AK Schmidt wrote:
Here’s one I can’t figure out: Why does replacing the luminance in lab mode with another luminance produce a different result than blending in a luminance layer? Time and time again, I’ve tried both procedures with identical files and the replacing of the lab mode luminance always produces a different result, but I really don’t think it should be that way (shouldn’t either method be producing the same results??). Any ideas
Lightness and Luminance are based on different mathmatics.
Lab’s Lightness channel is calculated using the profile of the source color space. This is a very accurate, and relatively slow calculation.
Luminance mode uses an approximate geometric calculation, converting to HSL space, and recombining the resulting luminance value with that of the target layer. The HSL calculation is relatively fast and does not involve any profile.
Normally the use of the HSL calculation is not an issue, but there are occaional quirks. For example, in CMYK, a layer in color mode will add a brownish cast to the underlying image. The same HSL mathmatics is used, BTW, for other Photoshop functions that support luminance and other modes.
does anyone else just replace their lab luminances?
It’s hard to think of a case where copying to the Lightness channel would be beneficial. For channel mixing, I would normally use L as a source channel and not a destination, since it tends to have good contrast and low noise.
The Lab gamut viewer at curvemeister uses an artificially created image with various L channel values to view a profile’s gamut at different Lightness values. <www.curvemeister.com/tutorials/gamutviewer/index.htm> —
Mike Russell
www.curvemeister.com
www.geigy.2y.net
