Saturation masking allows you to mask color in any image so you can work on distinct colors separately withing the image. This is different from working with separated colors as you do when separating RGB tones, because you are isolating colors that may have information in red, green, and blue components all at the same time.

I don't understand the distinction there. In addition, I'm not even sure what the phrase "because you are isolating colors that may have information in red, green, and blue components all at the same time" refers to: saturation masking or RGB separations? I'm guessing that it refers to saturation masking:

This is different from working with separated colors as you do when separating RGB tones because with saturation masking you are isolating colors that may have information in red, green, and blue components all at the same time.

However, I don't understand the reasoning because when you separate RGB tones, the tones in say the blue channel may have come from areas of the image that also have red and green components present all at the same time. Also, it seems to me that when you use a Hue/Saturation adjustment layer to desaturate "Blue" in an image, you isolate the blue channel just like you would with RGB separations. Or, maybe desaturating "Blue" with a Hue/Saturation adjustment layer works differently than I understand. Does it desaturate only blue components of RGB? Or, does it completely desaturate any pixel that has a blue RGB component.

Could someone explain the difference between saturation masking and working with separated colors like when you do RGB separations?

As far as I can gather, you are getting the wrong end of the stick. Saturation can refer to a range of tones from all 3 (or how ever many your colour space allows) channels.

It may be that mid green has the same saturation value as mid blue and as a result both will be altered with the values you assert.

Think of it as, including/excluding colours for the mask based on their saturation range and not their specific colour. Its just another way of measuring something at its core. In some cases you use hex values in others the saturation.

Hope I havent got hold of the wrong end of the stick;)

Hi,

Thanks for the reply. The way I understand it now, my confusion had to do with not realizing there were two different color spaces involved: RGB and HSB.

If you turn off the blue channel in RGB, it desaturates the B(blue)component of every pixel in the image. On the other hand, with a Hue/Saturation adjustment layer, you are dealing with pixels in the HSB(Hue, Saturation, Brightness) color space. So, when you desaturate "Blue", you are desaturating a 60 degree arc on the color wheel with pure blue in the middle of the arc. You can change the range of the arc and make it bigger to include adjacent colors or smaller to limit the blues that are included, but when you desaturate the arc, only those pixels in the image that have the exact same color as pixels within the arc are desaturated--all the other pixels remain unchanged.