Part 1 | What is CRI and TLCI? Demystifying Lighting Metrics with the Sekonic C-800

What is CRI? Or TLCI? TLMF? SSI? TM-30? Huh? Hey guys it's Dave here from Creative Path Films and in today's video we're going to be discussing the metrics used to evaluate the quality of

light fixtures featuring the incredible C-800 spectrometer from Sekonic. My goal by the end of this video is to demystify all of these metrics so that you can better understand the lights you're using and help you to make more informed buying decisions. Now I want to

say a huge shout out to Sekonic and Kayell, their Australian distributor, for lending us a C-800.

It's thanks to them that we can bring you this deep dive into lighting metrics. Now if you're new to lighting you might mistake the C-800 for a traditional light meter like this one. But

while a light meter is designed to take exposure readings and perform exposure based calculations, a spectrometer like the C-800 is very different. It assesses the quality of light by analyzing its spectral distribution. So how does it work? A spectrometer achieves its readings by analyzing

spectral distribution. So how does it work? A spectrometer achieves its readings by analyzing the light emitted from a light source. To do this the C-800 uses a CMOS linear image sensor in combination with a linear variable filter. An LVF is an optical filter that disperses a

light's various wavelengths linearly along the length of its surface from 380 nanometers to 780 nanometers. This is sort of similar to how a prism or linear grating breaks up or disperses

780 nanometers. This is sort of similar to how a prism or linear grating breaks up or disperses light waves when a light shines through them but far more advanced. The meter detects the intensity of each wavelength and then processes this data to generate graphs and numerical values

corresponding to each of the display modes. The C800 measures brightness in lux or foot candles, color information such as kelvin, delta uv, x/y and more, as well as a light's quality using CRI, TLCI, TM-30 and SSI. But just to be clear this does not perform many of the features that a light

meter does so it is not a light meter replacement. Both of these meters have their place and gaffers and directors of photography will often be seen carrying both. Today we're going to focus on the light quality metrics and we'll explore each of these one by one. Okay once I got into the edit

it quickly became apparent that this video was going to be far too long for a single episode so I've decided to break it up into two parts. This video will cover spectrum distribution, CRI and TLCI and part two will cover TM-30, SSI and a few other useful functions. All right let's

get into part one. First let's have a look at the setup that we have here today. Next to me I have two lights each pointed directly at the sensor on the C-800. I've got an older dedo light here in full spot to provide a tungsten source as well as an Amaran 200x which is a bi-color led light. I've

put a snoot on the Amaran light to get the beam as tight as possible so it doesn't bounce around the space too much. Now even though I'll be turning our main lights on and off to take my readings this is not a perfectly controlled technical test by any means. It's more of a showcase to

illustrate the principles. So this is the main menu of the C-800 and we're going to start out by having a look at the spectrum distribution mode which you can find under spectrum. As you

can see the meter has broken up the visible light spectrum into its various wavelengths from 380 to 780 nanometers. Everything outside this graph to the left is the ultraviolet spectrum and

780 nanometers. Everything outside this graph to the left is the ultraviolet spectrum and everything to the right is the infrared spectrum both of which are invisible to the human eye. In

its current configuration the meter will assign the wavelength with the highest intensity a score of 1.0 and plot the other wavelengths from there down to zero for wavelengths that aren't present.

This graph here is a reading from our Amaran light at 5600 Kelvin. Now let's have a look at a tungsten light spectrum by firing up the Dedo. Now this is the reason many DPs, myself included,

adore old-school tungsten lights. They're what we call a full spectrum light because tungsten light includes every wavelength in the visible spectrum. Because of this it will be able to accurately reproduce virtually any color. It is a beautiful full spectrum with no spikes

or divots. I will note that the Dedo light is a slightly cooler light source than other tungsten

or divots. I will note that the Dedo light is a slightly cooler light source than other tungsten lights at 3400 Kelvin. Now let's look at how you can overlap and compare two different spectrums. For this we're going into the spectrum comparison mode. Now let's take a reading of our Amaran light

which I've dialled in as close to 3400 Kelvin as possible. I'll save this reading in the memory by hitting the memory button. This allows you to recall it when needed. Now let's grab our tungsten

measurement once again and compare the two. As you can see it's far from a perfect match. The LED

light is missing the wavelengths at the farthest side of the spectrum and contains two spikes.

The first is an orange-yellow spike, which kind of makes sense for a light emulating tungsten, but the reason for this spike is probably to compensate for the second spike in the blue spectrum. You see all white LEDs start out as blue LEDs and the manufacturers apply a layer

spectrum. You see all white LEDs start out as blue LEDs and the manufacturers apply a layer of phosphor over the top to alter their colors. They're essentially gelling the diodes to make them hit different target color temperatures. What that means is that we're always going to

have some of that original blue wavelength bleed through, which is why LEDs currently can't match full spectrum tungsten light in terms of quality. Now let's have a look at the spectrum distribution of a light source that is the opposite of tungsten and a completely putrid light in terms of quality.

