A few years ago, while living in a small apartment in Montreal’s Plateau neighborhood, I noticed that the lightbulb in my kitchen had burned out. Naturally, I went to the hardware store to find a replacement. After struggling to read the French labeling on several different packages, I ultimately decided to go for the least expensive box of LED bulbs that the store had available. 
Once I installed the newly purchased lightbulb, I noticed a difference in the way our kitchen looked. Specifically, I noticed a difference in the appearance of the bowl of fruit that always sat on the counter. While the lightbulb illuminated the space, I remember thinking to myself how unappetizing and dull my fruit now looked. This exercise, though unintentional, clarified the importance of a light source’s color rendering capability.
More recently, I attended an Illuminating Engineering Society (IES) Webinar on lighting metrics. I had done some research in the past on lighting classifications but wanted to further understand what methods are currently in place for specifying characteristics of a light source. Upgrading lighting systems to LEDs is often one of the first measures considered when trying to improve building energy efficiency, yet the metrics we currently use to inform our design decisions are lagging when compared to the technology for which they’re used. With advances in semiconductor technology and an increasing demand for LED lighting, appropriate lighting metrics are becoming increasingly important. My hope is that this post will serve as a brief overview of the limitations of current metrics and point the reader towards new evaluation methods that are emerging in the lighting industry.
The Commission Internationale de l'Eclairage (CIE) established the CIE Ra metric (more generally known as the CRI, or Color Rendering Index) as a method of specifying a light source’s color rendering capability. A CRI score is found on most lighting cut sheets along with Corelated Color Temperature (CCT), lumen output, wattage, and color vector graphics. The score is typically established by comparing a set of colors rendered under the test source to the same colors rendered under a reference source (typically the sun, but the reference source can vary). More generally, the CRI is determined by the emitted light spectrum of a source. A continuous spectrum (incandescent source) would have a higher CRI value than a discrete spectrum (fluorescent source).
While the CRI is an acceptable average measure of color rendition, it has some shortcomings. First, the CRI method typically relies on only eight colors (or sometimes 14), meaning there are many colors that are not represented in a CRI score. Due to this lack of variety in test colors, lighting manufacturers could tailor their products to score highly on CRI specific colors while ignoring many other colors. This can result in high CRI scores for an LED that renders some colors improperly (think unsaturated, washed out, or unnaturally tinted visual effects). Additionally, the CRI only compares light sources with the same CCT. This suggests that in large lighting projects with fixtures of various CCTs, CRI comparisons of those fixtures are not necessarily valid.
Again, while the CRI may be perfectly adequate for many scenarios, the lack of specificity in this method makes it possible for the same scene to look very different under sources with the same CCT and CRI. Personally, I believe this is the most critical inadequacy of the CRI. The images below illustrate this issue – despite one scene appearing more vibrant, these sources could have identical CRI values. This may not be a problem for all lighting designs, but it could have significant impacts in food sale, display, and retail applications.
For LED lighting projects, equipment cost, performance, and energy savings potential are all important considerations. Today, the CRI remains the most common metric used for color rendering capability and is a useful indicator to consider. Eventually, though, new evaluation methods will likely be adopted in the lighting industry to more accurately compare equipment cost and performance. After all, the CRI was first introduced in the 1960s and was most recently updated in the 1970s. Since then, the lighting industry has undergone immense change, and it’s likely that lighting metrics will eventually do the same. While the lighting industry may not fully adopt a new metric soon, there are certainly options on the horizon that address the shortcomings of the CRI method.
