However, theory and practice are the same *only* in theory: in practice, they can be very different.
Practical results are what drive decision - if results *dont’* matter then it doesn’t matter what you do.
Hey whitebeard. Appreciate your very respectful point of view.
In practice they can be different. This is due to the issues of variabilities not accounted for. But this is not always the case and this depends on the area in which this is applied with. Many of todays technological advancements have been made through theoretical data only approaches. It has been proven many times through physics, electronics, data communication and more, how previsional data from theories, can predict a certain result. This makes the importance of theoretical data very important and useful. And is continuing to be applied on a daily basis.
Of course one can always say that in practice it may not apply. But you have to be aware with what you are referring to understand whether that will be true or not.
As far as light technologies go, this is very quantifiable and predictable due to the current theories and technologies available to us. Which through the work of lighting physics and measurement equipment we can ascertain accurate and usable data that will be reproducable in the field and used for predictions. This is how we can use theoreticle data and laws, to predict how light will behave in specific environments. Which can be used to calculate how and where lights can be placed in a commercial facility to obtain very specific levels of radiant or luminous flux.
The kind of theoretical data i am talking about is not mathematically calculated data. It is raw measured data with the use of radiometric equipment such as integrating spheres. Highly complex and expensive gear. But it measures the full capacity of a light source, including all accounted variables such as reflection and internal air losses, uniformity and can precisely quantify data such as electrical efficiency, photon efficiency, radiant and luminous flux and much more. All these are data which can be used to accurately predict certain variables with a given application. And has been used as such for a long time.
There is the issue of how plants behave to light, with respect to light quality. Which through my extensive research of photobiology and the work from researchers such as bruce bugbee, we can say with great certainty how plant anatomy and growth responds to specific wavelengths of light. As of the last 5 years, many lab studies have shown that light quantitiy is the biggest influence on plant anatomy and growth. And the quality of the spectrum is so very insignificant with multispectrum light sources. The only time light quality becomes a factor, is when monochromatic light sources are used. Which is not a standard type of technology used except for in non vegetative stages where plant morphology characteristics are abscent. The data from these researches have been used to develop some of the best grow lights for commercial applications today. Some examples are fluence, california lightworks, hortlix, and more. I know this for a fact because i have relayed emails with some of these companies, discussing their R & D.
As it stands, ceramic metal halide has not been replacing sodium discharge technology for commercial facilities. As the lighting engineers that work for these major companies are well aware of the capability and practicality of current technologies. The only technology that is looking to do so, is LED. Which based on haitz law, will inevitably happen with time. There is most certainly a place for ceramic metal halide for the hobby industry, as it adds a certain added practicality.
The Philips lamp is rated the highest. As the guide will tell you not the best choice for flowering. I bought this highly recommended system. 
