That's a funny story jon_slider..
Thanks Djozz.. so green has/needs a higher Voltage but will also produce more lumens at the same current.
So they will have a different runtime graph as well, with green dropping off quicker because of the higher Voltage it needs.
Green requiring more voltage than Red and less than Blue only applies to green emitting semis, not to green phosphor coated blues 
I’ve thought for a while that someone should take these PC color and white emitters, test them as is, then retest them all after carefully scraping off all the phosphor.
My hypothesis is that the green phosphor not only is more visually “efficient” that is radiant wattage vs lumens but also in conversion efficiency (watts to watts). I believe, based on the green nm1 test Djozz performed, that the peak current limiter in this emitter is the phosphor overheating. I will go as far as also presuming that the difference in Vf is also largely due to this. Less efficient phosphor = more heat = lower Vf. This can also be seen in some nichia and Samsung emitter tests posted around here IIRC when Ra70 vs Ra90 variations are compared
The human eye is like 2x as sensitive to green light as to balanced white light, so the phosphor could be less efficient and still get more lumens.
rated throw is calculated from actual lux rating measured, how our eyes see color has no bearing on the instrument, green just makes more lumens due to chemical composition of the chip,
Not true. The ‘lumen’ is 100% based on the human eyes sensitivity curve.
First paragraph of the Wikipedia page.
Why not to ask manufacturer?
Because you usually won't be able to reach the person who knows this stuff :) You will probably get in touch with somebody from support who know how to turn on a flashlight, and probably not more than that
the article has really nothing to do with the topic, since we are talking about lux measured by instruments, the fact that our eyes see colors differently and more sensitive to green than red, which is true, but it has no effect on measured lumens nor lux
The whole purpose of the units lumen and lux is to recreate the human eye’s perception.
So a measuring device for lux should do exactly that.
uh no, the purpose of the devices is to measure actual lumens and lux, not perceived ones. same as tape measure is there to measure actual distance, not how it looks to us. that is why we use instruments, for objective data, how we perceive it, is irrelevant as far as actual results measured.
What you don’t want to understand is that there are no “actual” lumens and lux other than the units that were created according to the human eye’s sensivity.
Again, there is no such thing as actual and perceived lumens.
lol good one, thanks for the laugh
Either you’re trolling or you’re not the brightest in perceived lumens.
or i’m correct and you are ignorant, but i do not care, believe what you want.
I’m sorry, I didn’t mean to be rude, but just read the wikipedia stuff please and then come back and tell me I’m wrong.
The lumens are already adjusted for eye sensitivity to different wavelengths according to some standard curve.
lamo, i did , and i posted it for you, maybe you need to take your own advice, and do not bother coming back when you do,
I thought lumens were weighted to human perception too but the SI unit lumen is derived from the SI base unit candela which is specified as being measured as 540THz.
So if it isn’t 540THz, which we perceive as green, it can’t be measured in lumens.
I guess white light that contains a 540THz component will register a lumens reading, and as 540THz is very close to the frequency at which the human eye is most sensitive this may be where we get the idea that lumens are weighted for human perception, but we can see light that doesn’t contain a 540THz component which would technically be measured as 0 lumens so the idea doesn’t hold.
I don’t know if light meters are weighted or if they just represent whatever eV the sensor picks up as lumens but my Dr Meter 1330B just responded strongly to the Cree 660nm XP-E2 in my Sofirn C01R (which shouldn’t produce any light at 540THz (555nm) according to Cree’s datasheet) while not registering anything from my infrared heater.
SI unit definitions: https://physics.nist.gov/cuu/Units/current.html
Cree XP-E2 datasheet: https://docs.rs-online.com/8831/0900766b8145b5b7.pdf