On Jun 7, 2014, at 7:49 AM, José Ángel Bueno García <jbueno61@gmail.com> wrote:
Please explain to me what's the difference. What about designers working with Illustrator?
Sorry -- I should have been more explicit. By, "art reproduction," I mean using a camera and printer like a photocopier to create reproductions of original paintings created with traditional paints on paper or canvas. The pigments used in oil or watercolor or acrylic paints are chemically and spectrographically similar to the ones used in inkjet printers. (Note: _similar_ -- there're no exact matches and some exceptions, but almost everything rhymes.) Also, in side-by-side comparisons, you are, by definition, using the exact same illuminant. As such, a nearly-indistinguishable match is relatively straightforward. In contrast, a computer display is radically different at every step of the way. Right up front, it's using three additive RGB primaries rather than four (or several) subtractive CMYK primaries. The resulting spectra are completely different, meaning you're now relying on observer-dependent metamerism functions to create the perceptual match. Worse, the gamuts only partially overlap; each will have significant swaths of color the other simply can't reproduce. And then, to top it all off, the printed copy is being viewed with its own illuminant, an illuminant that can't even in theory be shared by the onscreen copy. You have divergence, significant divergence, at every step of the process. The wonder is that it works at all; that it actually works very well (if you're careful enough) is almost beyond comprehension. If you do everything exactly right and you've got quality equipment and fulfill all the other caveats, yes, it is (generally) possible to make print and screen (practically) exact perceptual matches...but the print will only match in the viewing booth. Take that print home and look at it under a dim incandescent bulb in your living room in the evening, and it'll look radically different. Take it to the doctor's office the next day and it'll again look radically different in the cheap fluorescent lighting. However, if the print is a copy of an original watercolor (etc.) painting, if you view the original and the copy under all those different illuminants, they'll still be very good matches for each other. That is, the original is going to look bad under the bad lighting in (almost) exactly the same way the copy is going to look bad. And that's why I don't recommend that people obsess over side-by-side print-and-display viewing comparisons (unless you have some specific and expensive reason to do so). Profile all your devices. Judge output independently. Does it look good on screen? If so, make a print. Does the print look good in daylight? If so, you're done. If you're copying a painting, don't worry if the version onscreen matches the original; only worry if the print matches the original -- and, with quality profiles, no human intervention should be necessary in that particular case. Cheers, b&