White vs High CRI

Very interesting replies. I am still trying to fully understand it. Maybe someone can explain this. I build / mod mostly throwers for fun.
I had a MF04 with a XHP 70.2 P2 1A 6500K ( 70 CRI ) that burned up, so I replaced it with a XHP 70.2 N4 5000K ( 80 CRI ). I was pleasantly surprised at the down stream visibility increase because the reflection was less and this helped to see clearly. Lumens appeared less but I think it was because there was less reflection. Is this true ?
.

@CNC, was the reflected light maybe causing a glare or wash-out that distorted your vision, and with “less reflection” the long distance viewing was clearer?

The whole idea of CRI is related to how well you can see the true colors, like if you spotted someone walking out at a distance was their jacket brown, black or navy blue.

The temperature in Kelvins is related to the overall average tint color of the light, with the higher numbers toward blue-white and the lower kelvins toward orange-red.

It’s hard for me to see the real colors of things when the light is biased toward red, could be worn out eyeballs…

[edit] improper use of the technical meaning of “tint”

fwiw, Tint is not the same as Color Temperature

.

and white is not just one Color Temperature, warm white is 3000K and lower… Cool White is 5000K and higher… 4000K is Neutral White

Neutral Tint is not the same as Neutral White…
Neutral Tint is actually all along the Color Temperature line, neither green, nor magenta

there is a LOT more info on Color Temperature and CRI, in this thread, includes photos… scroll down to posts by BrokenRecordBot

I'll try to dumb it down a bit. If you mix red, green and blue light, you get white. So far so good. Imagine an object which only reflects yellow light, but no red and green. In sunlight this object will look yellow.

But in the RGB mixed light it would look dark and dull, because there's no yellow light to be reflected back.

Maybe, depending on the actual spectral reflectance of the object, it might even appear more orange or more greenish than it would be under sunlight, even if the RGB light source looks exactly like sunlight to your eyes. This is the effect of low CRI.

This is also true if you compare a high CRI and low CRI lightsource. They can have the same CCT (color temperature, for example 2700K for incandescent) and tint (green/magenta shift, pretty much perfectly neutral for incandescent) and look exactly the same to your eyes on a "perfectly white" surface, but once you introduce real world colors (inks, pigments etc.) the colors you see will be different.

I’ve seen lower numbers than that quoted. But it’s not entirely relevant. How often do you hold your flashlight directly overhead pointing straight down? How do you like that midday sun when the ground is covered in snow? When you’re holding a flashlight at night and light is reflecting back at you off of trees and or other surfaces that’s a whole lot different than the sun being straight over your head. Cooler temperatures including 5000k and above are going to reflect more light back at you off of objects than warmer temperatures. That affects your ability to see past the close objects in your light path. So we can’t compare the big flashlight in the sky to the little one in your hand.

Excellent Post!

one of the very best, explanations I have seen

Happy Holidays!

> But in the RGB mixed light it would look dark and dull, because there’s no yellow light to be reflected back

same thing happens with Red, and this is The BIGGEST difference between Low CRI and High CRI:

things that reflect red, will look Red, only if the LED produces Red, which Low CRI does not

if the light is Low CRI with Negative Red output, then the red object will look brown

here is one example:
High CRI and Low CRI Tomato sauce
.
(guess which is which)

extra credit quiz
which one tastes Redder? LOL

Very useful info from Jon Slider. I tend to agree with most of what he has to say. I’ve learned a bunch from his posts over the years.

One minor difference between him and me (which we have discussed in pms) is that I go with the photographer’s perspective on what makes neutral white.

Take a look at the image above. If you draw an oval around the white central region, I want to be on the black-body radiation line right in the middle of the white area. Roughly speaking, that puts me somewhere between the dot at 4000K and the dot at 6500K, closer to 5000K than 4000K.

sounds like the Higher CRI helped create better color separation… imagine how much better it could get with a 90+ CRI LED :wink:

have some fun with this Low CRI vs High CRI comparator:
https://techneesh.com/assets/juxt/cri1.html

I think the perception that you must sacrifice “white” to get “high CRI” comes in part from the available selection of LEDs.

