Is there a simple experiment I can perform which clearly demonstrates the difference between my high- and low-CRI torches?
Not a test, but skin colour on say the back of your hand looks really different/better with high CRI lights containing more red in their spectrum.
Works best if the LEDs are the same CCT.
Don’t worry if you can’t see a difference, just use whatever LED you like the look of… I’ve “low CRI” 4000K XPL-HI emitters that must have won the tint lottery, and look nicer than the high CRI SST20’s or Samsung LH351somethings I’ve compared them to.
Generally the Nichia 519a LEDs I have all look great, I only own the high CRI versions though…
People get worked up about what’s “best” forgetting that human vision is very personal and not everyone agrees on “best”.
It’s usually reds and browns that don’t show up well under low CRI light, so anything red or brown will show the difference. Rubik’s cubes, red apples, childrens toys - wooden surfaces are also good. If you shine one light then the other on your chosen surface you should see a real difference.
As gravelmonkey mentions though CCT also plays a part, I have a couple of lights with 5000K SFT-40s, they’re not high CRI but colours look much more natural under those than they do under a cold white version of the same LED.
Is that the same CCT?
3500K R9080, 3000K R70. Skin looks very similar imo. Skin only really seems to work on cold to neutral CCTs, with warm CCTs low and high CRI make almost the same skin tone. The camera heavily exaggerates the greenish skin color of the S21E, to my eyes it looked almost the same.
Did you try a skin color test with your own lights? Do you see a difference?
Another way to test is to shine at wood grain, can you tell which light is High CRI:
I need to dig out two lights with the same CCT. Will be interested to see the results.
My first test… Convoy S21E (5000K, 519A) vs Sofirn SC31 Pro (5000K, SST40).
I’d never noticed how disgusting the SC31 Pro was until I did this comparison! Oh God, I’m becoming a CRI snob…
That is a point worth bearing in mind - you didn’t notice it until you did a side by side comparison. I’m prepared to bet that in a couple of hours time if you pick up the SC31 Pro and just use it, it’ll be just fine.
Not really… it depends on what you want to use it for. If you have no idea yet… just start somewhere in your house, or point it to your skin. My favorite is actually pointing it to some print paper or anything really white. My preference goes to cooler white, since whites are closer to what we see during daytime.
Grab a 2700K 90CRI LED, and you get an orange hand (CCT).
Grab a 5700K 90CRI LED from Samsung, and you get a green hand (positive duv) ![]()
I woud suggest getting a few flashlight with diferent color temperatures, and then see what color temperature you like. Cooler vs Warmer.
And then include LEDs with and without CRI.
Don’t overthink this… it’s really not very important for most use cases.
I was actually testing a high CRI light a few days ago, and pointed a 519A to a fake (pink) flower, and the flower was way too bright/deep pink, compared to what I would see during day time. So enjoy the learning, and decide for yourself what is or is not important.
Oh, and one more idea I just discovered. This is using HS21 headlamp: one source is high-CRI and the other is much less so. True, they are also not the same CCT (some 5000K vs. some 6600K), but I wonder if in this particular situation, even if both sources had warmer tints, you could see the difference:
So the test is: find a light pink wall, as I did :-), maybe with an object hanging from it to get some colour anchor, and illuminate it with low and high CRI source. The wall appears white in one and much more pink in the other.
Tint plays a difference here also, the 519a is -Duv (pinkish) and the SST40 +duv (greenish).
The CRI plays less of an important role, in my opinion.
Can you have pinkish LEDs with low CRI, or low R9 at the same time?
yes
Tint DUV is an independent variable from CRI Ra and CRI R9 (red)
There are LEDs w Low CRI and either positive or negative DUV
There are LEDs w High CRI and either positive or negative DUV
Tint DUV is also an independent variable from Color Temperature (CCT)
However as a general rule of thumb,
-
Low CRI LEDs are often Cooler CCT and have Greener Tint DUV
-
High CRI LEDs tend to have less Cool CCT, and less Green Tint DUV
To reduce confusing CCT, CRI and DUV, it helps to have an instrument that can measure each variable and report them separately.
When outdoors at night, in environments with lots of green plants, Low CRI, Cool White, w Green Tint, works fine… it just makes the Grass Look Greener ![]()
otoh, when looking at things with Red pigments, such as some foods, High CRI will make red foods look more colorful and appetizing.
In regard to different Color Temperatures, Cooler CCT will more closely match daylight. Warmer CCT will more closely match Incandescent light, candle light, and fire light.
Trivia… Sunlight is nominally 5200K, but varies based on cloud cover. Moonlight is nominally 4500K, but varies by the amount of atmosphere and pollution the light is moving through.
Sun and Moonlight from closer to the horizon, travels through more atmosphere and looks warmer. When Sun and Moon are overhead, they travel through less atmosphere and look cooler.
That’s primarily a tint comparison, not a CRI comparison.
I used to chase high cri leds. Once it clicked with me that cri is often showed as a composite score it really made sense how there can be leds with really ugly green tints yet still high cri. At that point I became more focused on duv as a metric.






