>how accurately a cell phone camera can reproduce that spectra Not accurately at all. In terms of spectrum analysis, they're essentially as useless as the huma

>how accurately a cell phone camera can reproduce that spectra Not accurately at all. In terms of spectrum analysis, they're essentially as useless as the human eye. You're trying to get information from a detailed continuous function that varies wildly all along visible wavelengths (and beyond), and all you get are three data points. In fact, I've already analyzed the spectrum and I can confirm: Yup, it's red. I suppose you could try gathering the exact spectral response of each camera having filmed the same thing (something like this [1]) and combine the resulting data to attempt getting a more detailed spectrum, but it would be very difficult and likely to yield disappointingly vague results. [1]: http://www.gamutvision.com/docs/images/Kodak_KAF-10500-CX_response.gif http://www.gamutvision.com/docs/images/Kodak_KAF-10500-CX_re...