Naresh Varnakavi, Ph.D.MATERIALS SCIENTIST
← All ToolsColorimetry & Lighting

PHOTONICS / COLOR SCIENCE

CIE Color Coordinates & CCT Calculator

Calculate CIE 1931 x,y and CIE 1976 u′,v′ color coordinates, correlated color temperature (CCT), Duv and color gamut from emission spectra or coordinates. Explore white points, color rendering, LED efficiency and biological weighting with the measurements each method requires.

Your 1° viewing field

Chromaticity calculations use the CIE 1931 2° standard observer, which ISO/CIE specifies for fields of about 1°–4°. The official color-matching functions are tabulated every 1 nm from 360 to 830 nm. Rendering metrics and biological weighting use the observers and action spectra prescribed by their respective methods, as documented below.

01 / INPUT

Start with what you have.

All calculations run in your browser. Your uploaded spectra are not sent to a server.

Enter numerical coordinates or click either CIE diagram. XYZ results are reported on a relative scale with Y = 100; this does not supply absolute luminance.

Enter values to calculate a color.

02 / OUTCOME

One color. Complementary measures.

Comments & suggestions

Help improve CIE color coordinates, CCT and Duv. Feedback is public and requires a GitHub account to post.

Leave feedback on GitHub

03 / CIE DIAGRAMS

Locate the color.

Click a valid position to calculate that color. Save multiple points to compare emitters.

CIE 1931

x, y · standard 2° observer
Applicable to your 1° viewing field under the CIE 1931 standard. Boundary derived from official 1 nm CMFs.

CIE 1976 UCS

u′, v′ · transformed from the same XYZ
u′v′ chromaticity is distinct from CIELUV L*u*v*. Duv is evaluated in CIE 1960 uv, not this diagram.
Current colorReference white WReference RGB gamutYour RGB gamut
Saved color coordinates
Pointxyu′v′CCT / KDuvAction
No saved points yet.

04 / RGB GAMUT

Compare RGB gamut and coverage.

A single emission point cannot define a display gamut. Supply three RGB primary coordinates or assign three calculated emitters.

Red primary

Green primary

Blue primary

Enter all three primary coordinates to calculate area ratio and coverage.

Two-dimensional chromaticity-triangle comparison
DiagramYour triangle areaReference areaArea ratioCoverage of reference
RGB inputs are required.

Area ratio100 × area(your triangle) / area(reference triangle). This can exceed 100%.

Coverage100 × area(overlap) / area(reference triangle). This cannot exceed 100%.

Values depend on the selected coordinate diagram. These are 2D geometric measures, not perceptual color volume or a measurement of the browser/display hardware. Display P3 and DCI-P3 share primary coordinates but have different reference whites.

Comments & suggestions

Help improve RGB color gamut and coverage. Feedback is public and requires a GitHub account to post.

Leave feedback on GitHub

05 / ACHROMATIC REFERENCES

Common white points.

Rounded standard coordinates for the CIE 1931 observer. Choosing a reference white changes the purity construction and reference distance; it does not alter a source’s measured chromaticity.

White-point reference coordinates
Referencexyu′v′Use as input

06 / COLOR RENDITION

How faithfully does the light render color?

Calculate CRI/Ra, R9, R15, CQS and TM-30 from the emission spectrum selected above. Coordinates alone cannot determine these scores.

Upload or load a spectrum in the emission input above.

For color rendition, supply 380–780 nm with sample spacing no greater than 5 nm. Use the zero-extension option only when missing emission is known to be negligible. It is also applied to photobiological calculations below, where wider ranges may be required.

Color rendition needs a numerical spectrum. The advanced engine downloads on first use and runs locally in your browser.

Principle and interpretation

CRI/Ra and R1–R15 use the CIE 2024 sample set in Colour 0.4.7; Ra remains the average of the first eight samples. CRI is withheld when the source–reference distance in CIE 1960 uv exceeds 0.0054. CQS reports NIST CQS 9.0 Qa, Qf and Qg. TM-30 reports the ANSI/IES TM-30-18 implementation, including all 99 sample scores in the CSV export. The method edition is recorded with each result.

