High CRI... not what I expected? (Happy ending in post 118)

I would get high CRI lights, but high CRI emitters make compromises that I don't like.

Sometimes they're not as bright at lower CRI emitters, and the bigger problem is that they don't handle heat as well as some lower CRI emitters.

“Warm white” and “neutral white” refer to color temperature, not CRI. CRI is separate.

Most high-CRI leds used in flashlights tend to be 4000K or warmer. However, it is also possible to have cool white that is high CRI.

I see a clear difference in high CRI LH351D 4000K and the SST20 4000K vs. cool white Cree XP-G and XM-Ls. There is a difference not just in colour rendition but in depth perception on a subtle level (such as a PC aluminium fan grating, if it is partially full of dust I see the difference in where what I see is, i.e. behind it and at the front (dust level), whereas with the cool white LEDs it looks to be all at the front, flat like a photo. This was quite interesting. I had tested long ago with cool vs. neutral white LEDs outdoor in relation to use on bicycles, but now, for inside the house I found high CRI to be better in this respect here too, and something else: low CRI LEDs give me a feeling that something is off (missing in colours).

For me inside the house high CRI is more important than outside…

I tested with various LEDs, e.g. with a yellow spirit level (Stanley Fatmax): 1. 6000K, 2. ca. 4000K neutral white (XP-G 5B1), 3. 4000K neutral white 90 CRI LH-351D, 4. 4000K neutral white SST 20 95 CRI. The result is that between each of them there is a clearly visible difference of increase of vibrance of the yellow. It looks most washed out with 6000K and most vibrant with the SST20 4000K.

The SST20 lights up everything in a manner that is pleasant, I don’t care much for the LH351D though it’s better in some ways than the low CRI neutral white LEDs but it has a weird yellowish tint shining on white, whereas with the SST20 I don’t get that impression at all.

excellent points, I agree

the SST has higher CRI R9 (Red) than the LH351D
both have tint above the BBL
they produce more lumens than 219b,

SST matches the CRI Ra and CRI R9 of 219b

Only the 219b has Tint below the BBL, more pink than SST and LH351

any of them has better red output, aka CRI R9, than xml or xp cree…

CRI R9 Red is separate from the Pink Tint of the sw45k, which has both, pink tint and 90+ CRI R9 Red spectrum content.

examples of different amounts of Red output, aka CRI R9:

sw45k, my benchmark, just focus on the Red bar:

Nice graph Jon Slider.

CRI with high R9 looks especially nice.

But that said, none of my in-use flashlights use SW45k 219B.

I have some of those LEDs laying around, but I find the output to be just too dim. I need more lumens.

CRI is important to me … but only to a point.

R9 of the LH351D 4000K is closer to 75 at medium currents. The 5000K washes out reds a bit by comparison.

had the same experience when I received my very first high CRI light. It takes a while to get used to the light quality. Using the light in the garden, on skin, on flowers, etc will also help to see the difference.

I’ll add my 2¢ here I suppose. LH351d has not impressed me with color representation, esp with the dome on. Sliced and at high currents and a tint correction film/spray it’s a different story- but then by that point any advantage of the samsung is gone imo. Domed SST20 in 4000K under narrow pebbled TIRs are my go to for now. Best tint IMO comes from 9080 nichias with a duv around –0.0020 and 4000-5000k. My absolute favorite tint/cct/everything is an e17a quad with mixed 3500 and 4500k. For warm, 2000K e21a is my go to. For throwers, I find the osrams cw to be somehow acceptable. I have 75cri 4000K MTG2s that look great next to an FC40 95CRI due to lower duv of the CREE. I nearly puke everytime I turn on my Armytek Dobermann with 6500k XPL HD :confounded: Domed cw XHPs have the same effect. 3A/3D HIs though become bearable again. The lower duv+cct and better color-over-angle of the HI make the same CRI night and day difference to me.

Moral? R9, duv, COA, CCT, Rg, Rf/CRI in mostly that order. Any single variable can ruin it all though IMO.

I’m hopeful for lower duv bins of the Luminus to arrive in 3500/4000k. Praying for a 95cri SST40. Also hopeful the B35AM works out for us with below bbl tints.

Speaking of MT-G2s :heart_eyes: , where ya get pcbs for ’em anymore? I got a few bare emitters, need boards for ’em.

Well personally time of day (or night) doesn’t change whether seeing color is important?

I’m okay with the efficiency and output compromise, because 90+ CRI emitters now are as efficient as CW, 70 or lower CRI ones were not very long ago.

CRI at night can matter to some people. My truck has the old style HIDs (xenon) with a deep purple/dark blue color. They don’t blind oncoming traffic, but I always thought that other cars had weird colors whenever I had my headlamps switched on.

It is the same with any other light application, riding a bike, hiking, walking your dog, etc…

I can’t say I am feeling it either. CRI I think is critical as a doctor or a lab technician but for the rest of us us that walk their dogs and work on cars, not really feeling it.

I’d be interested to know if long-term exposure to lower-CRI lighting like indoor LED bulbs leads to color vision atrophy.

I get what you mean. I’ve been walking through meadows and along the edge of the woods at night lately. All the different plants and grasses at different heights with scrolling shadows look richer with a warmer tint. I don’t know how much CRI plays a role. I’d have to check back to back but I seem to get the same effect with XHP50.2 and XHP70.2 at 3000K as I do with SST-20 at 2700K.

More likely IMO that the same people not bothered by and/or unable to notice one, wouldn’t mind the other.

If something appears or looks different at different times and in different circumstances, how do you know which one is the real one. We know an apple is supposed to be red but if one led shows it as bright red and one shows it as a duller red, which one is closer to reality?

Even two people using the same light source will see the apple differently.

good questions… do some tests :slight_smile:
see what you learn

numerically speaking
If an LED does not produce red, then no red will come back from the apple to our eyes.

High CRI LEDs produce more Red, show the red pigments better.
High CRI is more Full Spectrum, closer to Real Daylight, but not the same.

the LED on the left wont show Red, it cant, because it does not produce Red.

one light completely fails to show the red content of broiled chicken:

to me, the Low CRI chicken looks disgusting…
the High CRI Chicken, is much more appetizing…
it Tastes more like Real Chicken
lol!

My inbred kids won’t be able to complain about always having chicken for dinner anymore, now they can choose between low CRI chicken and high CRI chicken.

and make you commit adultery.

That is 90% the tint and/or CCT of the two lights making a difference, and the other 10% being the high cri led producing more depe reds.

CRI is such a limited way to objectively measure light quality. R9 is just a sliver of deep red taking from a broad spectrum of light— or rather one single color sample. Just because a light has -R9 on its CRI graph doesn’t mean it can’t produce any red content whatsoever.

I bet if you compared a high CRI light with amazing R9, but green tint and high CCT, and compared it to a lower CRI light with pink tint and warm tint but only so-so R9, you’d probably walk away preferring the lower CRI light.

And on the topic of CRI as a whole— it’s a subtle quality to light. It is miniscule compared to CCT and tint. The only real way to accurately qualify how CRI affects light is to pit two light sources of the same CCT and tint against each other, but with varying CRI levels.