  • Many people think of “white” as “cool white”. This is a misconception. A color that looks pure white under one lighting condition may look too warm or too cool in another. Human eyes have an auto-white balance which tends to correct for color temperature, to an extent.
  • Most high CRI leds were only available in warmer color temperatures. This is true for XPG, XPG2, SST-20, XPL-HI. Nichia 219B at 4500K is a bit of an outlier on the cooler side, but has low output.
  • It is only recently that high-CRI has become commonly available in some cooler tints. LH351D at 5,000K for example. But even then, these high-CRI LEDs tend not to be perfect, often featuring a greenish tint.
  • Result is some people associate “high CRI” with “not white LED”. This despite the fact that color temperature and tint (which together determine how white an LED appears), are separate factors from CRI.

@jon_slider regarding the tomatoe sauce - I thought red was more to do with R9 which is not part cri calculations…?

I thought R9 is part of CRI calculations.

It’s just that CRI is calculated based on the average score for a whole range of colors, including R9. If all the other colors score high and R9 scores low, an LED may still rate as high-CRI, because the average is high. However, without R9, reds will still look washed out or the wrong color.

I think this is why the top-rated leds for CRI on BLF are typically also the ones with decent R9: SST-20 and Nichia 219B.

EDIT: see Luxwad’s explanation in post 31. My explanation in this post is wrong.

CRI is an average score of specific (consistent) shades of color. R9 is the ninth color used, which happens to be red:

Note that each color is “Rx”, and has a rating from 0-100. Really bad LEDs can actually have negative scores, usually in R9.

However “CRI”, which is usually CRI (Ra) only covers R1-R8, so R9 is not included in that measurement. The extended CRI (R96a) covers R1-R15. In either case, a low measurement in one area may not be enough to rate an LED as low-CRI, since it’s an average value.

There are other scales, such as TLCI (Television Lighting Consistency Index), that use more & different colors.

Good to know. Thanks for clarifying Luxwad.

Oh my mistake. I thought i saw somewhere previously that it was separated for some reason.

There is CRI Ra
and also other separater bars like CRI R9

the sw45k is a 9080 LED… the 90 is CRI Ra (an average of the first 8 bars)

the 80 is the minimum CRI R9 (Red)

imo it is the lack of Red that is the MOST significant difference in the Spectrum output of Low vs High CRI LEDS

not just CRI Ra, but also CRI R9, which is the most limiting variable

some High CRI Ra LEDs, like the LH351d, are 9050… that means the CRI R9 is a minimum of 50…

the SST-20 is better in Red output, it is a 9080 LED

maukkas spectrum posts show not only CRI Ra, they include CRI R9, and also Tint DUV, in addition to Color Temperature

here is a low CRI LED with very neutral tint, but negative CRI R9… to highlight the different factors realting to LEDs… CCT, CRI Ra, Tint DUV, and the one I pay a lot of attention to, CRI R9

.
in that image you can see CRI Ra is 73 and CRI R9 is –17 (negative red output)… as well as a Tint DUV of 0.0002 (essentially smack dab on the BBL), and a Color Temperature of 4500K…

So, which sauce do you think will taste Redder? LOL

Lol, that’s a whole lot of confusion going on there :laughing: so how does it relate to tomato sauce :person_facepalming: :wink:

It does look like a good combo might be a Lower CRI emitter of your choice coupled with a separate red XPE to fill in the missing R9.

tomato sauce looks red when an LED with High CRI R9 illuminates it

tomato sauce looks brown when illuminated by an LED with negative CRI R9

CRI is 14 color bars. CRI Ra is the white bar in the spectrum graph below:

typically when flashlight sellers mention CRI, they only mean the average of the first 8 bars, hence CRI Ra is the actual complete name for that factor

so, If Im cooking, I will choose the Light with the higher CRI R9 value:

.

cause Redder food looks more appetizing and “tastes” better… than brown food… LOL

Low CRI is useless for judging how pink a steak is on the grille… High CRI is a Must Have around food.

That’s okay. Firelight2 got my point

:+1: :sunglasses: :+1: