Are all neutrals created equally (assuming same color temperature)?

[tobrien]

so I've been wondering a lot lately about what the best neutral white LED would be. part of me says, "well if they have the same color temperature, regardless of emitter (XM-L or MC-E or XP-G etc.) it should basically be the same." but the other part of me says, "well the beam profile could possible affect things"

so what's correct? is there a certain emitter that's known to be the best neutral white? let's say we're talking about the exact same color temperature, regardless of emitter. also, the other factors such as reflector and so on are kept as close to the exact same as possible in an ideal setup.

what prompted this whole question of mine is because I know Prometheus Lights/Dark Sucks sets the MC-E as their default choice, despite it being an older LED, and I know some of y'all still like those, right?

thoughts?

[Yoda4561]

Not really, while your eyes might have trouble picking up the difference at first glance, the spectral output of two lights at 4500k can be different. Some leds also have color fringing issues that others don't, so each will behave different depending on what kind of optics or reflectors are used. Stuff like yellow fringing, a cold center spot, or vice versa. In practice, walking around at night using a flashlight, I'd say that two lights of similar color temp will be hard to distinguish unless you do a careful A/B comparison.

[Colonel Sanders]

All neutrals absolutely are NOT created equal. Two lights can both be rated at 4300k, for example, and one can lean green...while the other might be truly neutral. I have neutral lights that look just fine...until placed side by side with my Mac's neutral Tri-EDC which is by far the best example of a truly beautiful neutral that I've come across. Nothing really that wrong with the other neutrals...until they are side by side with a truly great one. Then, the difference is VERY noticeably! And of course, opinions vary.

[Th232]

As Colonel Sanders has said, two lights with the same colour temp can look completely different, here's an image from Wikipedia that describes it quite well. Note how colour temperature isn't defined by a dot but by a line (which extends past what's drawn there), so you can have a whole range of colours.

Then you'd also have CRI thrown into the mix to further differentiate them.

[blasterman]

Best way to think of this is a color temp, or CCT, is for the sake of argument just an average of various colors. That's why there's a line in the graph Th232 linked and not a point.

LEDs are mostly composed of four colors; royal blue, green, amber, and red. Those four colors can vary in relation to each other, but still add up to the same number / color temp. So, you can have two different LEDs with different amounts of the individual colors and yet the color temp or CCT is the same. They'll look quite different to your eyes.

In practice I think the issue with neutrals being inconsistent is due more to lazy binning and marketing than anything else. Cree, Rebel, and the rest *know* what's preferable in the neutral category for general lighting, and that's certainly not neutrals with an annoying green cast. However, they have zillions of 'off' bin neutrals they have to sell, and those tend to be the bargains on various web-sites. Since our eyes are most sensitive to blue-green, neutrals that are on the green tint side tend to be pretty obvious.

[Anders Hoveland]

First, color temperature does not really tell much about the type of light when the light source is not incandescent. "Color temperature" most accurately describes black body radiation light sources, like a normal incandescent bulb, halogen bulb, or the sun.

But in the case of current white LED's, I am fairly sure that two LED's with the same color temperature will generally put out the exact same type of light. Unless we are talking about RGB or those new "full spectrum" LED's, the only differences between different color types of LED's are how much yellow phosphor is used. "Cooler" color LED's have more blue light, while "warmer" color LED's have more yellow light. It is actually very simple. So in this case, color temperature just tells you what ratio of yellow to blue light there is. But you cannot compare different types of light sources (LED versus incandescent, for example) that have the same "color temperature", they just have different types of light.

Another confusing thing, light sources that have "warmer" colors (red, yellow) are said to have a lower color temperature.
"Cooler" colors, such as blue, have a higher color temperature. It is a little ironic that higher temperature black body radiators (such as stars) give off "cooler" looking light. Why do humans instinctively find blue to be a colder color?

[Gunner12]

Ice and snow is bluer and fire is yellow, and have been around far longer then color temperature measurements, so it's probably natural for us to call the lower color temperatures warmer (fire) and higher color temperature cooler (ice).

As many have said, color temperature doesn't mean better CRI. I was looking at some 1/4-1/2 watt neutral LEDs (might start making PR drop-ins) and some have 65 CRI at ~4000k temperature, and others have 80+ CRI at the same color temperature. Both are similar in output.

