CIE Illuminants & SPD Lab

CIE 1931 xy

●—

Expanded uncertainty, k = 2

Quantity	Value	u	U (k=2)	Relative
Not run yet.

Contribution to u(CCT)

95% coverage region in u′v′

●Monte Carlo per JCGM 101:2008 · inputs uncorrelated unless a correlation matrix is supplied

What the pipeline is worth

Variation	x	y	CCT	ΔD_uv

CCT method

Method	CCT	D_uv	Validity

Rendition metrics

Query Ctrl+Enter to run

Query

Recipes

Grammar

select names or ranges · with observer, interval, interpolation · report columns · order by column asc/desc · limit n

Parsed, never evaluated as code. The same engine answers here as on the tabs, so a console row and a Lab reading cannot disagree.

Results

Any two columns, plotted

Horizontal axis Vertical axis

The register

What the statuses mean

Verbatim the published table, copied.

Computed derived exactly from a defining formula or a verbatim table.

Synthesised a stand-in that is not the published data. Never reported as a standard value.

Absent not held. The features that need it stay switched off rather than guessing.

Every dataset this tool touches

Dataset	Source	Tabulated	Status	Why it matters

Switched off, and why

Multi-illuminant comparison table

Compare all 18 standard illuminants side-by-side: CCT, CIE xy, CRI Ra, and white-point swatch.

Click button above to generate.

Reference

Four tabs became one view. The counts are what each holds.

Filter

Matches across all four, so a term you half-remember does not need you to guess which tab it was on.

Status

Counted from the data register, so the filter cannot claim more than the tool holds.

All rows

Standards — and what each one is actually backed by here

CIE 15:2018 — Colorimetry

CIE 15:2018 (4th edition) is the fundamental reference for colorimetric computation. Defines the CIE 1931 2° standard observer, illuminant SPDs (A, D50, D55, D65, D75, E), XYZ tristimulus values, CIELAB, and recommended illuminant-observer computations. Tables provide SPD data at 1 nm and 5 nm intervals from 300–830 nm.

CIE 13.3:1995 — Colour Rendering Index

Defines the CIE Colour Rendering Index (CRI). Ra is the general index (mean of R1–R8). R1–R14 are special indices for 14 test-colour samples. Reference illuminant is a Planckian radiator for CCT < 5000 K and a CIE daylight phase for CCT ≥ 5000 K. Adaptation uses CIE 1960 UCS Von Kries, and colour difference is in CIE 1964 U*V*W* space.

IES TM-30-18 — Colour Rendition

The Illuminating Engineering Society TM-30-18 method evaluates colour rendition using 99 colour evaluation samples (CES). Rf (fidelity index, 0–100) measures average colour shift from a reference. Rg (gamut index, ~60–140) measures change in gamut area. Together they address CRI’s known limitations — particularly with saturated colours and narrow-band sources.

ISO 10526:2007 — CIE Standard Illuminants

ISO 10526 specifies the spectral distributions of CIE standard illuminants A, D65, and supplementary illuminants D50, D55, D75. It establishes the normative SPD tables and computation rules for the daylight series using S0, S1, S2 basis vectors and the CIE daylight chromaticity formula.

ICC.1:2022 — Profile Connection Space

The ICC specification defines D50 as the PCS illuminant. All ICC profiles must express colorimetric data relative to D50. Bradford chromatic adaptation is mandated for converting from device white-points (often D65) to D50. The illuminant CIE xy values for D50 (0.3457, 0.3585) are normative.

CIE Standard Illuminant A

CIE Illuminant A represents typical domestic tungsten-filament lighting. It is defined as a Planckian radiator at 2856 K. The relative SPD is computed as:
S_A(λ) = 100 · (560/λ)^5 · (exp(c2/(560·T)) - 1) / (exp(c2/(λ·T)) - 1)
where c2 = 1.4388 × 10^7 nm·K. This has a warm yellowish spectrum peaking in the infrared.

CIE D-Series Daylight Illuminants

D-series illuminants (D50, D55, D65, D75) are defined by the CIE daylight formula using three basis vectors S0(λ), S1(λ), S2(λ) measured from real daylight data by Judd, MacAdam & Wyszecki (1964). The SPD is:
S(λ) = S0(λ) + M1·S1(λ) + M2·S2(λ)
where M1 and M2 are computed from the daylight chromaticity xD, yD derived from CCT.

