Photoshop 6 for Windows Bible- P23

Chia sẻ: Cong Thanh | Ngày: | Loại File: PDF | Số trang:30

Thêm vào BST

Báo xấu

61
lượt xem 4
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Photoshop 6 for Windows Bible- P23:If you are reading this foreword, it probably means that you’ve purchased a copy of Adobe Photoshop 6.0, and for that I and the rest of the Photoshop team at Adobe thank you. If you own a previous edition of the Photoshop Bible, you probably know what to expect. If not, then get ready for an interesting trip.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Photoshop 6 for Windows Bible- P23

Chapter 13 ✦ The Wonders of Blend Modes 631 If you want all layers in the clipping group to blend as a unit, select the Blend Clipped Layers as Group check box (it’s at the bottom of the spotlighted area in Figure 13-16). If you selected different blend modes for the individual layers in the clipping group, Photoshop uses the mode applied to the base layer for all layers. To adjust blending of the base layer only, deselect the check box. Now all blending options except the Opacity slider have no impact on any layer in the clipping group except the base layer. Blending individual color channels Photoshop 6 Directly underneath the Fill Opacity slider, the Channels check boxes enable you to blend color channels independently of each other. (When you’re working on a grayscale image, this option is unavailable because the image has only one chan- nel.) For some examples of the creative possibilities afforded by the Channels fea- ture, see Color Plate 13-5. Knocking out layers Photoshop 6 The Knockout pop-up menu is another of the blending options new to Photoshop 6. It turns the contents of the active layer into a floating hole that can bore through one or more layers. You specify how deep the hole goes using the Knockout option; use Opacity and blend mode options to define the translucency of the hole. Select a Knockout option depending on what layers you want to blend together: ✦ None blends normally. The layer is treated as a standard layer, not a hole. ✦ Shallow cuts a hole through a layer set to expose the layer immediately below the layer set. In a clipping group, Shallow burrows down to the base layer of the group. ✦ Deep cuts all the way down to the Background layer. An exception arises when you’re working with a layer in a clipping group. In that case, Deep blends the layer with the base of the clipping group, just like Shallow. Figure 13-18 shows examples of each of these options. In all the examples, the top layer contains black text, to which I applied the Outer Glow and Bevel and Emboss layer effects. The next layer down holds an umbrella and a few associated layer effects. I grouped these two layers into a layer set, as shown in the Layers palette in the figure. The clouds occupy the layer immediately below the layer set. The back- ground layer contains a gradient created with the new noise-gradient feature (see Chapter 6). In the upper-left example, I set the Knockout option to None for all layers. I used the Normal blend mode and set the Opacity value and Fill Opacity value to 100 percent throughout. With this setup, each layer blends normally with the one below.
632 Part IV ✦ Layers, Objects, and Text Original 25% Fill Opacity, None 25% Fill Opacity, Shallow 25% Fill Opacity, Deep Figure 13-18: Here you see how the original image (top left) appears after I lowered the opacity of the text layer to 25 percent and then applied each of the three Knockout options. To create the top-right image, I lowered the Fill Opacity value for the text layer to 25 percent and left the Knockout setting at None. You see the pattern of the umbrella through the faded text, as you would expect. The effects (Outer Glow and Bevel and Emboss) remain unchanged because they lie beyond the bounds of the filled portion of the layer and thus don’t respond to the Fill Opacity slider. For the bottom-left example, I left the Fill Opacity at 25 percent but set the Knockout option to Shallow. Now the text interior blends with the cloud layer, which is the layer immediately beneath the layer set that contains the text. For the final example, I selected Deep from the Knockout pop-up menu, blending the text interior with
Chapter 13 ✦ The Wonders of Blend Modes 633 background layer and creating the nuclear rain environmentalists have worried about for years. Knocking out by brightness value The Blend If sliders, found at the bottom of the Blending Options panel of the Layer Style dialog box, enable you to drop out pixels in the active layer and force through pixels from lower layers according to their brightness values. Photoshop 6 When you use the sliders, Photoshop applies your changes taking the Knockout option into account. For example, if you set the Knockout option to Deep, you force through pixels from the Background layer instead of from the layer immediately below the selected layer. Select a color channel from the Blend If pop-up menu to apply the effects of the slider bars beneath the menu to one color channel independently of the others. Each time you select a different Blend If option, the slider