I've been studying and researching the subject lately, including real physics applied.
I've managed to get some light throught a spectrometer. I used commom lights to compare the results to other known measurements around the web and confirmed the accuracy. Note it doesnt measure the amount of light, so take that into account. For now, I have these 3 results:
This is a home fluorescent light bulb, that has characteristic spikes that may be used for calibration. As seen, it doesnt have a nice spectrum of colors although it has some light throughout the color spectrum.
So, if you put a red ball it will look red, or a green ball it will look green, and so on, but the quality of that color will not be great, it will lack some gradations or gradients between most of them. (kinoflos with CRI above 95 doesnt have that poor spectrum, of course)
- Fluo GE
- Fluo GE.jpg (342.18 KiB) Viewed 2853 times
Here is a home LED bulb, those you usually put on the ceilings fixture. It's not that China-cheap LEDs fixtures, but not an expensive one. You can see it has a much nicer spectrum of colors than the normal fluo, but it lacks some color right where the skin tones fall on the spectrum. Although not as great as the Sun, it certainly can be used and has its place.
- Philips LED
- LED Philips.jpg (317.19 KiB) Viewed 2853 times
And thats the direct sunlight, outdoors on a cloudy day. The broader spectrum of all, as already known.
(Notice the sensor has some IR cut filter so data above 650, 700, is certainly shown much lower than real values)
- Sunlight, outdoors, cloudy
- Sol_02.jpg (319.38 KiB) Viewed 2853 times
I'm waiting for some new halogen lights to arrive so I can compare to the ones I already have and other kind of light sources. Thats just a start of the study. Lately, I'm using fluos only for hair/back light. Will trust a bit on some (not high end) LEDs and probably build something out of it, maybe for fill and even key in certain cases.
Hope it can help some of you with less knowledge about lights just like me.
Will post more results when I test more.