D50 (5003 K): ICC PCS, printing industry standard.
D55 (5503 K): Mid-morning/afternoon daylight.
D65 (6504 K): Average noon daylight, sRGB reference white.
D75 (7504 K): North sky daylight, overcast conditions.

CIE Fluorescent Illuminants (F1–F12)

CIE defines 12 fluorescent illuminant types in three groups:

F1–F6: Standard (halophosphate) fluorescent lamps. Broad emission with mercury lines at 405, 436, 546, 578 nm.
F7–F9: Broadband (de luxe) fluorescent. Higher CRI, simulate daylight.
F10–F12: Narrow-band three-phosphor lamps. Efficient but low CRI for saturated colours.

Every status here is read from the data register, so a document cannot be marked implemented by writing prose about it.

Mathematical Models and Formulas

Planck’s Law — Blackbody Spectral Radiance:

B(lambda, T) = (2hc^2 / lambda^5) * 1 / (exp(hc / lambda*k*T) - 1)

h = 6.62607e-34 J·s (Planck constant)
c = 2.99792e8 m/s (speed of light)
k = 1.38065e-23 J/K (Boltzmann constant)

Second radiation constant: c2 = h*c/k = 1.4388e-2 m·K

Wien displacement: lambda_peak = 2.898e6 / T (nm)

Illuminant A uses this at T = 2856 K, normalised
relative to 560 nm for the CIE standard SPD.

Citations — and where each one is used

CIE illuminants and colorimetry

[1] CIE (2018). Colorimetry, 4th Ed. CIE 15:2018. cmf-2, d-basis Runs here

[2] ISO 10526:2007. CIE standard illuminants for colorimetry. the ISO adoption of the same illuminants Context only

[3] Judd, D.B., MacAdam, D.L. & Wyszecki, G. (1964). Spectral distribution of typical daylight as a function of correlated colour temperature. JOSA 54(8), 1031–1040. the S₀ S₁ S₂ basis Runs here

[4] CIE (1986). Colorimetry, 2nd Ed. CIE 15.2:1986. the edition the older formulae came from Context only

[5] CIE 167:2005. Recommended practice for tabulating spectral data for use in colour computations. Sprague resampling Runs here

Colour rendering

[6] CIE (1995). Method of measuring and specifying colour rendering properties of light sources. CIE 13.3-1995. R₀ and R₁–R₁₄, on a stand-in TCS set Data missing

[7] IES (2018). IES method for evaluating light source colour rendition. TM-30-18. no CES set held — no result shown Data missing

[8] David, A., Fini, P.T., Houser, K.W., et al. (2015). Development of the IES method for evaluating the colour rendition of light sources. Optics Express 23(12), 15888. why TM-30 exists Context only

Chromatic adaptation

[9] Lam, K.M. (1985). Metamerism and colour constancy. PhD thesis, University of Bradford. the Bradford matrix Runs here

[10] Estévez, O. (1979). On the fundamental data-base of normal and dichromatic colour vision. PhD thesis, University of Amsterdam. the Von Kries (HPE) matrix Runs here

[11] CIE 159:2004. A colour appearance model for colour management systems: CIECAM02. the CAT02 matrix Runs here

[12] Li, C., Li, Z., Wang, Z., et al. (2017). Comprehensive colour solutions: CAM16, CAT16 and CAM16-UCS. Color Res. Appl. 42(6), 703–718. the CAT16 matrix Runs here

[13] Finlayson, G.D. & Süsstrunk, S. (2000). Performance of a chromatic adaptation transform based on spectral sharpening. Proc. IS&T/SID 8th Color Imaging Conference, 49–55. the Sharp matrix Runs here

[14] Fairchild, M.D. (2013). Color Appearance Models, 3rd Ed. Wiley-Blackwell. how to choose between them Context only

[15] ICC (2022). ICC.1:2022 Image technology colour management. why the PCS white is D50 Context only

Colour temperature

[16] Ohno, Y. (2014). Practical use and calculation of CCT and Duv. LEUKOS 10(1), 47–55. every CCT and D_uv on this page Runs here

[17] McCamy, C.S. (1992). Correlated colour temperature as an explicit function of chromaticity coordinates. Color Res. App. 17(2), 142–144. the legacy figure, shown only for comparison Runs here

[18] Robertson, A.R. (1968). Computation of correlated colour temperature and distribution temperature. JOSA 58(11), 1528–1535. the isotherms here are computed, not tabulated Context only