triangles change to the positions at which you last set them for that color channel. Select a color channel from the Blend If pop-up menu to apply the effects of the slider bars according to the contents of a single color channel. Choose Gray to base the changes on the grayscale composite. Each time you select a different Blend If option, the slider triangles change to the positions at which you last set them for that color channel. Regardless of how you set the sliders, Photoshop applies your changes evenly to all channels in the image; the selected channel is merely used for the calculation. Note You can select any channel regardless of the settings of the Channels check boxes. For example, if you select Red from the pop-up menu, Photoshop applies the Blend If values according to the contents of the red channel even if you deselect the Red check box. In other words, the contents of the red channel are used to calculate how the green and blue channels in the layer merge with the rest of the image. ✦ This Layer: This slider bar lets you exclude ranges of colors according to brightness values in the active layer. When you exclude colors by dragging the black triangle to the right or the white triangle to the left, the colors disappear from view. ✦ Underlying Layer: This slider forces colors from the underlying layers to poke through the active layer. Any colors not included in the range set by the black and white triangles cannot be covered and are therefore visible regard- less of the colors in the active layer. ✦ Preview: Select the Preview check box to continually update the image win- dow every time you adjust a setting. The slider bars are far too complicated to fully explain in a bulleted list. To find out more about these options, read the following sections.
634 Part IV ✦ Layers, Objects, and Text Color exclusion sliders The first slider bar, This Layer, hides pixels in the active layer according to their brightness values. You can abandon dark pixels by dragging the left slider triangle, and abandon light pixels by dragging the right slider triangle. Figure 13-19 shows examples of each. ✦ To create the first example, I first set the blend mode to Screen. Then I dragged the left slider bar until the value immediately to the right of the This Layer label read 170, thereby hiding all pixels whose brightness values were 170 or lower. ✦ To create the second example, I changed the blend mode to Multiply. I reset the left slider triangle to 0 and dragged the right slider triangle to 120, which hid those pixels with brightness values of 120 or higher. Screen, This Layer: 170, 255 Multiply, This Layer: 0, 120 Figure 13-19: Two examples of modifying the blend mode and This Layer settings inside the Layer Style dialog box.
Chapter 13 ✦ The Wonders of Blend Modes 635 Drag the triangles along the Underlying Layer slider bar to force pixels in the under- lying layers to show through, again according to their brightness values. To force dark pixels in the underlying image to show through, drag the left slider triangle; to force light pixels to show through, drag the right slider triangle. Here’s how I achieved the effects in Figure 13-20: ✦ To achieve the effect in the top example in Figure 13-20, I started off by applying the Hard Light mode. (Those blend modes, they’re keepers.) Then I dragged the left slider triangle until the first Underlying Layer value read 140. This forced the pixels in the sunset that had brightness values of 140 or lower to show through. ✦ In the second example, I changed the blend mode to Overlay. Then I dragged the right Underlying Layer slider triangle to 150, uncovering pixels at the bright end of the spectrum. Hard Light, Underlying Layer: 140, 255 Overlay, Underlying Layer: 0, 150 Figure 13-20: Here I changed the Underlying Layer slider bar settings to force through the darkest (top) and lightest (bottom) pixels in the sunset.
636 Part IV ✦ Layers, Objects, and Text Bear in mind, every single adjustment made inside the Layer Style dialog box is temporary. The slider bars hide pixels; they don’t delete them. As long as the layer remains intact, you can revisit the Layer Style dialog box and restore hidden pixels or hide new ones. Fuzziness The problem with hiding and forcing colors with the slider bars is that you achieve some pretty harsh color transitions. Both Figures 13-19 and 13-20 bear witness to this fact. Talk about your jagged edges! Luckily, you can soften the color transitions by abandoning and forcing pixels gradually over a fuzziness range, which works much like the Fuzziness value in the Color range dialog box, leaving some pixels opaque and tapering others off into transparency. To taper the opacity of pixels in either the active layer or the underlying image, Alt- drag one of the triangles in the appropriate slider bar. The triangle splits into two halves, and the corresponding value above the slider bar splits into two values sep- arated by a slash, as demonstrated in Figure 13-21. Alt-drag Figure 13-21: Alt-drag a slider triangle to split it in half. You can then specify a range across which brightness values fade into transparency. The left triangle