Color accuracy in lighting: artificial and natural lighting
In our various discussions about color rendering and lighting quality, we oftentimes reference color accuracy. But what does it mean to be "accurate" and why do we care?
We discuss this question in detail below.
Our discussion of color accuracy must begin with clarification on natural vs artificial lighting.
Human vision has primarily evolved and developed under natural light sources - the sun during the day, and fire at night.
Over time, however, as humans invented a range of artificial lighting technologies that allowed for illumination in both dark locations as well as evening hours, it became evident that some are more similar to natural light sources than others.
The invention of gas lamps and incandescent lighting may put us in "artificial lighting" territory, but in terms of color and spectrum, we are still not too far off from natural daylight and fire.
In incandescent and halogen lighting, light is generated by heating an object (the filament) to an extreme temperature such that it begins to emit heat. This is why incandescent light bulbs are so inefficient - its filament needs to be warmed to a sufficient temperature before it begins to emit enough light.
This temperature, commonly known as color temperature, happens to be 2700 Kelvin for incandescent lighting, and 3000K for halogen. On the other hand, the temperature of the sun is approximately 5500 Kelvin (about twice as hot as the filament in a light bulb) and therefore takes on a bluer tint.
The mechanism behind incandescent light bulbs and the sun are similar in that they both emit light as a result of an object being warmed to extreme temperatures.
They exhibit different colors because of the difference in physical temperature, but because of the similarity in mechanism, we perceive incandescent lighting to also be "ideal" the same way we consider daylight to be "ideal" despite having a different color.
With the advent of fluorescent and LED lighting, however, we take a big step into "artifical lighting" territory. The main reason for this is that the mechanism behind generating light is completely different.
The first type of color accuracy asks: how similar is this light's emitted color to a natural light source? Our answer here uses chromaticity coordinates (CCT & Duv) to analyze.
The second type of color accuracy asks: how similar do objects' colors appear under this light compared to a natural light source? Our answer here uses CRI and other color quality metrics to analyze.
The questions are a bit nuanced, so it can be a bit tricky. The first type is concerned with the light coming out of the light. The second type is concerned with the color reflected off of the object it illuminates.
The first type of color accuracy is concerned with whether the emitted light color is accurate when compared to a natural light source.
But before we can ask this question meaningfully, we need to determine what "natural light source" we want to reference.
If you're working on a painting at night, and want it to appear the same under natural daylight, you will need a light bulb with emitted color that matches that of natural daylight.
On the other hand, if you're working on a painting and care about how it appears in an art gallery with halogen lighting, you will want to use a light bulb with emitted color that matches halogen lighting.
We use CIE 1931 xy coordinate chart to define the color points of our natural light source, and the color of the light emitted from the artificial light source.
By comparing the proximity between the natural light source and our artificial light source, we can determine how similar, or accurate, the two light sources are in their emitted color.
We describe the second type of accuracy as the ability of a light source to reveal an object's colors the same as under natural lighting. This is the core definition of the color rendering index.
Under this definition, we assume that the artificial light source's emitted light color is already deemed "accurate" and is what we are looking for.
However, that is not to say that the colors of the object will appear "accurate."
To extend our painting example above, we may have already determined that our light bulb's emitted white light is satisfactorily similar to natural daylight or halogen.
But once we shine the artificial light source onto the painting, do the colors look accurate? Do they look the same as under our natural light source?
In order to determine the CRI, various standards and procedures have been developed. Specifically, the CIE has developed what is called the Illuminant D series to simulate daylight.
These daylight standards are used so that we can ask ourselves, "how accurate is this light source when compared to daylight?"
Similarly, black body radiation can be used to simulate warmer color temperatures less than 5000K.
When artificial light sources emit light, they may be missing significant portions of the visible spectrum, even if the light appears white.
Instead of emitting light as a result of heating, LEDs emit light as a result of electrons being converted to photons - a significantly different process compared to daylight and incandescent lights.
Therefore, the emitted light tends to be of a very specific color, such as blue, red or green, rather than a combination of colors.
Chemicals called phosphors are applied to these devices in order to tweak the spectrum in a way that the resulting light appears white. Unfortunately, this resulting spectrum is oftentimes far different from the natural light sources we are used to.
The light emitted from the light source might appear bright and white, but once you shine it onto an object, the reflected color may appear inaccurate.
This is where CRI will help explain the likelihood that a light source shining on an object will allow it object to appear accurate.
We discuss this question in detail below.
The transition from natural to artificial lighting
Our discussion of color accuracy must begin with clarification on natural vs artificial lighting.
Human vision has primarily evolved and developed under natural light sources - the sun during the day, and fire at night.
