You need to look at the spectrums and see what the Kelvin and CRI ratings are affecting. General rule with LED, 80 CRI indicates the red peak is at 600nm. 90 CRI indicates the red peak is at 630nm. The result is 90 CRI having substantially higher levels of light in the 630-730 range than 80 CRI even when looking at 2700K -vs- 3500K for instance. Kelvin affects ratio between red and blue peaks but doesn't indicate ratio of red:deep red.
Deep red causes stretch in the vegetative period, therefore 90 CRI, even in a higher Kelvin such as 3500K 90CRI will be associated with more stretch (in veg). This is not necessarily a bad thing unless the grow will be in a cab with limited space. Most tents can handle whatever color temp you throw at them.
Typical old school HPS: 5% in the blue spectrum.
Typical 2700K 90CRI: 8% in the blue spectrum.
Typical 3000K 90CRI: 10% in the blue spectrum.
3000K 80CRI: 12% in the blue spectrum.
3500K 80CRI: around 15% in the blue spectrum.
In theory 2700/90 should produce shorter plants than HPS, not just because of the blue ratio but also because LED generally has less deep/far red than the HPS spectrum even in the case of high CRI.
Anything above 4000K when talking about LED achieves it's rating in part by shifting the orange/red peak toward green/yellow. Take a look at some 5600, 6500K spectral charts. Nothing necessarily wrong with green in the spectrum but orange/red provides the most efficient conversion of light energy to plant energy. It's questionable whether there is any point in going higher than 4000K when it comes to flowering a plant, or even vegging, not just due to the loss of orange/red but because there's no indication that boosting the blue ratio above 20% will further reduce vertical growth.
If testing was done that indicated 5600/6500 provided shorter plants, one would also want to look at plant mass, meaning they might end up shorter because there was less photosynthesis on a watt per watt basis. OTOH, if electrical use isn't a concern it might be worthwhile to do some testing with higher K samples since levels of deep red, though generally low at 70/80 CRI will be even lower at higher K temps... however I suspect the effect would be minimal...
3000K 80CRI, 3500K 80CRI, 3000K 70CRI will produce similar plant height. 2700K 90CRI, 3000K 90CRI (and likely 3500K 90CRI) will produce similar plant height.
High CRI is also associated with less stretch during the flowering period! More research needs to be done but that has been my observation.
In a test, full cycle with 90 CRI ended with plants being only 3-4 inches taller than 80 CRI full cycle, while 90CRI was a foot taller at the beginning of 12/12. With final full cycle plant height around 5 feet this comes out to about 5% difference. Plant height at the end of the veg period was about 25-30% difference.
Untested but, shortest plants possible, less or equal to 80 CRI in veg, and 90 CRI in flower. In theory this will result in shorter finish than running 80 CRI full cycle. For these reasons I suggest 4000K 80CRI for dedicated veg and 3000K 90CRI for dedicated flower.
It should be noted that observations are based on available data I'm aware of. Certain combinations are generally avoided that would be helpful in rounding out the testing data. If someone took one for the team and vegged with 3000K 70 CRI -vs- 6500K 70 CRI for instance and compared height -vs- plant mass, or compared plant height and yields between 3000K 90CRI -vs- 5600K 70 CRI for instance.
@REALSTYLES has vegged with high K so he may have some input. Testing data seems to indicate around 5% increase in yield with 90 CRI samples over 80 CRI. More testing is needed in this area. Testing data also indicates reduced flowering times using high CRI. Because of the small variances and limited testing these results also need further validation.
@The Dawg has played around with various samples so he may be able to provide some more insight.