This workspace reports rendition for near-white sources at 2,000–25,000 K and |Duv| ≤ 0.05. CRI and CQS use the 2° observer; TM-30 uses its prescribed 10° sample colorimetry and 2° CCT calculation. The chromaticity diagrams retain your CIE 1931 observer setting. Numerical values may vary slightly with sampling and method edition; use the current official IES tool for a standards report.

Comments & suggestions

Help improve Color rendering: CRI, CQS and TM-30. Feedback is public and requires a GitHub account to post.

Leave feedback on GitHub

07 / LED & WHITE-LIGHT EFFICACY

Connect optical output to electrical input.

Enter measured device values. The selected spectrum supplies luminous efficacy of radiation and the photon-energy weighting used for EQE.

Use values from the same operating point. Total radiant flux requires a calibrated total-output measurement, such as an integrating sphere. A spectrum used with that flux must represent the same output and include all relevant emission. Current is needed for EQE even when electrical power is entered directly.

Add the measurements available; each result shows which inputs it needs.

Principle and outcome

Electrical power = V × I. Device luminous efficacy = luminous flux / electrical power (lm/W electrical). Wall-plug efficiency = radiant flux / electrical power. Spectral LER = 683 ∫S(λ)V(λ)dλ / ∫S(λ)dλ (lm/W optical), using all supplied radiant power in the denominator and the 360–830 nm photopic function in the numerator. Missing ultraviolet or infrared emission biases a whole-source LER.

EQE = emitted photons / injected electrons. For radiant-power spectra, photon rate = radiant flux × the power-weighted mean wavelength / (hc); electron rate = I/e. Without a spectrum, the entered wavelength supplies a narrow-band approximation. IQE = EQE / extraction efficiency only when extraction has been independently determined. Photoluminescence quantum yield is a separate measurement. Current efficiency = measured luminous intensity / I (cd/A); no angular-distribution assumption is applied.

When measured lumens are absent, luminous flux can be estimated from matched radiant flux × spectral LER. Such output is explicitly labeled as spectrum-derived.

Comments & suggestions

Help improve LED efficiency and efficacy. Feedback is public and requires a GitHub account to post.

Leave feedback on GitHub

08 / PHOTOBIOLOGICAL QUANTITIES

Weight the spectrum for biological response.

Explore CIE S 026 α-opic metrics and selected exposure-weighted quantities. Absolute results need a matching photopic or radiometric measurement at the same location and direction as the spectrum.

Load a spectrum above. DER and spectral weighting ratios can be calculated without absolute calibration.

CIE S 026 · five photoreceptor classes
ResponseELR / mW·lm⁻¹Daylight efficacy ratio · DEREquivalent daylight illuminance / lxα-opic irradiance / W·m⁻²
A spectrum is required.

Principle, required ranges and scope

α-opic irradiance = ∫Eλsα(λ)dλ. ELR is weighted irradiance divided by photopic illuminance; DER divides that ratio by the corresponding D65 value. EDI = illuminance × DER. With source luminance, the analogous results are α-opic radiance and equivalent daylight luminance (EDL). Alternatively, scale the spectrum to a measured irradiance or radiance integrated over the exact uploaded wavelength interval; this supports UV/IR-only sources without a photopic normalization. ELR and DER are undefined when photopic output is zero. The five action spectra are the official CIE S 026 data. Calculations use 380–780 nm, including V(λ) normalization over that interval.

The blue-light factor uses B(λ) over 300–700 nm from ICNIRP 2013, Table 2. Multiplying the factor by illuminance gives EB (W·m⁻²); multiplying by luminance gives LB (W·m⁻²·sr⁻¹). Steady dose = weighted value × duration. Retinal thermal radiance uses the revised R(λ) over 380–1400 nm and is reported only for luminance input. UV-A (315–400 nm) and infrared (780–3000 nm) outputs are unweighted band quantities.

