with a gray or white card for accurate skin tones under LEDs, window light, and mixed lighting.
- Kelvin vs Tint: Fixing Green and Magenta Color Casts
Learn the difference between Kelvin color temperature and green-magenta tint (Duv), why LEDs shift green, and how to fix color casts.
- Lighting Gels: CTO and CTB for Color Matching
Match tungsten, daylight, and LED sources using CTO and CTB lighting gels, mired shift math, and light transmission loss.
- Matching LED Lights to Daylight
How to match LED video and photo lights to window daylight using Kelvin presets, mired shifts, and camera white balance without mixed color casts.
- Mixed Lighting and Color Casts: Why Skin Looks Orange or Blue
Fix orange and blue color casts caused by mixing window light, household lamps, and LEDs. Learn how to match light color temperatures on set.