RAGB LED lighting and photography ideas wanted

Hi All,

I've just built a 4R4A4G4B LED array, to investigate diversifying into RAGB (red/amber/green/blue) LED lighting rather than pure RGB lighting. I had the idea that RAGB LED lights would be better for photography/videography than RGB lights.

Any suggestions for suitable test examples to test if the RAGB lighting is better than RGB lighting and compare to flouro/incandescent.....if results are good I'll track down a photographer with suitable 'studio' lighting and filters to see if these devices will be handy in that market.

Some examples I've thought of already:
- human subject, skin tones
- box or colourful crayons or pens
- a cooked meal (ie if RAGB/RGB lights are used in a restaurant...we don't want the food looking strange)
- any others???

Any ideas on the best way to examine the results....preferably in an objective way (eg can you determine the CRI using a digital camera+paint shop pro??), but some subjective areas to look at would be OK.

Lastly, would it be better to look at RAGB lighting to replace incan+filters and to try to reproduce 'normal' sunlight/incan light or should I look at new ways of using this lightings abilities....any help from professional photographers would be handy here.

Phil

rgbphil wrote: "Any suggestions for suitable test examples to test if the RAGB lighting is better than RGB lighting and compare to flouro/incandescent....."

The photography industry standard color checker is the

Macbeth ColorChecker Chart

The gretagMacbeth site has a page of various color checker charts

Big can of worms here.

The short answer is that the more wavelengths of light you have available in your light source, the better that light will be at rendering color. If a wavelenght of light that your colored surface reflects is not present in your light source, your surface won't appear as it does under a full spectrum light. But the eye is neither film nor camera, the wavelength response of your film/camera may be such that the missing wavelength would not be recorded anyway.

So to the eye, assuming no color blindness, RAGB will be better at color rendering than RGB. And a light using red, amber, yellow, cyan, blue, turquoise, and green LEDs will render color better than RAGB. Better color rendering would come from a high Kelvin Tungsten lamp source, or perhaps ceramic metal halide, and some "temperatures" of sunlight are even better than that.

Unfortunately the CRI of a light source is based upon the subjective opinion of a small group of people. Spectrographic analysis of the light source would be a better and more reproducible metric, but still won't really tell you how a given light source will interact with a given colored surface unless you've also analyzed the light reflected off of the surface from a full spectrum source.

Robert Gerlach did some research on a 7 color LED based light source sponsored by USITT and published in TD&T, Vol. 39 No.4 (Fall 2003). It would be worth a read. I don't know if Joe Tawil of Great American Market has written any articles on the subject, but he occasionally gives a talk on light and color that is probably the most informative lecture on the subject I've ever seen.

One of the props we use for the "light and color" demos that we do for the design students is a large board with painted areas. Paint an area with paint made from a pure secondary color pigment like green, next to an area painted with paint mixed from blue and yellow pigment to match the green under a full spectrum light source. Repeat with other secondary colors. This will show very interesting variations in color under light sources that are not full spectrum.