Lighting Quality

by Nick Gromicko, CMI® 


Efforts to enhance home energy efficiency have spurred a growth in a variety of alternative lighting sources that use less energy than conventional Compact fluorescent bulbs are often poor color renderers, despite being economicalincandescent bulbs. With this improvement comes a greater variety of light quality, although this latter property has become obscured amidst the excitement generated by energy savings. How well do these new, “green” lighting sources actually render color to the human eye? They can illuminate a room at a lower cost, but is it necessary to forfeit the ability to tell whether your clothes match just to save money and energy? Many consumers and inspectors should be aware of a metric designed to quantify this aspect, known as the color rendering index, known as CRI. CRI is a measure of how well light sources render the colors of the things they illuminate, such as skin tones and fabrics.

How is CRI measured?

The appearances of eight color samples are compared under the light in question and a reference light source on a scale of 0 to 100. The average measured differences are subtracted from 100 to arrive at the light’s CRI, which correlates positively with light quality. Thus, small average differences will result in a higher or better score, while larger differences result in a lower CRI and poorer light quality.

Criticisms of CRI

While CRI is a useful and generally accepted form of measurement, it has its criticisms.  Among them:

  • All errors made during testing are weighed equally, while human perception tends to favor or ignore certain errors over others.
  • By valuing the arithmetic mean of the errors, single large deviations become under-valued. Thus, two light sources with identical CRI values may perform differently if outlying deviations exist in a spectral band that is significant for the application.
  • More than eight samples may be required. With more samples, it would be harder for manufacturers to optimize their lamps to reproduce the test hues faithfully, which otherwise perform poorly. CRI merely measures the faithfulness of a light source to an idealized source with the same correlated color temperature, or CCT, but the ideal source itself may render colors poorly if it has an extreme color temperature due to a lack of energy at either short or long wavelengths (i.e., it may be excessively blue or red).

What is CCT, and how does it relate to CRI?

A common misconception is that CRI alone can be used to gauge the quality of a light source, perhaps as a result of deliberately misleading marketing campaigns. To compare lights solely based on its CRI would be akin to comparing baseball players based on batting averages while neglecting their league (i.e., your 9-year-old son may have a better batting average than Alex Rodriguez, but the two cannot be logically  compared). Similarly, CRI is a useful metric only when it is compared with light sources that have a similar CCT. This metric describes the temperature (measured in degrees Kelvin/K) that a black body radiator -- an idealized material that absorbs all radiation that falls into it and emits a temperature-dependent spectrum of light -- would need to be at in order to glow at a given wavelength. Higher CCTs (5,000 K or more) are considered “cool” (blueish white) colors, while lower CCTs (2,700 to 3,000 K) are “warm” (yellowish white through red) colors.

Despite their CRI rating of 100, incandescent bulbs are far from ideal for color rendering because they have a CCT of just 2,700 K. These lights are weak at the blue end of the spectrum, muddying the differences between different shades of blue. The perfect CRI rating merely means that sample hues look exactly the same as they would if illuminated by a black body radiator at the same CCT, but neither light source would render the color faithfully. Similarly, lamps that exceed 6,000 K in color temperature are too weak in the red end of the spectrum, making it difficult to distinguish reds and oranges, resulting in a washed-out appearance. An ideal light source for color rendering will have a high CRI value as well as a color temperature similar to daylight.

Warmer or lower color-temperature lights are often used in public areas to promote relaxation, while cooler or higher color-temperature lights are used to enhance concentration in offices. Thus, the best color temperature ranges are based on application, but for general indoor lighting, it is best to match the color temperature of mid-day sunlight, or approximately 5,400 K. 

CRI and Color Temperatures of
Common Light Sources

Light Source
CCT (degrees K)




incandescent bulb
tungsten halogen
Solux bulb
natural sunlight
5,000 to 6,000
Bell & Howell sunlight lamp
80 to 85


white fluorescent bulbs
50 to 98

 Other Factors That Affect Choice for Indoor Lighting:
  • lumens, which indicate light output;
  • lumens per watt, which indicate how much light is produced for the energy used.  This measurement shows how efficient the different types of lamps are. Standard tungsten incandescent lights generally produce around 15 lumens/watt, while some CFLs can emit more than 100 lumens/watt. Cree, Inc., recently announced that it created an LED of 208 lumens/watt, an industry-best.
  • price. CFL, LED and halogen lights vary based on initial cost. Incandescent lights are generally the cheapest option, and LEDs are sometimes prohibitively expensive. 
  • disposal. Fluorescent tubes are long and fragile, making them difficult to dispose of, and all fluorescent lights contain mercury.
In summary, consumers should consider CRI and CCT before they purchase lighting.