• Here is a link to the full explanation: https://rollitup.org/t/welcome-back-did-you-try-turning-it-off-and-on-again.1104810/

Mammoth mint white emerald green 880w

Rocket Soul

Well-Known Member
I’ve waited 3 1/2 weeks to say’s this… he thinks the study he posted proves that “blurple” spectrums provide the best yield and quality…. But it doesn’t… nor can those conclusions be drawn from a single study, let alone from the one previously posted. Delusions of grandeur
The numbers provided can be taken as an exploratory study. If you want to show proof there are a few statistical test that can be run byt you need much more numbers to actually have a shot at getting to statistical significance. The golden standard: T-test with p:0.05. this would mean that differences in mean values and variance has only 5% chance of looking different due to noise rather than signal. These test extremely dependant on variance, standard deviation and n: the amount of measurements or in this case yield per plant. Here n is 2 (in one case 1!), which would mean no firm conclusions can be drawn.
This doesnt mean that its useless info as it can orient next step/studies in the science.
I have seen vey good studies on spectrum/yield at same ppfd and the clear winner was led 3, but this study didnt include blurple. I think we have been missing out on including 640nm in growlights. Way back when lights were coming out of blurple era it seems to have been the goto red supp but since it was not as efficient as 660, especially in ppf/w it seems to have fallen out of style. Nowadays the 640 has much better efficiency so hopefully we might see more of it.
 

cdgmoney250

Well-Known Member
@Rocket Soul
I’m not saying the numbers are useless, but like you said, the data set is way too small with too much variance to draw any type of real conclusion. Exploratory is being generous if we’re out here trying to prove things definitively.

Not only that, but the plants in the study were obviously a hemp type variety based on the cannabinoid numbers. I would want to see the same study done on drug type with a larger data set, way more plants, and replicated as well tested over and over at different light intensities. I don’t believe spectrum manipulation effects THC & CBD synthesis the same.

Also, LED 4 from the study KitnerPush posted is indeed a blurple spectrum, with 0.26% amount of green wavelengths. It’s just 660nm and 450nm wavelengths. What we don’t know was ppfd the plants were grown at. From previous studies I’ve seen comparing spectrum vs yield, the blurple spectrums never match full spectrum at higher ppfd ranges in yield, cannabinoid or terpene production.

Most warm white LED’s have a pretty significant amount of 640nm already, especially higher CRI chips. I’m more interested in supplementing from 670-750 & 365nm-450nm to try to hit the full breadth of Chlorophylls, phytochromes and cryptochromes, as those are the wavelengths lacking from traditional white spectrums.
 

Rocket Soul

Well-Known Member
@Rocket Soul
I’m not saying the numbers are useless, but like you said, the data set is way too small with too much variance to draw any type of real conclusion. Exploratory is being generous if we’re out here trying to prove things definitively.

Not only that, but the plants in the study were obviously a hemp type variety based on the cannabinoid numbers. I would want to see the same study done on drug type with a larger data set, way more plants, and replicated as well tested over and over at different light intensities. I don’t believe spectrum manipulation effects THC & CBD synthesis the same.

Also, LED 4 from the study KitnerPush posted is indeed a blurple spectrum, with 0.26% amount of green wavelengths. It’s just 660nm and 450nm wavelengths. What we don’t know was ppfd the plants were grown at. From previous studies I’ve seen comparing spectrum vs yield, the blurple spectrums never match full spectrum at higher ppfd ranges in yield, cannabinoid or terpene production.

Most warm white LED’s have a pretty significant amount of 640nm already, especially higher CRI chips. I’m more interested in supplementing from 670-750 & 365nm-450nm to try to hit the full breadth of Chlorophylls, phytochromes and cryptochromes, as those are the wavelengths lacking from traditional white spectrums.
This is my thinking aswell. Nowadays theres +3.5 ppf/w 640s so theres options to use colder 80cri whites/evos in order to not suffer any efficiency loss for 90 cri but youd have to add a massive amount of redsup to compensate. This is the study i was refering to above, very interesting numbers though expanded blues/violets wasn't the purpose here, but it shows quite clearly that the "all reds are the same" doesnt hold true (so why not same on the blue side? ) and that wider spectrum has a benefit (wide red sup and widest white base were the winners all the way thru while 660 red sup only looks lacklustre, especially when looking at terps)


It has full statiscal analysis showing when were getting signal and when were getting noise. Yes, a whole lot of noise but you can still see very clear trends
 

cdgmoney250

Well-Known Member
This is my thinking aswell. Nowadays theres +3.5 ppf/w 640s so theres options to use colder 80cri whites/evos in order to not suffer any efficiency loss for 90 cri but youd have to add a massive amount of redsup to compensate. This is the study i was refering to above, very interesting numbers though expanded blues/violets wasn't the purpose here, but it shows quite clearly that the "all reds are the same" doesnt hold true (so why not same on the blue side? ) and that wider spectrum has a benefit (wide red sup and widest white base were the winners all the way thru while 660 red sup only looks lacklustre, especially when looking at terps)


It has full statiscal analysis showing when were getting signal and when were getting noise. Yes, a whole lot of noise but you can still see very clear trends
I haven’t seen this study yet, I’ll have to check it out. Thanks for the link!
 
Top