A sodium vapor street lamp. This is what we call a narrow spectrum or narrow band light

source. It has one huge spike in the yellow-orange spectrum, a small spike in the green-blue band,

source. It has one huge spike in the yellow-orange spectrum, a small spike in the green-blue band, and very little color information anywhere else. Here it is again overlaid over our tungsten and LED sources. Huge difference. These sorts of light sources are terrible to work with because

LED sources. Huge difference. These sorts of light sources are terrible to work with because there's almost no color information available outside of that narrow band. You'll never be able to get skin tones, blues, purples, or other colors to reproduce because the information in

those wavelengths just isn't there. Everything is being tinted by this one narrow band of color.

Where a spectrometer is very useful though is if you're in a situation where you need to try and simulate a light source like this, which is something that I'm going to be attempting to do in another video. Now let's have a look at the most well-known metric, CRI, or color rendering index.

another video. Now let's have a look at the most well-known metric, CRI, or color rendering index.

CRI is a score that ranges from 0 to 100, with 0 being nothing and 100 being a perfect score for color reproduction. In film and video lighting, anything with a 90 or higher is good, 94 to 96

color reproduction. In film and video lighting, anything with a 90 or higher is good, 94 to 96 is very good, 94 to 96 is very good, and 97+ is excellent. One important thing to remember about CRI is that it evaluates colors and light quality based on how it appears to human vision. This

perspective is called the observer, and different metrics have different observers. Assuming human

vision as the observer introduces a bias, as it's not necessarily an accurate representation of how a camera's sensor will perceive these colors. The biggest limitation with CRI is that it uses only a very limited number of test colors to score color accuracy. CRI measures eight muted colors,

which are labeled R1 through to R8. There is another version of CRI called CRI Extended, which adds a further seven test colors for a total of 15. But most lighting manufacturers don't adopt the extended standard in their testing and marketing. Unfortunately, it's

fairly easy for lighting manufacturers to game the system and tune their lights to represent each of these eight colors very accurately, resulting in higher scores on a light that may not be accurate overall. Many colors, including skin tones, aren't well represented by those eight color patches,

overall. Many colors, including skin tones, aren't well represented by those eight color patches, so it's not necessarily the best metric to use, but it's sort of become the standard. It can be useful, however, in identifying light sources that aren't great and should be avoided. Let's have a

look at the CRI scores of our lights here today, and then we'll see how they compare. We'll start

with our Amaran light set to 3400 Kelvin. Let's fire it up and take our reading. As you can see, the C-800 measures CRI Extended, and we have our 15 color swatches represented on this page,

R1 through R15. Each of these have their own individual scores, as well as the overall CRI score, which is denoted as Ra. This light comes in at 97.7, an excellent result. Only R12 scored

under 90. Now let's have a look at our Tungsten light. As expected, an almost perfect score, 99.9

under 90. Now let's have a look at our Tungsten light. As expected, an almost perfect score, 99.9 overall, with all 15 colors individually scoring 99.5 or higher. You can also compare the CRI

results of two light sources in the CRI comparison mode. Let's jump over there and you can see our current measurement on the right-hand side. Now let's go into the memory and load up our 200x for

comparison. Here are the results. As you can see, they're fairly close. Now for fun, let's pull up

comparison. Here are the results. As you can see, they're fairly close. Now for fun, let's pull up

our street lamp again. Awful, 18.7, and some of the swatches are even scoring in the negative.

I definitely wouldn't want to light someone with one of these if I can avoid it. Now we're going to have a look at another metric called TLCI, or Television Lighting Consistency Index. Now,

an important thing to note is that TLCI does not use human vision as the observer. Instead, it uses a three-chip broadcast camera as the observer, which will see colors differently to the human eye, as well as many single-chip CMOS cameras that we use today. Instead of 8 or 15 colors,

TLCI uses the first 18 colors of an X-Rite ColorChecker Classic chart instead. Like CRI,

TLCI also measures its scores from 0 to 100. All right, let's take some measurements. Our Amaran

200x scored a 99, so an excellent score. But what about our tungsten light? Full marks yet again.

This mode also shows you three other readings in addition to the TLCI score. It shows you the CCT, as well as a measurement called Delta UV, which measures any green or magenta shifts away from the Planckian curve. A positive score indicates a green shift, and a negative score indicates

a magenta shift. There's also a companion metric called TLMF, or Television Lighting Matching Factor, which is designed to compare two different light sources to each other using their TLCI results. This metric uses the same 18 colors, but this time it also utilizes the six grayscale

TLCI results. This metric uses the same 18 colors, but this time it also utilizes the six grayscale swatches at the bottom of the chart, which helps assess any color temperature differences, as well

as green or magenta shifts. So let's recall our Amaran light once again from the memory. When you

do this, it overlays the recalled spectrum over the top of our latest reading. It also shows the CCT and Delta UV of the recalled light. You can see under TLMF, we've got a score of 93, so a very close match to our tungsten light. Now let's recall the same light, but measured at 5600

Kelvin instead. Let's see what happens. As you can see, these spectrums are quite different,

Kelvin instead. Let's see what happens. As you can see, these spectrums are quite different, and our score has dropped all the way down to a 3. If we recall our street lamp once again, it drops down even further to a matching factor of only 1. Alright, that's it for part one. Thank

you so much for watching, and please keep an eye out for part two, which will be coming out very shortly. Thank you again to Sekonic and Kayell for making this series possible.

I hope you have an absolutely wonderful day, and I look forward to seeing you in the next video.