The CRI method dates to 1965 and still averages eight moderately saturated samples. Television cameras needed their own answer, TLCI. Then IES rebuilt the framework on 99 samples, TM-30 (Rf for fidelity, Rg for gamut area). Three numbers, three jobs.
Four ideas to carry into the next specification conversation you have.
Ra is the mean of R1-R8, eight moderately saturated samples. The saturated reds and blues sit outside that average. Current edition: CIE 13.3-1995.
R9 measures a strong red and is not in Ra, so a source can pass on Ra and still render skin and red badly. WELL v2 accepts Ra ≥ 80 only when R9 ≥ 50.
Rf = fidelity. Rg = gamut area. TM-30 Annex E's top preference level asks Rf ≥ 78, Rg ≥ 95 and −1% ≤ Rcs,h1 ≤ 15%. There is no upper limit on Rg.
It models a camera and display, not the eye. EBU R 137 recommends the method to identify problem luminaires; it sets no pass mark. Different number, different question.
What each document actually asks for. Anything beyond these rows is a project decision, not a standard.
| Where it applies | What to ask for | Source |
|---|---|---|
| Interior workplaces, general | Ra ≥ 80 | EN 12464-1:2021 general rule, with defined exceptions for particular tasks |
| WELL v2 projects | Ra ≥ 90, or Ra ≥ 80 with R9 ≥ 50, or Rf ≥ 78, Rg ≥ 100, −1% ≤ Rcs,h1 ≤ 15% | WELL v2 feature L07 part 1, any one route passes |
| Pleasing, natural-looking spaces | P1: Rf ≥ 78, Rg ≥ 95, −1% ≤ Rcs,h1 ≤ 15% | TM-30 Annex E, colour preference intent, 200-700 lux polychromatic |
| Colour-critical work | F1: Rf ≥ 95 · F2: Rf ≥ 90, Rf,h1 ≥ 90 | TM-30 Annex E, colour fidelity intent |
| Display and feature lighting | V1: Rg ≥ 118, Rcs,h1 ≥ 15% | TM-30 Annex E, colour vividness intent |
| Anything that gets filmed | A Qa figure and the production type | EBU Tech 3355. The scale reads two ways and no minimum is mandated, so the project sets its own |
| Saudi street lighting | 3000-6000 K, 5-step MacAdam | SASO 2927:2019 regulates colour consistency and CCT, not colour rendering |
Five reasons the lighting industry never settled on one number.
The method dates to 1965 (CIE 13), was updated in 1974 to account for chromatic adaptation, and the current edition is CIE 13.3-1995. Ra is the average across eight moderately saturated samples (R1-R8) against a reference at the same CCT: a Planckian radiator below 5000 K, CIE daylight at or above it. LEDs render those eight well but can fail saturated red (R9), deep blue (R12) and complexion (R13). A high Ra hides those failures.
R9 measures a strong red, and it is not in Ra. It also comes from a non-uniform colour space, so R9 values swing far wider than Ra values and can fall below zero. That is why specifiers started writing R9 as its own line instead of trusting Ra alone, and why TM-30 reports a red-bin fidelity value, Rf,h1, on the ordinary 0-100 scale instead.
IES TM-30-15 (2015) introduced 99 Colour Evaluation Samples, down-selected from over 100,000 measured object reflectances. TM-30-18 harmonised the framework with CIE 224:2017 and became an American National Standard; Annexes E and F were added in 2019; TM-30-20 republished it with no technical change; the current edition is ANSI/IES TM-30-24. Rf is the fidelity analogue of Ra. Rg is the gamut area ratio: above 100 means more chroma on average, below 100 means less. Because it is an area, hue shifts move it too, so it is not a saturation dial.
Television Lighting Consistency Index (EBU Tech 3355) simulates a broadcast camera's spectral response. A TLCI score reflects how the light will look on-camera after auto white balance. EBU R 137 recommends the Tech 3355 method be used to identify problem luminaires and specifies how to measure; it does not set a minimum Qa. The scale itself carries two readings, one for film-style work and one for live multi-camera, so the same Qa means different things depending on the production. Irrelevant for spaces that aren't filmed.
Spec the number that fits the space. For an interior workplace, EN 12464-1's Ra ≥ 80 general rule is the starting point. For colour-critical retail, name a TM-30 Annex E cell, the CCT it applies at, and the red-bin value, rather than inventing limits of your own. The metric is a contract, not a marketing slogan.
GRANDEUR also publishes a TLCI value for venues that get filmed. R9 and TM-30 figures are not published per product today, so if a project is judged on them, raise it with us at specification stage.
A 15-slide explainer, ready to post straight to LinkedIn as an image carousel. Same content, faster reading.
15 slides · LinkedIn-ready. Download all 15 as high-res PNGs and post them straight to LinkedIn as a carousel. PDF version still available.
Practical lighting know-how from NLC Academy. No spam, unsubscribe anytime.
By subscribing you agree to receive NLC emails.