Over time, however, as humans invented a range of artificial lighting technologies that allowed for illumination in both dark locations as well as evening hours, it became evident that some are more similar to natural light sources than others.
The invention of gas lamps and incandescent lighting may put us in "artificial lighting" territory, but in terms of color and spectrum, we are still not too far off from natural daylight and fire.
In incandescent and halogen lighting, light is generated by heating an object (the filament) to an extreme temperature such that it begins to emit heat. This is why incandescent light bulbs are so inefficient - its filament needs to be warmed to a sufficient temperature before it begins to emit enough light.
This temperature, commonly known as color temperature, happens to be 2700 Kelvin for incandescent lighting, and 3000K for halogen. On the other hand, the temperature of the sun is approximately 5500 Kelvin (about twice as hot as the filament in a light bulb) and therefore takes on a bluer tint.
The mechanism behind incandescent light bulbs and the sun are similar in that they both emit light as a result of an object being warmed to extreme temperatures.
They exhibit different colors because of the difference in physical temperature, but because of the similarity in mechanism, we perceive incandescent lighting to also be "ideal" the same way we consider daylight to be "ideal" despite having a different color.
With the advent of fluorescent and LED lighting, however, we take a big step into "artifical lighting" territory. The main reason for this is that the mechanism behind generating light is completely different.
Why would a light source be inaccurate?
The first type of color accuracy asks: how similar is this light's emitted color to a natural light source? Our answer here uses chromaticity coordinates (CCT & Duv) to analyze.
The second type of color accuracy asks: how similar do objects' colors appear under this light compared to a natural light source? Our answer here uses CRI and other color quality metrics to analyze.
The questions are a bit nuanced, so it can be a bit tricky. The first type is concerned with the light coming out of the light. The second type is concerned with the color reflected off of the object it illuminates.
Accuracy type 1: emitted light color accuracy
The first type of color accuracy is concerned with whether the emitted light color is accurate when compared to a natural light source.
But before we can ask this question meaningfully, we need to determine what "natural light source" we want to reference.
If you're working on a painting at night, and want it to appear the same under natural daylight, you will need a light bulb with emitted color that matches that of natural daylight.
On the other hand, if you're working on a painting and care about how it appears in an art gallery with halogen lighting, you will want to use a light bulb with emitted color that matches halogen lighting.
We use CIE 1931 xy coordinate chart to define the color points of our natural light source, and the color of the light emitted from the artificial light source.
By comparing the proximity between the natural light source and our artificial light source, we can determine how similar, or accurate, the two light sources are in their emitted color.
Accuracy type 2: reflected color accuracy
We describe the second type of accuracy as the ability of a light source to reveal an object's colors the same as under natural lighting. This is the core definition of the color rendering index.
Under this definition, we assume that the artificial light source's emitted light color is already deemed "accurate" and is what we are looking for.
However, that is not to say that the colors of the object will appear "accurate."
To extend our painting example above, we may have already determined that our light bulb's emitted white light is satisfactorily similar to natural daylight or halogen.
But once we shine the artificial light source onto the painting, do the colors look accurate? Do they look the same as under our natural light source?
In order to determine the CRI, various standards and procedures have been developed. Specifically, the CIE has developed what is called the Illuminant D series to simulate daylight.
These daylight standards are used so that we can ask ourselves, "how accurate is this light source when compared to daylight?"
Similarly, black body radiation can be used to simulate warmer color temperatures less than 5000K.
Why do some artificial light sources have low color rendering accuracy?
When artificial light sources emit light, they may be missing significant portions of the visible spectrum, even if the light appears white.
Instead of emitting light as a result of heating, LEDs emit light as a result of electrons being converted to photons - a significantly different process compared to daylight and incandescent lights.
Therefore, the emitted light tends to be of a very specific color, such as blue, red or green, rather than a combination of colors.
Chemicals called phosphors are applied to these devices in order to tweak the spectrum in a way that the resulting light appears white. Unfortunately, this resulting spectrum is oftentimes far different from the natural light sources we are used to.
The light emitted from the light source might appear bright and white, but once you shine it onto an object, the reflected color may appear inaccurate.
This is where CRI will help explain the likelihood that a light source shining on an object will allow it object to appear accurate.
Other Posts
Browse Waveform Lighting Products
A-Series LED Bulbs
Our A19 and A21 lamps fit in standard lamp fixtures and are perfect for floor and desk lamp fixtures.
Candelabra LED Bulbs
Our candelabra LED bulbs offer soft and warm light output in a decorative bulb style that fits E12 lamp fixtures.
BR30 LED Lamps
BR30 lamps are ceiling lamps that fit in residential and commercial fixtures with 4-inch or wider openings.