[19] Wyszecki, G. & Stiles, W.S. (2000). Color Science: Concepts and Methods, Quantitative Data and Formulae, 2nd Ed. Wiley. the reference values checked against Context only

Uncertainty, and datasets not held

[20] JCGM 101:2008. Supplement 1 to the GUM — Propagation of distributions using a Monte Carlo method. the whole Uncertainty view Runs here

[21] CIE S 026/E:2018. CIE system for metrology of optical radiation for ipRGC-influenced responses to light. no α-opic spectra held Data missing

[22] MacAdam, D.L. (1942). Visual sensitivities to colour differences in daylight. JOSA 32(5), 247–274. no ellipse parameters held Data missing

Nine of these were missing until the engine was read back against its own source: it names its methods in code, and the register names the standards behind the datasets it does not hold. Each addition was checked against the publisher record — which is how the Sprague reference turned out to be CIE 167:2005, not the ASTM practice the code comment claimed.

This tool computes illuminant SPDs, CCT and D_uv by Ohno's method, chromaticity in three planes, six chromatic adaptation transforms and Monte Carlo measurement uncertainty, entirely in the browser. CRI is computed against a stand-in sample set and labelled as such; TM-30 is not computed at all. Not a substitute for calibrated measurement or official CIE/IES software.

Research Backend

Backend

CRI Limitations Analysis

CRI has known limitations with modern narrow-band LED and 3-phosphor fluorescent sources. The 14 TCS reflectances are relatively unsaturated, meaning CRI can rate a source highly even if it renders saturated colours poorly. TM-30 addresses this with 99 colour evaluation samples and separate fidelity/gamut metrics.

Key CRI limitations:
1. Only 14 TCS (mostly desaturated Munsell samples)
2. Uses CIE 1964 U*V*W* (non-uniform colour difference)
3. No gamut shift information (expansion vs contraction)
4. No hue-specific rendering data
5. Narrow-band sources can score Ra > 80 while
distorting saturated reds and deep blues significantly

TM-30-18 improvements:
- 99 CES spanning full hue/saturation range
- Uses CAM02-UCS and CIEDE2000
- Separate Rf (fidelity) and Rg (gamut) indices
- 16-bin hue-angle gamut vectors for directionality

Metamerism Analysis Guide

Two objects can appear identical under one illuminant but different under another (illuminant metamerism). The metamerism index quantifies this risk. High values between illuminant pairs indicate that colour samples at risk — critical for textile, automotive, and packaging industries where products are viewed under varying store/office lighting.

Metamerism Index (MI) computation:
For each of 14 TCS (or 99 CES in full TM-30):
XYZ_A = integral(TCS * SPD_A * CMF)
XYZ_B = integral(TCS * SPD_B * CMF)
Lab_A, Lab_B from respective XYZ
dE_i = CIEDE2000(Lab_A, Lab_B)

MI = mean(dE_1..dE_N)

Interpretation:
MI < 0.5: very low metamerism risk
MI 0.5-2: moderate risk — visible to trained observers
MI 2-4: high risk — clearly visible shifts
MI > 4: severe — different appearance guaranteed

Practical Lighting Evaluation Workflow

Step-by-step method for evaluating a lighting source against requirements:

Measure or select the test illuminant SPD.
Compute CCT and Duv — confirm Duv < 0.006 for the source to be “on the Planckian locus.”
Compute CRI Ra — general acceptability threshold is Ra ≥ 80 for most applications, Ra ≥ 90 for critical colour work.
Check R9 (saturated red) separately — many LED sources score Ra > 80 but have R9 < 50.
Compute TM-30 Rf and Rg for a more complete picture. Rf ≥ 85 and Rg 95–105 is considered excellent.
If comparing two illuminants, compute the metamerism index — MI < 1 ensures minimal colour shift risk.
Use the surface reflectance analysis to visually inspect how critical samples (skin, textiles, prints) appear.
For ICC workflows, confirm the Bradford adaptation from test white to D50 produces acceptable ΔE₀₀.

Research backend provides: 18 standard illuminants, SPD spectral charts, CRI Ra + R1–R14, TM-30 Rf/Rg approximation, metamerism index, Bradford + Von Kries chromatic adaptation, surface reflectance analysis, multi-illuminant comparison, full export (CSV/JSON/PNG). All computation is on-device with zero network dependency.