Frogiswrong wrote:
it is well worth converting your RGB images to CMYK
not only to follow the examples in your book, but often for color correction in general, even if your eventual use of the image is in RGB.
You have got to be kidding…
Not at all.
Certain images correct more easily, and give a better final result, when corrected in CMYK. The same is true of Lab and RGB.
CMYK is good for controlling shadows, making very fine color adjustments, and for restoring detail to highly colored items, such as flowers. Channel mixing from RGB to CMYK can add punch to certain color ranges. It’s true that some blues get fried when you convert to CMYK, but this may be largely avoided, when it matters, by using the Wide Gamut CMYK profile:
http://www.curvemeister.com/tutorials/cmyk/widegamutcmyk/wgc myk.htm Lab is good for manipulating saturation, correcting very intense color shifts that are not dependent on luminance, such as deteriorated film emulsions, and large changes in brightness without causing color casts.
Lab is also the basis of Photoshop’s selective color adjustment, and using curves in Lab gives you more control over selective color operations. Lab is the best color space for sharpening, and for getting rid of chroma noise.
RGB is good for luminance dependent color casts, such as mixed illumination, and for removing other color casts, provided they are not so extensive that the overall luminance of the image is affected.
More generally, if you are looking for a channel to use as a mask, look outside the RGB cube. You may find just the mask you need in one of the other color spaces, rather than limiting your search to RGB only. The a and b channels of Lab can be a particularly rewarding source of masks.
If you use only one color space you’re limiting your options for certain images.
Take the other color spaces for a spin. You paid for them after all 🙂 —
Mike Russell
http://www.curvemeister.com http://www.zocalo.net/~mgr http://geigy.2y.net