LED Light Panels: 82CRI vs 90CRI
Regarding: LED Light Panels: CRI 82 vs CRI 90
Flo Lights: 500 LED Panel - at CRI 90
Cool Light: 600 LED Panel- at CRI82
Does anyone have experience with 82 vs 90? Would appreciate the feedback.
The higher the CRI number, the "cleaner" the light. A CRI of 100 would have no unwanted green or other spikes in it.
With gels such as 1/8 or 1/4 minus green, you can usually correct a cheaper fluorescent unit. However, even with the gels, it's never as clean as a Kino Flo. How important this depends a lot on how you use the units. If you light everything in the scene with the same units, then in post it's easy to pull out the unwanted green. However, if you are mixing these fluorescents with clean tungsten or daylight, you will probably have a difficult time cleaning up the spikes. Pull the excess green from the scene, and the area lit with the other units will suddenly go magenta. So you end up masking areas of the scene, which gets really complicated when you pan, tilt, or dolly the camera.
director of photography
and custom lighting design
I have 3-LED 500's from Flo-Light, and as Rick states the higher the CRI value the more even the representation of a natural light source will come from the lighting unit. LED units are the future and these LED 500 units are a little green but they are so handy it is worth the effort to correct them with some 1/4 minus green gel. I will give a thumbs up to the Flo-Light LED 500 unit. I added an Anton Bauer Gold mount plate on the back so I can use them with batteries and they run almost all day on one battery.
My name is Richard Andrewski and I am the owner of Cool Lights USA, the maker of the LED 600. As many of you know, we spend a fair amount of time educating on our site as well as on other community sites and not with marketing hype but true, well-researched and backed up facts that usually culminate in blog articles and/or products as well. I don't know any other manufacturer that gives the detail behind products like we do for example. Its there to read or not read at your choice.
A few facts on LEDs and CRI to those who are interested because there does seem to be a lot of "differing" information out there on the subject and LEDs are very new in lighting so not well understood by many yet. Lets talk about two things that are in opposition: A claim of CRI 90 and needing a magenta correction for this "CRI 90" light.
1). Its practically impossible to find a 90 CRI daylight LED at this point especially in the 5mm round family as used on both these competing lights. (I would love to see a spec if you know of one but a spec supplied from the manufacturer of the LED, not from some other source like a dealer of a particular LED component). We also have seen the hard facts of CRI and how it is reported (or mistakenly reported) by LED manufacturers. Our ultimate LED manufacturer was pretty blunt about it and what the current capabilities of LED manufacturers are in general, but you can also find plenty that will tell you what you want to hear as long as a sale is forthcoming. You can find some higher CRI tungsten (3200Kish) type LEDs (very expensive) but that's not what's in question here. The two competing products in question use 5mm daylight LEDs of a certain price range to allow the finished product to be priced in the range it is. If a better one existed I'm sure we'd have seen it and be using it if it allowed the product to remain cost effective.
2). LEDs in general aren't full spectrum so they don't mix well under camera with full spectrum lighting (like real daylight) and require a minus green correction to do that in most cases. They are at a very early stage of development like fluorescent was 10 years ago. This lack of full spectrum is also related to the lower CRI of 5mm white daylight LEDs in general and why they are having difficulties in achieving a higher CRI. Are there LED products out there that are better? Sure. But they all use RGB type LEDs to approximate a full spectrum device and color temperature can be dialed in to custom mix the output. And, they are far more expensive too than those like the LED 500 or LED 600 that use a mono color LED. Thus you cannot truthfully use the marketing term "full spectrum" when selling a mono color LED panel.
3). Color temperature is the measure of white shifting toward red or blue at opposite ends of the scale comparing to a theoretical black body radiating light. Daylight and real tungsten lighting are really the only things we have that are close to this theoretical black body. Everything else (fluorescent, LEDs, HMI/metal halide, etc.) is measured with a correlated color temperature (CCT) and must mimick the black body with varying degrees of success. What this means is you can have a CCT of 5600K CRI 90 that has a very balanced light close to real daylight and a CCT of 5600K CRI 80 or lower which may have a bit of green shift that's affecting the color rendering. Both are still a 5600K CCT light. The green shift doesn't affect color temperature at all, just CRI.
4). High CRI in electronic lighting (every type of lighting that's just mimicking the theoretical black body radiator) means there is a good balance in the green and magenta shift of its spectrum and no need for extra "shifting" of the light so it renders colors well.
