Adjustment Methods and Principles

Over the decades since Photoshop’s first release, there have been many approaches to adjusting images with the tools it offers. In the early days, we had only destructive methods, those that truly alter the color numbers of the pixels. We ended up with many copies of images if we were cautious and feared we may need to go back to an earlier version. Currently, Photoshop offers many nondestructive methods for adjusting images or parts of them. We can readjust or unadjust at will. Sadly, many tips and tutorials you might find online continue to suggest techniques that are tedious and unnecessary. Only a very few Photoshop adjustments and functions still require these methods.

Almost all adjustments allow you to create time-saving presets that can be chosen from that adjustment’s preset menu. Use either the Properties panel menu (for adjustment layers) or the gear icon (for adjustments from the Image > Adjustments menu).

Best Practices

There are several general adjustment techniques, but some are better than others!

Direct to Pixels

When the active layer is either a Background or pixel layer, an old and usually unnecessary method of performing an adjustment is to go to Image > Adjustments and choose one of the listed adjustment types (outlined later in this chapter).

Unless you immediately undo such an action, there is no way to remove the adjustment or to even see which adjustment you just performed. A trick as old as this method is to duplicate the layer that needs such an adjustment. In this way, you can delete the adjusted layer should it prove to be problematic. This is almost never necessary! Read on for newer, better methods.

Adjustment Layers

In the previous chapter, I had a little bit to say about this: see “Adjustment Layers” (page 149). The key takeaway is that these allow us to affect many layers at once, not affect others, and to easily mask adjustments so they affect only the areas we wish them to. I will focus on this method since the most important adjustments can be applied this way.

Adjustments as Smart Filters

When we have Smart Objects in our documents, we can use adjustment layers and we can also apply adjustments like we would filters. There’s at least one filter disguised as an adjustment and it hides in the Image > Adjustment menu: Shadows/Highlights. And there are filters, like the Camera Raw filter, that are very useful as adjustments! Both are best applied to Smart Objects, as they then become editable and removable Smart Filters.

Interestingly, nearly all the items in the Image > Adjustment menu can be applied this way. So my advice becomes more nuanced (some might pun, layered): if the layer is a pixel layer, don’t use that menu, but if you make it a Smart Object, feel free!

Histograms

Although a great deal of this chapter is about color, I have not forgotten that tone is where the details are. Indeed, I create many grayscale (aka, black-and-white) images. In these, we are careful to maintain gradation where it’s important. A picture of a bride in a white dress requires us to take pains to maintain highlight detail, whereas a portrait of a friend on a snowy slope may be more dramatic if we allow the snow to lose all detail as a field of white.

To monitor how many pixels may be completely white or black, if any, we can consult an image’s histogram. Imagine this image, composed of ~10,000 pixels:

Now, imagine we take all those pixels and carefully arrange them, each stack containing pixels of the same shade of gray. Since a standard grayscale image has up to 256 levels (shades of gray), we’d have up to that many stacks, with the darkest to the left, and lightest to the right:

In this histogram, the stacks are very tiny at each end, meaning there are very few pixels that are absolutely white or black with no detail. To use a technical word, few pixels are “clipped.”

If we wish to maintain either highlight or shadow detail in an image (or both), we should monitor the image for clipping. If the darkest or lightest pixels are some distance from the ends, our image may be lacking in contrast. Admittedly, in a photo of a foggy Seattle morning, all the pixels would be huddled near the middle of the histogram.

Since each color channel is a grayscale (black-and-white) image, each has a histogram and can be evaluated for contrast and clipping. Some adjustments show us these histograms (e.g., Levels and Curves). But we can see them at any time with the Histogram panel.

Get that panel by going to Window > Histogram. I like to adjust what it shows me by opening the panel menu and choosing either Expanded View or All Channels View, and selecting Colors from the Channel menu at the top of the panel. Just below the Channel menu, the three color channels are superimposed so we can discern each one. Clicking or dragging in the histogram shows data about those specific levels—not often needed, but you never know.

Bit Depth

I wrote earlier that images commonly have up to 256 levels of tone in each channel (0 to 255). Why that many? Because computers calculate with binary digits (zeros and ones, bits for short), and 8 bits provide 256 combinations. Photoshop can handle more: 16 bits/channel, for 65,536 levels or 32 bits (used differently), to give us an unlimited tonal range! This is good news, since many camera images use up to 16 bits/channel. This gives us silky transitions by preventing abrupt jumps in tone called “banding” or “posterization.” We also gain more latitude when adjusting.