So hard to find a good high cri light nowadays

[jon_slider]

spectrum tests on the Yuji BC3030 G04
...
underwhelmed by the fabled R9080

that Yuji is a great LED!
here are some spectrum test data

the Yuji is a 4000k 9080 with a duv of -0.0030 which I think is excellent,
not quite as pink as the 219b 4500k 9080 that has a duv of -0.0055
but the Yuji is still below the BBL, like I prefer

also great is that the Yuji has an R9 CRI of 96.. this is fantastic, and similar to the 219b

personal preference for color temperature varies, I dont use 4000k 9080 (also available in 219b), because I prefer the 4500k color temperature
but I have total respect for the Yuji option. Jason at darksucks is really savvy at picking high quality LEDs. :goodjob:

IF you have both the Yuji and the 219b, I hope you post a beam photo, side by side, to show the difference in color temperature and tint.

I suggest you take the photo using a folded piece of white printer paper, like this:

 

[staticx57]

Let's play a game. Rank these, both by CRI and CRI R9

 

[aznsx]

Let's play a game. Rank these, both by CRI and CRI R9

They appear to span a big range of color temperatures, and are on a white target. Wouldn't 'ranking' CRI be impossible? (That's not a rhetorical question. I know very little about this and am trying to learn

 

[staticx57]

It is a bit of a perception not equal to reality question. But theres only two CCTs here: 4500k and 6500k

 

[twistedraven]

I just want to say something real quick about a common misconception of CRI. CRI is a measurement of how a certain CCT accurately stacks up against its black-body radiant of the same color temperature, therefor you can't really compare R9 values of two different CCTs. For example, an LED that's 3000k or 3500k, but with an R9 of only 50 might have more deep red within its output than an LED with a CCT of 4500k or 500k, but with an R9 of 70 or 80, because the lower CCT will inherently make the output warmer and trending towards the red side of the spectrum anyways.

Also, just because an LED is outputting a green-tinted light, doesn't mean it has low CRI or low deep red rendering. The highest CRI led I have that's in a flashlight is a 5600k Optisolis, with a CRI of 98-99, with really high R9. However, it is green tinted. Likewise, you can find very pleasing to look at lights that are rosy tinted, but with CRIs of only 70-80.

CRI is very much a subtle quality compared to CCT and tint, that I feel it's quite easy to put too much emphasis on it. I could very well post a picture of two leds illuminating wood, or skin, or meat... one 4000k 70cri and rosy tinted, vs the aforementioned Optisolis that's 5600k and 98-99 cri, and I guarantee most of the people would say the 4000k 70cri is higher CRI.

 

[KITROBASKIN]

Fascinating, twistedraven.

More than once, comments have been made here at CPF about the shortcomings of comparison photos regarding tint and color temperature as well as the ability of our eyes to accommodate existing conditions. On the second detached retina surgery with my other eye, one could say that the surgeon was wanting to be sure the repair work took. As a result I lost a band of lateral peripheral vision midway (vertically) on the right. It was horribly distracting and frustrating to lose that capability. But you know what? The brain adjusted and now it is rarely noticeable. Definitely need to turn the head to make sure when driving, etc....

 

[Dan FO]

My newest is a Lumintop 2.0 tool Nicha because of it's great UI, being rock solid and will take 14500s. The reversible clip is awesome because you just reverse the body for ball cap wear.

 

[jon_slider]

I could very well post a picture of two leds illuminating wood, or skin, or meat... one 4000k 70cri and rosy tinted, vs the aforementioned Optisolis that's 5600k and 98-99 cri, and I guarantee most of the people would say the 4000k 70cri is higher CRI.

I would like to see that photo

I agree that many people dont recognize the difference between low CRI and high CRI, because the difference IS subtle.

top is low CRI xp-g2, bottom is High CRI N219b:

both those chicken breasts taste the same, if I close my eyes
but, I prefer seeing the one illuminated by High CRI, it gives a more complete picture.
Low CRI omits red content information.. I happen to like Red foods... lol

otoh, green leaves look nice (and greener) under Low CRI...
foliage is possibly one of the worst ways to attempt to learn to discern red spectrum output from a light

 

[twistedraven]

I would like to see that photo

I'll grab my SC600 MK3 HI from my car later tonight and slap some Lee minus green filters on it to take a pic later.