What all this means in a nutshell is that if a light is reported as CRI 90 or greater and someone says it needs a minus green filter--those two facts don't add up so one or the other fact is mistaken. I'll leave it to you to decide which one now that you have the above 4 facts which can all certainly be validated elsewhere.
These type LED mono color panels have been widely reported as "a bit green" across the net--not just our's but the competition as well. One poster here in fact acknowledges they need a minus green. Minus green correction gels come in handy in instances where you need to mix with other full spectrum or electronic sources. Cool Lights customers have reported using either nothing or commonly a 1/4 minus green--the opinions vary. Most often, when someone is using no correction it is the case of exclusive use of one or more LED 600s in a controlled environment with no other lighting and a digital medium that allows a custom white balance. This alleviates the green spike very well.
In conclusion, that's just the state of 5mm daylight LEDs and the products that are based on them at the moment. People that are buying our product are appreciating it but we also set expectations correctly at this point that we believe there are no true CRI 90 5mm LED "mono daylight" product out there. Anything else is not necessarily backed up with facts and the facts speak for themselves as long as you understand CRI, what it is and how its measured. Any manufacturer is of course entitled to their opinion of "high CRI" or "CRI 90" but if customers require the use of correction filters then perhaps the spec is mistaken. As is the marketing term "full spectrum" when applied to mono color LED products.
Till the day that LEDs mature as fluorescent and HMI have, we at Cool Lights like to sell LEDs on their current merits: great for portable use, where power is an issue, on the street, in a car, etc. We don't sell them however on their color rendering merits. Fluorescent and HMI/metal halide remain great alternatives available in high CRI, mix well by themselves with other lighting and are hard to beat for cost per watt/lumen of output.
For anyone interested we have an article series that's started on LEDs in my blog at the Cool Lights website. The first in the series explains why LEDs are far more powerful in output than their specs suggest and backs it up with real world examples. The next in the series will deal with color rendering and LEDs, much as this posting has.
Cool Lights USA
A brief addendum to Richard's long post: color is measured along two different axes.
One of them is the blue-red (or orange) axis. When you are talking about "daylight" or "tungsten" bulbs, that's the axis you are referring to.
Good color meters measure both. The blue-red axis is measured in Kelvin units (3200K = tungsten, 5500K is roughly standard daylight.) The magenta-green axis is measured in CC (Color compensating) units.
ALL lights emit light that can be measured along each axis. "Full Spectrum" lights are neutral along the magenta-green axis, but will be balanced either for daylight or for tungsten (blue-red axis) and be either more blue (daylight) or more orange (tungsten.) With orange gel a daylight bulb can become a temporary "tungsten" bulb, and equally a tungsten unit can be turned into a daylight unit with the addition of a blue gel. These days, we mostly add only 1/2 orange or 1/2 blue for this last conversion. Full correction looks false.
director of photography
and custom lighting design
Yes the two axises which are really color temperature and CRI. I just made a post on DVXUser yesterday where we were talking about color temperature, CRI and color meters. The Kenko KCM 3100 which I have and used to be the very well thought of Minolta III is a very comprehensive meter and about as good as you'll get out there. It tells you not only whats going on the RED/BLUE axis but also the GREEN/MAGENTA access so you know what kinds of correction you may need. Ultimately, a photospectrometer and integrating sphere can't be beat for measuring CCT and CRI accurately though. The color meter can only indirectly tell you about CRI by whats going on in the magenta/green axis. You can approximate the CRI value by looking at the amount of shift necessary.
One other thing. The question may come up: Why a magenta/green axis in addition to the red/blue one?
This is very simple. Because a predominance of green is very common in all electronic lighting. The reason being that electronic lighting is attempting to be more efficient so it's no accident that the color our eyes are most sensitive to is green. Thus the green spike makes the light more "green" in power usage as well.
Why is magenta included as the other side of that? Because you use the color opposite another color on the color wheel to counteract the original color. Magenta and green are opposite each other. Thus you'll hear about the magenta / green shift in a light. Shift more toward magenta to get rid of the green spike. White balance does that, correction filters do it and so do bulb manufacturers by boosting magenta when they want to "get the green out". Its also worth noting that since the green spike is ultimately responsible for the efficiency of electronic lighting that as we engineer it out with whatever method, we lose some output and may have to make up for that with the light harder. Many pro fluorescent manufacturers do this to make up for the reduced lumen output of higher CRI bulbs.
Hope this helps.
EXCELLENT posts Richard!!!!
Thanks for your clarifications!