degarb
Flashlight Enthusiast
I know there is alot of talk about cri. But in real world, I just want the blues, red, and yellow to all pop.
I got my kaidomain xpg r4 (3c?) wired up to replace a low bin china, bluish xml. (So, I tried to dedome the low bin cold xml with soldering iron, which didn't work; so I tried to pop dome off from the side which got me total phosphor removal and pretty blue.)
Anyway, this kaidomain xpg2 r4 was a horrid yellow, and not too efficient. I shown it on a blue shirt- that looks blue in any light- and it looked purple. ... Then, I was examining the skin of my caucasian children; hey they reds and yellows really looked good. So, I overlay the yellowish beam with the typical neutral xml and xpg, to find a nearly perfect color rendering of all the skin tones and blues. The hybrid beam is white but yellows and reds pop, much better than with my neutral xpgs ( my xmls' color rendering are barely passable to me, even with the t6 yellowish bt20), I HATE YELLOW LIGHT! AS BAD As BLUISH, maybe worse. The yellow beam shots makes me suspicious of the Nichia 219.
This resurrected the color mixing idea from 2008, when leds were worse at rendering color. Now before you poo poo mixing leds, remember philips L prize bulb is mixing them.
So is there a perfect led for color rendering, which is north of 140 lpw. Or I wonder what combination of bin's would get me north of 140 lpw, each led driven at 1 watt. Also, the ansi color chart means little to me other than temperature range and some shift in tint towards yellow or green: I would like a more comprehensive beam shot catalog of color v. each bin and tint, for all Cree and competing leds. I know the lottery, but isn't that mostly behind us with the silicon?
Maybe there is no one with strong convictions. So, no reply is needed, then. And, I should post the beam comparisions with mixing and each separate (time, I don't have, in theory). But, just thought I would note it for anyone's future design consideration.
I also will note that the better the two beams match, the better the mixing of color. So mixing the wide neutral xml with yellowish xpg, was no good.
I got my kaidomain xpg r4 (3c?) wired up to replace a low bin china, bluish xml. (So, I tried to dedome the low bin cold xml with soldering iron, which didn't work; so I tried to pop dome off from the side which got me total phosphor removal and pretty blue.)
Anyway, this kaidomain xpg2 r4 was a horrid yellow, and not too efficient. I shown it on a blue shirt- that looks blue in any light- and it looked purple. ... Then, I was examining the skin of my caucasian children; hey they reds and yellows really looked good. So, I overlay the yellowish beam with the typical neutral xml and xpg, to find a nearly perfect color rendering of all the skin tones and blues. The hybrid beam is white but yellows and reds pop, much better than with my neutral xpgs ( my xmls' color rendering are barely passable to me, even with the t6 yellowish bt20), I HATE YELLOW LIGHT! AS BAD As BLUISH, maybe worse. The yellow beam shots makes me suspicious of the Nichia 219.
This resurrected the color mixing idea from 2008, when leds were worse at rendering color. Now before you poo poo mixing leds, remember philips L prize bulb is mixing them.
So is there a perfect led for color rendering, which is north of 140 lpw. Or I wonder what combination of bin's would get me north of 140 lpw, each led driven at 1 watt. Also, the ansi color chart means little to me other than temperature range and some shift in tint towards yellow or green: I would like a more comprehensive beam shot catalog of color v. each bin and tint, for all Cree and competing leds. I know the lottery, but isn't that mostly behind us with the silicon?
Maybe there is no one with strong convictions. So, no reply is needed, then. And, I should post the beam comparisions with mixing and each separate (time, I don't have, in theory). But, just thought I would note it for anyone's future design consideration.
I also will note that the better the two beams match, the better the mixing of color. So mixing the wide neutral xml with yellowish xpg, was no good.
Last edited:
