General Blending Options

Ordinarily, layers higher in the Layers panel obscure those below them. Likewise, painting in Photoshop deposits color that overwrites what was there previously. However, there are ways to cause layers and brush strokes to blend with what’s below.

Opacity

If we lower a layer’s Opacity, we can see through it. If the layer’s Opacity is 50%, it is literally half as visible, and the layer(s) below makes up the other half. The same is true for any tool that uses a brush metaphor (the Brush tool is the most obvious, but others are painterly too). Both layers and tools that use brushes have other settings that seem to do the same thing as the Opacity setting. For layers, it’s Fill, and for painterly tools, it’s Flow. We’ll discuss Flow in the “Brushes & Painting” chapter (see “Controlling Transfer: Opacity and Flow” on page 261). Fill is identical to Opacity unless Layer Styles or certain advanced blending options are used. More on that in the next section. There are several ways to adjust a layer’s opacity:

• Click the word Opacity, type a value in the now-highlighted Opacity field, then press Enter. This is the least efficient way.
• Click the small arrow to the right of the field, then use the resulting slider. This, too, is rather slow.
• Rather than clicking the word Opacity, press and hold it, then drag left or right to “scrub” the value in the field. Release the mouse when you have the desired value.
• If the currently selected tool doesn’t have its own Opacity setting, you may simply tap numbers on your keyboard: 3 produces 30% opacity, for example. Typing two numbers swiftly uses both for a more accurate result: 8 and 3 yield 83% opacity. If the active tool does have an Opacity setting (like the Brush tool), then typing numbers changes the opacity of the tool rather than that of the layer.

Blend Modes

Blend modes are another way to make layers visually interact. They are often both wonderful and confusing. Names like Multiply and Difference may not be intuitive—unless you know what’s meant by those terms. Even then, it is often profitable to try several blend modes to see which might yield the most pleasing result. In the table below, X indicates a general purpose for which a mode is used. A lighter X means that the mode may technically be in a category, but may not work as well as others in it. Note that some blend modes work only when painting or using the fill command. “NC” refers to a mode’s “neutral color,” one that vanishes when the mode is applied. If a mode has a neutral color, it will be black, white, or 50% gray.

You’ll notice that several have names derived from arithmetic (Subtract, Multiply, Divide). Indeed, most of the blend modes use a calculation involving each pixel in the layer to which the mode is applied and the one below it. Sometimes it’s useful to know what that calculation is, but most often, a metaphor serves better. As I say to my students, “definitions may be necessary, but they’re not necessarily definitive.” Thus, I’m going to define most of the blend modes, showing examples of the most useful ones. Your projects and workflows will inevitably lead to your own favorites.

In the following groups of examples, I’ll show a consistent pair of layers, one slightly offset from the other, so the blend modes can be compared. The top layer, a photo of a plush fruit named Peachy, is the one to which I’ve applied the blend modes. In some cases, I’ll show additional, practical examples as well. In each grouping, I’ll start with the most useful or definitive blend mode, then show the others that appear in the same section of the blend mode menu. As the table above shows, some modes can be put in multiple categories, but I’ll show them with others nearby in the menu.

Advice: Experiment with blend modes as often as you can! In time, you’ll have a visual memory of many of the main ones, and you’ll know which one (or few) to try in a given circumstance.

Special Cases

These few (Normal, Pass Through, Dissolve, Clear, and Behind) don’t really “blend” layers the way the others in this long list do. Let’s get them out of the way first.

Normal

Every layer’s default blend mode: Normal means no blending at all.

Pass Through

This is the default mode for groups of layers. The name means that the layers within the group blend with layers outside the group in exactly the way they would had they not been grouped at all. That is, each grouped layer blends with layers inside and outside the group equally.

Even changing a group’s blend mode to Normal can make things look significantly different. This forces the layers in the group to blend only with each other. Think of this as a sub-composition within a larger document. In the past, we would have had to merge those layers together, losing the ability to edit them individually. A group with a blend mode other than Pass Through may be a substitute for a Smart Object in certain circumstances.

Dissolve

Dissolve is the only blend mode that requires another setting to be changed to see a result. With an Opacity of less than 100%, Dissolve causes a scattering of the layer’s pixels (or those of a group or Smart Object) to disappear. If the Opacity is set to 70%, for example, 70% of the pixels remain fully opaque, and 30% completely vanish in a seemingly random scatter. In fact, most features that render partial opacity (erasure or layer masks) will give us Dissolve’s effect, as seen in this example.

Clear

Clear is available only with the Brush tool, the Pencil tool, and the Fill command. It’s essentially erasure. That’s it: it deletes!

Behind

Like Clear, Behind works only when painting or filling. In this example, I painted on the “back” of this fading version of the Peachy layer by setting the Brush tool blend mode to Behind.

To Be Technical…

If you wish to have a deep understanding of what the rest of the blend modes actually do, you’ll have to know a little bit about how Photoshop does math with color numbers when you apply a blend mode. For example, when we use the Multiply blend mode, just what numbers are being multiplied?

When blending RGB colors, instead of using the usual 256 values (0–255), the color numbers are “normalized” to a range of 0–1. When we use a blend mode called Subtract or Multiply, it’ll be numbers between 0 and 1 that are subtracted or multiplied. As fascinating as that is, the arithmetic doesn’t necessarily bring an image to mind. So I rely more on visual metaphors to explain the blend modes whenever possible. For a few modes, I’ll also let you know what math is happening.

Darkening Modes

All of these make white disappear; that is, white is the so-called “neutral color” for these modes. Multiply is the most definitive mode of this group.

Multiply

Multiply is the most useful of the modes in its section. Indeed, it’s probably the most frequently used blend mode in Photoshop.

Recall that when blending, each pixel’s colors use numbers in the range of 0–1 (instead of 0–255). When we choose the Multiply mode, those are the numbers that literally get multiplied. Since each RGB value for white would be exactly 1, white has no impact on the colors below, and essentially disappears. A fine metaphor for this effect is slide (transparency) film. Hopefully some readers will remember this twentieth-century technology. Picture two slides on a light table that’s shining light through them.

Now imagine overlapping the slides. Wherever the top image is white, the slide is clear, and we see the image under it. Where it’s black, it’s completely opaque, obscuring the image below.

Darken

With Darken, Photoshop chooses the lower red, green, and blue values for each pixel.

So if a pixel has color values of 60R, 170G, 190B (), and the pixel below it has values of 230R, 140G, 90B (), the result will be the lower of each: 60R, 140G, 90B ().

Darken often looks similar to Multiply, but because of the way it works, it could look quite different. And sometimes this mode looks nearly identical to Darker Color, but sometimes it’s radically different! When I suspect one of those might be appropriate, I always try both in case it’s one of those circumstances in which they differ substantially.

Color Burn

Color Burn differs from Multiply in that it gives more intense colors under midtones in the layer to which the mode is applied. More highlight detail is lost or diminished too.

Linear Burn

Linear Burn is darker and has more contrast than Multiply. I use this mode when the image ordinarily would demand Multiply but is a little too faint.

Darker Color

Darker Color not only makes white invisible and gives a dark result like the others in this section, but it does a direct pixel-by-pixel comparison to do so. Imagine that both the layer being blended and the one with which it is blended are completely desaturated (with the Hue/Saturation adjustment, for example). Each pixel’s luminance is then compared, and the darker of the two is shown. Of course, neither layer is visibly desaturated, but Photoshop uses that adjustment invisibly to determine the luminance values to compare.

For an interesting use case, see the description of the opposite mode, Lighter Color (page 179).

Lightening Modes

Screen

Imagine projecting two images onto the same screen from two different projectors. They would fill in each other’s shadows, giving a result lighter than either one by itself. Where either image is black, it has no effect on the other. In this way, Screen is the opposite of Multiply, and black is its neutral color since it disappears. Handy for simulating chalk on a blackboard!

Lighten

Lighten sometimes looks similar to Screen, but because of the way it works, it often looks quite different. For each pixel, Photoshop chooses the higher red, green, and blue values.

So if a pixel has color values of 230R, 140G, 90B , and the pixel below it has values of 60R, 170G, 190B , the result will be the higher of each: 230R, 170G, 190B .

Sometimes this mode looks nearly identical to Lighter Color, but sometimes it’s radically different! When I suspect one of those might be appropriate, I always try both in case it’s one of those circumstances in which they differ substantially.

Color Dodge

Color Dodge is the opposite of Color Burn: it gives a light result and hides black, as the others in this section do, and it can intensify the color of the layer to which its applied more than the one below it (as Color Burn does).

I use this as the Highlight Mode for the Bevel & Emboss layer effect when I want a strong specular highlight to help simulate chrome or another shiny substance. See “Bevel & Emboss” (page 195) for more on this effect.

Linear Dodge (Add)

Mathematically, this is one of the simplest blend modes. Linear Dodge literally adds the normalized RGB values (0–1 rather than 0–255) of each pixel. If the result exceeds 1, it’s simply set to 1 (white).

Lighter Color

Lighter Color not only makes black invisible and gives a light result like the others in this section, but it is more comparative than the others. Imagine that both the layer being blended and the one with which it is blended are completely desaturated (with the Hue/Saturation adjustment, for example). Each pixel’s luminance is then compared and the lighter of the two is shown. Of course, neither layer is visibly desaturated, but Photoshop uses that adjustment invisibly to determine the luminance values to compare.

Contrast Modes

Hard Light

Hard Light and Overlay tie for the title of “most definitive” in this section of blend modes.

In fact, Hard Light and Overlay are closely related: you can achieve the same visual result as Hard Light by inverting the layer order and using Overlay mode on the layer that’s now on top.

Think of Hard Light as a combination of Screen (when lighter than middle gray) and Multiply (when darker). Middle gray vanishes. In the window image above, a black-to-white gradient layer above the window is set to Hard Light. The center of the gradient disappears.

Overlay

You can think of Overlay as a visually less intense version of Hard Light.

You can see above that neither black nor white are truly opaque; they merely darken or lighten the pixels under them. In the crumpled paper example, the paper looks less violently crumpled. The wall surrounding the window is less obscured than with Hard Light.

Soft Light

As the name implies, this mode is the least intense of this group. Think of it as a softer version of Overlay, to which it is very similar but less intense.

Vivid Light and Linear Light

Think of Vivid Light as a combination of Color Dodge (when lighter than middle gray) and Color Burn (when darker), and Linear Light as a combination of Linear Dodge and Linear Burn. Middle gray vanishes with both.

Pin Light

When Pin Light is applied to a color, that color will replace colors below it, depending on which is lighter or darker. If Pin Light is applied to a color lighter than middle gray, it will replace colors below it that are darker than it is. If Pin Light is applied to a color darker than middle gray, it will replace colors below it that are lighter than it is.

With what does it replace the colors below? That’s slightly tricky. Consider applying Pin Light to a color lighter than middle gray. If it’s 80% of the way from middle gray to white, then darker colors below will be replaced with something ~80% of the way from black to white. The visual result is similar to (but not the same as) Lighten. The analogous case applies when applying Pin Light to colors darker than middle gray, and the result is somewhat similar to Darken.

Hard Mix

Hard Mix is brutal. The resulting pixels have red, green, and blue values of either 0 or 255! That means those pixels will be red, green, blue, yellow, cyan, magenta, black, or white. That’s it. It does this by simply adding those values for each pixel. If the total is less than 255, it’s set to 0. If it’s 255 or more, it’s set to 255. So if a pixel has color values of 130R, 140G, 45B , and the pixel below it has values of 60R, 170G, 190B , the result will be the higher of each: 0R, 255G, 0B , since both 130+60 and 45+190 are less than 255, and 140+170 is more.

Arithmetic Modes

Subtract

Although Difference is listed first in this section of the menu, the Subtract mode represents the group better. Mathematically, this is one of the simplest blend modes: Subtract literally subtracts the RGB values of each pixel. If the result is negative, it’s simply set to 0 (black).

The lighter the pixels to which this mode is applied, the darker the result, since you’re subtracting higher values from the underlying layer. If you subtract black (0), you’ve subtracted nothing and the underlying layer remains the same. The example below gives a hint why these are sometimes called the inversion modes.

Difference

The word “difference” means the result of a subtraction. And indeed, Difference mode is very similar to Subtract. The only difference (I couldn’t help it!) between the two is that this mode always subtracts the darker pixel values from the lighter ones. So when the pixels to which Difference is applied are lighter than the ones underneath, the underlying pixels get inverted.

The most common use for the Difference mode is getting two similar layers aligned or registered to each other. Photoshop’s Auto-Align Layers feature is great most of the time, but when we need to manually get content lined up, Difference can help.

Consider two identical layers, one atop the other. That is, the difference between them is exactly 0. Thus, if the top one is set to Difference mode, everything turns black.

Exclusion

Think of Exclusion as a weaker form of Difference. It uses the same rules, but produces a lower-contrast result. When two identical images are aligned to one another, the result will not be completely black.

Divide

More math! Divide produces a lightening result by dividing the normalized color numbers (0–1 rather than 1–255). The values of the pixels below are divided by the values of the pixels to which the mode is applied. If you divide by white (1), the underlying layer remains unchanged. Dividing by black, which is dividing by 0, creates a singularity that destroys civilization. Or it produces white. I’ll let you find out!

It appears that Photoshop allows dividing by 0. Those tricky engineers!

Component Modes

This group can be both very practical and wonderfully creative. To understand these modes, consider separating an image into two parts: its color and its luminosity. A simple black-and-white image is one in which the color is removed, leaving only its luminosity behind. The color is, well, everything else. And that, too, can be thought of as having two components: hue (which part of the color wheel) and saturation (how intense the hue).

I find it easier to explain them in nearly the opposite order than they appear in the menu. We’ll use the following images to explore these four modes:

Note that the image that will be our top layer, the one to which we’ll apply these blend modes, is very colorful except for the grayscale across the top. Where there is color, it’s quite saturated (except for the clouds and little else).

Luminosity

Luminosity mode essentially removes the color component, leaving only the luminosity (the tonal part) of the image. Any color we see comes from the layer(s) underneath—in this case, the blue-green below Peachy or the beige-and-brown window image below the tree layer. Where the underlying layer is either white or black or otherwise desaturated, the result is desaturated.

Color

This mode is the exact opposite of Luminosity. Color retains only the color from the layer to which it’s applied, and any tonal detail you see comes from below. The grayscale at the top of the tree image has no color, and thus gives a grayscale look to the underlying image (the window). The rich, saturated colors elsewhere get applied to that underlying image.

Hue

At first glance, Hue is very similar to Color. The grayscale makes the underlying image colorless, and tonal details come from the layer below. But note that the colors are not that saturated in the tree/window image, but are in the Peachy image. The level of saturation, like the tonal detail, comes from the layer below. Where the browns of the window image are saturated, so is the result. Since the layer below Peachy is very saturated, so is that result.

Saturation

I think you’re on to me at this point. When we set the top layer to Saturation, the hue and luminosity come from below. However intense (saturated) the bottom layer is, and however light and dark, so is the result.

Blend Modes & Smart Filters

When a filter is applied to a Smart Object, it becomes an editable Smart Filter. Double-clicking the filter’s name in the Layers panel opens that filter’s dialog so it can be adjusted. However, many Photoshop users don’t realize that the small icon to the right of most Smart Filters can be double-clicked as well, resulting in a very different dialog box: the filter’s Blending Options.

The details of filters are covered in the “Filters & Transforms” chapter of this Compendium.

In this image, I wanted to simulate the look and feel of the images made by the nineteenth-century photographers Hill and Adamson. My wife, a photo historian, and I visited some of their old haunts in Edinburgh, Scotland, to capture some of the same subjects they did. Speaking of haunts, many of their photos were made in graveyards.

With about five Smart Filters, I was able to achieve a fairly good match. But I wanted the image to look a little bit more spooky than theirs, so I added a blended blur to the image.

I applied the Gaussian Blur filter, then opened its Blending Options by double-clicking .

The effect of the blur is obvious: the image lost all focus. But when using the Smart Filter’s Blending Options, something interesting happens. I lowered the opacity of the blur to 70%. This replicates an effect in which we would duplicate a layer, blur the duplicate, then lower its opacity. So, essentially, you’re seeing a bit of the blurred version and a bit of the sharp version at the same time.

Now imagine changing the blend mode of that duplicate layer. That’s what changing the blur’s blend mode does—without the duplicate layers! I chose Overlay to give both light and dark elements of the photo a kind of ghostly glow.

I then used several other filters and adjustments to achieve this spookier version of a Hill and Adamson print.

Blend Modes & Adjustment Layers

Sometimes when we apply an adjustment to an image, we get more than we intend. For example, when I increase an image’s contrast, I sometimes dislike the increased saturation that comes with it. By changing the blend mode to Luminosity, only luminosity, and not color, is affected.

In Photoshop’s earlier days, we would try to lighten an underexposed image by duplicating its layer, and setting the blend mode of the duplicate to Screen. If the image was very dark, we might do that a couple of times. Of course, this bloated the file size. Later, we’d create an adjustment layer (it didn’t matter which one, as we wouldn’t actually make any adjustment per se) and set its mode to Screen (or Multiply for overexposed images). Even with no adjustment, this was identical to a duplicate layer set to Screen, but with little or no effect on file size.

In short, the effect of applying a blend mode to an adjustment layer is the same as adjusting a duplicate layer with that blend mode applied.

Blend Modes & Groups

By default, a group uses a blend mode called Pass Through. With this mode, layers in the group blend with layers outside the group as they would if they were not grouped. That is, those layers’ blending passes through to the layers outside the group.

Any other blend mode is very different, including Normal. The layers in the group act as if they’d been merged, with the resulting layer blending with the rest of the document. Luckily, they aren’t merged and can still be edited on their own. The effect is similar to converting the layers into a Smart Object and changing its blend mode.

In the following example, a color fill layer is blended with an image of an illustration, and the result, grouped, is blended with the image below it.