 

[aznsx]

I just want to say something real quick about a common misconception of CRI. CRI is a measurement of how a certain CCT accurately stacks up against its black-body radiant of the same color temperature, therefor you can't really compare R9 values of two different CCTs. For example, an LED that's 3000k or 3500k, but with an R9 of only 50 might have more deep red within its output than an LED with a CCT of 4500k or 500k, but with an R9 of 70 or 80, because the lower CCT will inherently make the output warmer and trending towards the red side of the spectrum anyways.

Also, just because an LED is outputting a green-tinted light, doesn't mean it has low CRI or low deep red rendering. The highest CRI led I have that's in a flashlight is a 5600k Optisolis, with a CRI of 98-99, with really high R9. However, it is green tinted. Likewise, you can find very pleasing to look at lights that are rosy tinted, but with CRIs of only 70-80.

CRI is very much a subtle quality compared to CCT and tint, that I feel it's quite easy to put too much emphasis on it. I could very well post a picture of two leds illuminating wood, or skin, or meat... one 4000k 70cri and rosy tinted, vs the aforementioned Optisolis that's 5600k and 98-99 cri, and I guarantee most of the people would say the 4000k 70cri is higher CRI.

Thanks, twistedraven. In lieu of a simple 'yes/no' answer yet, I believe your explanation just answered my simple question with a 'YES'. Please let me know if I'm misinterpreting your post, because as I said, I'm trying to learn more about this subject.

That question was: They appear to span a big range of color temperatures, and are on a white target. Wouldn't 'ranking' CRI be impossible?

 

[twistedraven]

Yes, because CRI is a value that compares a specific CCT against an ideal black body radiant of the same CT, comparing CRI from differing CCTs isn't as meaningful as you would think.

Below is a picture of 6 different emitter setups, they are as follows:

Cree XHP35 HI
4500k
80 CRI
Unknown R9

Cree XHP35HI with 2 layers of Lee 804 minus green filters over it
4500k
80 cri
Unknown R9

Nichia Optisolis
5000k
99 CRI
93+ R9

Osram CSLPM1.TG
5700k
65 CRI
Unknown R9

Luminus SST-20 with 1 layer of Lee 804 minus green filter over it
4000k
95 CRI
90+ R9

Osram CSLPM1.TG with 2 layers of Lee 804 minus green filters over it
5700k
65 CRI
Unknown R9


All pictures were taken with a Sony A6300 and Sigma 35mm F1.4 in manual mode. 1/10 sec exposure, F / 3.2, iso 100, and in RAW format. RAW format gives me more control to fine-tune their exposure in Lightroom to closely match one another, and more importantly, adjust the color temperature of each picture's white balance to match the CCT of the emitter being used, without introducing color artifacting.

Can you guess which one is which?

 

[defloyd77]

Can we stop calling 70 CRI low CRI? I mean, since when on a scale of 0-100 is 70 considered low? Just sayin'....

 

[staticx57]

Let's play a game. Rank these, both by CRI and CRI R9

Might as well post the answers to this question then. I measured these actual lights with my spectrophotometer. So I ranked them based upon their test results not what I saw on the beamshot. So

SW45K in an HDS Rotary: CRI 96 R9 92
LED 1 is an E21A Quadtrix 6500k in a PK Paladin with some diffusion CRI 96 R9 99
LED 2 is an Optisolis 6500k in an Balaton also with some diffusion CRI 99 R9 96

Heres the numbers:

 

[staticx57]

Also interesting note, that while many here think CRI is about how much of one color there is positively affecting CRI, heres the R9 score of a Sofirn C01R which is a red light.
You might think all that red would really help but no.

 

[jon_slider]

adjust the color temperature of each picture's white balance to match the CCT of the emitter being used

Im not sure I understand how changing the white balance in each photo, produces useful info
it appears to have made the wood background inconsistent
maybe it would help to have a daylight control shot as a reference.. ?

tbh, I dont like guessing games
I would prefer to see an example of the one you like, and share its CRI, CCT, and DUV specs.

I originally thought you offered to post just 2 photos.. I find Six images, with modified white balance to boot, too complicated. I do hope you share your own personal prefferences.

Can we stop calling 70 CRI low CRI? I mean, since when on a scale of 0-100 is 70 considered low? Just sayin'....