[Anders Hoveland]

As many have said, color temperature doesn't mean better CRI.

Higher color temperature does generally mean better CRI for blackbody radiation sources (incandescent, halogen).

But otherwise, color temperature has little to do with CRI.
Of course, obviously CRI is not going to be very good if it is mostly red-orange light, or if it is mostly blue light. So in one sense, it is not really possible to have good CRI if the color temperature goes to either extreme.

[SemiMan]

SemiMan your nasty comment toward another member has been deleted, if you feel you have knowledge to pass on, surely you can find a nicer way to communicate that knowledge without the need to belittle a fellow member. - Norm

A blackbody radiator, by definition, under 5,000K is 100CRI period. It makes no difference what the color temperature is.

After 5K, it is based on sun spectrum at ground level.

You don't get warm leds by using more phosphor any more than you get cool leds by mixing less phosphor. They will use different phosphor combinations or in the case of high CRI, multiple phospors.

[srfreddy]

Quote removed - Norm

No need to be so hostile, SemiMan... even if you are right about the CRI. I'm not sure about the warm/cool LED's though.

[SemiMan]

Quote removed - Norm

No need to be so hostile, SemiMan... even if you are right about the CRI. I'm not sure about the warm/cool LED's though.

- Unfortunately logic has been tried. Didn't work. Trying to explain things has been tried (by many posters). Didn't work. Some people read the odd Wikipedia article and the claim to be experts. Others point out the errors in what they believe. A normal person would accept their mistake. Some people, for whatever reason, will not admit their mistake and keep not only repeating their false information, but try to spread it around in hopes they get support for it.

- Certain posters who can be Googled have been recently banned on other boards. I wonder why that is?

[SemiMan]

Quote removed - Norm

No need to be so hostile, SemiMan... even if you are right about the CRI. I'm not sure about the warm/cool LED's though.

CRI is a definition. It only relates lights under 5K to a blackbody radiator and lights over that to the sun. It does not indicate the ability to render colors well across all color temperatures though. It is very difficult to see the "richness" of a blue/blue-green object under 2700K incandescent even though that light source is 100CRI. That said, depending on what you are looking at, you will notice the difference between a good 85CRI LED bulb at 3000K and a 100CRI halogen at 3000K. If either was just turned on in a room though, 95%+ of people would not think anything was wrong with the LED bulb and would just assume it was a halogen incandescent. Yes that has been tested. If a comparative test was done and people were told what to look for, then yes some (not all) would be able to tell the difference.

As you get to cooler temperatures of almost any light source, the relatively shift in intensities to more blue / blue-green, means that things that are being lit that are those colors become brighter with great color contrast. Those particular objects will become more "pleasant" to look at. Unfortunately, with many cool white LEDs that also comes with elimination as opposed to the reduction of deep reds (and the addition of some spectral holes). So now while blue and blue-green objects look pleasant, red and yellow objects look "unnatural" and washed out.

There are however very high CRI sources in both the 3,000 and 4,000K ranges, but unfortunately not really at higher color temperatures. Some of these are made with the use of multiple phosphors which gives an almost blackbody response. They achieve high CRI both in the saturated colors of R8 CRI, as well as the pastels of the extended CRI color set. Unfortunately, these tend to give up a lot of efficiency as to achieve that blackbody response, they also create light outside the visible range which is wasted and even in parts of the visible range where our luminous efficiency is low. Other high CRI sources, typically at low color temperatures, mix a particular cooler white LED with red LEDs. This fills in the missing red, but without the loss of efficiency of phosphors that have output outside the visible range and in low response ranges of the visible range. This particular measure can still achieve high CRI with the saturated colors and pretty good CRI across most of the extended CRI color set. This is how Cree can achieve 100+ lumens/watt at the fixture level with warm white.

There is work being done on other color measurement systems including CQS (color quality system) that seek to solve the deficiencies of the CRI system, one of which is that color differences are treated the same for all colors and two that currently only the smaller color set of saturated colors is used for CRI. Unfortunately, like any standard, getting a bunch of experts to agree on the same thing is like herding monkeys. It is not easy.

Semiman