Coverage is checked for each quantity. Missing wavelengths are treated as zero only when you explicitly select the zero-extension assumption. A visible-only file cannot establish absence of UV/IR. These results characterize spectra and steady exposures; they do not assign an IEC 62471 risk group, a safe viewing time, or predict a health outcome. A safety assessment also needs calibrated spectroradiometry, source geometry, averaging field and the applicable exposure conditions.

Comments & suggestions

Help improve Photobiological spectral quantities. Feedback is public and requires a GitHub account to post.

Leave feedback on GitHub

09 / PHOTOLUMINESCENCE QUANTUM YIELD

Calculate PLQY and PLQE.

Compare integrating-sphere traces or a known reference standard; report absorbed- and incident-photon normalization explicitly.

The PLQY workspace includes a blue/red S₀–S₁–S₂ example, blank subtraction, photon-weighted integration, relative and slope methods, absorptance, energy yield, uncertainty and lifetime-based rates.

Open the PLQY / PLQE workspace ↗

METHOD / SCOPE

Principle and interpretation.

Coordinates and spectra

Emission XYZ is obtained by integrating spectral radiant power with the official CIE x̄, ȳ, z̄ functions. Linear interpolation is used between supplied samples; products of the interpolated functions are integrated across their combined breakpoints. XYZ is normalized to Y = 100.

x = X/(X+Y+Z), y = Y/(X+Y+Z); u′ = 4x/(−2x+12y+3), v′ = 9y/(−2x+12y+3). For CIE 1960, u = u′ and v = 2v′/3.

CCT and Duv

A numerical minimum search finds the closest Planckian radiator in CIE 1960 uv over 1,000–100,000 K. Duv is the signed distance: positive above, negative below the locus. CCT is withheld for |Duv| > 0.05 or when the nearest result reaches the calculation limits.

A saturated red, green or blue emitter can have valid color coordinates while a white-light CCT is not applicable.

Wavelength and excitation purity

Extend a line from the chosen white through the sample to the spectral boundary. Excitation purity is the sample-to-white distance divided by boundary-to-white distance in xy. For purple-side colors, the complementary wavelength is reported and purity uses the purple boundary.

At the reference white, purity is zero and dominant wavelength is undefined. These quantities depend on the chosen reference white.

What needs more information?

RGB gamut needs three primary colors. Peak wavelength, FWHM, spectral centroid and visible-band luminous efficacy need spectral data. Absolute luminance/flux needs calibrated radiometric data and measurement geometry.

The added color-rendition and photobiology sections calculate these quantities from spectral data. LED device efficiency additionally requires electrical and calibrated optical measurements. Coordinates alone supply chromaticity, CCT/Duv and white-point relationships.

Observer data: CIE (2019), Colour-matching functions of the CIE 1931 standard colorimetric observer, 1 nm, 360–830 nm. Data and the derived diagram boundary: CC BY-SA 4.0. CSV converted to JavaScript without changing values; original MD5: 17cca777db64b17170f06f67ce9d3ab7.

CIE observer field-size scope · CIE CCT definition · IES excitation purity · ICC RGB specifications · Primary-coordinate references.

Related reference calculators: xy to Duv · CCT to xy. Chromaticity uses original numerical integration and geometry code. Advanced rendition uses the pinned Colour 0.4.7 implementation (BSD-3-Clause); its scientific libraries load on demand and uploaded data stays in your browser.

Additional methods: CRI implementation · NIST CQS · TM-30-18 implementation · Official IES spectral tools · CIE α-opic data (2018), DOI 10.25039/CIE.DS.vqqhzp5a · CIE D65 (2019), DOI 10.25039/CIE.DS.hjfjmt59 · ICNIRP 2013 weighting functions and errata.

CIE α-opic and D65 data, including their JSON conversion, are CC BY-SA 4.0. Undefined cone sensitivities outside the tabulated ranges are filled with zero according to CIE metadata. Colour license. The bundled Colour package excludes documentation, examples and tests; calculation code and required data are unchanged.