I also feel funny calling 70 "Low", but, I dont think 0CRI is a real thing..
Ive never seen an LED w less than 65 CRI, and most Low CRI LEDs are 70CRI these days
also, it seems consistent that LEDs with 70CRI have a Negative R9 CRI component (red things look brown)

thank you for the data
I dont think it would be possible to see the effect of R9 (Red) CRI on a White target..
maybe it would have a more visible effect on showing the color of a Red object?

In any case what are your personal observations and preferences among those options?

heres the R9 score of a Sofirn C01R which is a red light.
You might think all that red would really help but no.

interesting example, definitely suprising

maybe the CRI test is not relevant or applicable, for measuring monochromatic sources?

I believe that is also true for lumens..
a meter set up to measure the output of a "white" LED,
does not have correct calibration to measure lumen output from monochromatic sources.

 

[twistedraven]

Im not sure I understand how changing the white balance in each photo, produces useful info
it appears to have made the wood background inconsistent
maybe it would help to have a daylight control shot as a reference.. ?

tbh, I dont like guessing games
I would prefer to see an example of the one you like, and share its CRI, CCT, and DUV specs.

I originally thought you offered to post just 2 photos.. I find Six images, with modified white balance to boot, too complicated. I do hope you share your own personal prefferences.

Here's the photo with the names applied:

The idea of changing the white balance for each Emitter is simple: Our eyes work differently than a camera's. Our eyes adapt and white balance quickly whenever we are introduced to different color temperatures. If I just kept my camera in its default white balance I set of 'Sunlight' (5200k), then the Optisolis would have a huge advantage, showing the most amount of contrast of color, because it is closest to the whitepoint, whereas warmer LEDs would look overly orange. I'm not sure on the exact science of how our eyes white balance, but it's definitely something to factor when comparing tints.


My original point was that CRI is such a subtle quality to light, and that tint+CC make a far greater difference. I've white balanced each of these 6 different setups, so all that's left to pick apart is tint+CRI differences. Even when white balanced equally to the Optisolis, you can see that the SST20 has more red saturation within it, because it has more red within its spectrum, but also just because it's a warmer LED. It's also a little more rosy as well. You can also see just how good you can came lower CRI sources look with minus green filters. Tint goes a looong way.

My preference is the SST20 btw. It's my ROT66 gen2-- my go to hiking/camping light.

 

[staticx57]

interesting example, definitely suprising

maybe the CRI test is not relevant or applicable, for measuring monochromatic sources?

I believe that is also true for lumens..
a meter set up to measure the output of a "white" LED,
does not have correct calibration to measure lumen output from monochromatic sources.

That one bit was meant to be humorous. You cannot measure CRI or any component without white light. Without white light you don't have a CCT, without CCT what do you compare the light's output against.

 

[defloyd77]

Can you guess which one is which?

I'll take a shot at it. Number 2 sticks out to me the most as it pops the least, I'm guessing that's the Osram. 4 and 6 to my eyes look to to pop the best, I'll guess 4 is the SST-20 and 6 the Nichia. Number 1 I think is the Cree with minus green. 3 and 5 have me stumped, but I think they're the Osram with minus green and the Cree.

 

[staticx57]

thank you for the data
I dont think it would be possible to see the effect of R9 (Red) CRI on a White target..
maybe it would have a more visible effect on showing the color of a Red object?

In any case what are your personal observations and preferences among those options?

Here you go, Here is the E21A and the 219B against each other, trying my best to hold them far enough apart. I grabbed the E21A first so I just used it, I can try and compare with Optisolis if you are curious.

What you see is a comparison mostly between CCT, the lower ~4500k has more red content and the ~6500k has less, this is expected given the difference between the two CCTs. If you look at the chart though you will see the SW45k actually produces too much red which lowers its R9 score. In the spectral output graph, red is tested blue is ideal. Look at how much better Optisolis is than both at matching its ideal source CCT. Then look at its CRI score, no wonder it scored so well.

As for my preference, what I should really do is build a multi emitter 5700k Optisolis (since each LED doesn't produce much light. Glorious sunlight. In general though I prefer the cooler temps when I want things to be bright, this gives me the most color information as each color is pretty even in its output. When I want a warmer light ill go closer to 2700-3500k. 4000-5000k is in an awkward spot these days.

 

[ma tumba]

I have a 6x optisolis mule, 5700K. 3x 5000 + 3x 6500.

this is a fantastic tint but I think I prefer the 6500k. nothing has been better to me than that

in any case, the optisolis is head and shoulders above the e21a at any cct