half represents the beginning of the fuzziness range — that is, the brightness values at which the pixels begin to fade into or away from view. The right half represents the end of the range — that is, the point at which the pixels are fully visible or invisible. Figure 13-22 shows some fuzziness applied to the This Layer slider. Here are the specifics: ✦ In the top example, I set the blend mode to Multiply. I left the first This Layer triangle set to 0. I Alt-dragged the second triangle to split it. And I moved the left half of the split triangle to 55 and the right half to 128. The result is a grad- ual drop off. All pixels with brightness values of 0 to 55 are opaque, the pixels become gradually more translucent from 56 to 127, and pixels brighter than 128 are transparent.
Chapter 13 ✦ The Wonders of Blend Modes 637 ✦ Next, I duplicated my layer and switched the blend mode to Screen. After splitting the first slider triangle with an Alt-drag, I set one half of the triangle to 128 and the other to 220. I dragged both halves of the second This Layer triangle back to 255. The darkest pixels are transparent, they fade into view from 129 to 219, and they become opaque from 220 on up. As shown in the bottom example in Figure 13-22, the result is a perfect blending of Multiply and Screen, with the sunset showing through in the gray areas. Multiply, This Layer: 0, 55/128 Duplicate layer, Screen, This Layer: 128/220, 255 Figure 13-22: By Alt-dragging a This Layer slider triangle, you can create gradual transitions between the opaque and transparent portions of a layer.
638 Part IV ✦ Layers, Objects, and Text Using the Underlying Layer slider is a bit trickier. It typically works best when you’re trying to force through very bright or dark details, such as the highlights in the sun- set sky and the shadows in the water. It also helps to work with a foreground layer that has lots of flat areas of color for the background to show through. Here’s what I did to create Figure 13-23: ✦ For starters, I applied Filter ➪ Other ➪ High Pass to my thinker layer, as in the first example of Figure 13-23. This created lots of gray areas for the underlying pixels to shine through. ✦ I applied the radical Color Dodge mode to this layer. I left the first Underlying Layer triangle at 0. Then I split the second one and moved the left half to 80 and the right half to 200. This forced through the darkest pixels, fading them out as they got lighter. ✦ Next, I duplicated the layer, applied the Color Burn mode, and fiddled with the Underlying Layer triangles until the values read 100/150 and 180/255. The result is a vibrant composition that nicely sets off the thinker’s tattoos. Using Channel Operation Commands Image ➪ Apply Image and Image ➪ Calculations provide access to Photoshop’s chan- nel operations, which composite one or more channels with others according to predefined mathematical calculations. Although once hailed as Photoshop’s most powerful capabilities, channel operations have been eclipsed by the standard and more accessible functions available from the Layers and Channels palettes. One day, I suspect Adobe will scrap Apply Image and Calculations altogether. But until that day, I will dutifully document them both. The Apply Image and Calculations commands allow you to merge one or two identi- cally sized images using 12 of the 17 blend modes discussed earlier plus 2 addi- tional modes, Add and Subtract. In a nutshell, the commands duplicate the process of dragging and dropping one image onto another (or cloning an image onto a new layer) and then using the blend mode and the Opacity settings in the Layers palette to mix the two images together. Although Apply Image and Calculations are more similar than different, each com- mand fulfills a specific — if not entirely unique — function: ✦ Apply Image: This command takes an open image and merges it with the fore- ground image (or takes the foreground image and composites it onto itself). You can apply the command to either the full-color image or one or more of the individual channels. ✦ Calculations: The Calculations command works on individual channels only. It takes a channel from one image, mixes it with a channel from another (or the same) image, and puts the result inside an open image or in a new image window.
Chapter 13 ✦ The Wonders of Blend Modes 639 High Pass filter Color Dodge, Underlying Layer: 0, 80/200 Duplicate layer, Color Burn, Underlying Layer: 100/150, 180/255 Figure 13-23: After combining a High Pass effect with the radical Color Dodge and Color Burn blend modes, I used the Underlying Layer slider bar to force through pixels from the background so that the sunset and ocean didn’t get lost.
640 Part IV ✦ Layers, Objects, and Text The primary advantage of these commands over other, more straightforward com- positing methods is that they allow you to access and composite the contents of individual color channels without a lot of selecting, copying and pasting, cloning, floating, and layering. You also get two extra blend modes, Add and Subtract, which may prove useful on a rainy day. The Apply Image and Calculations commands provide previewing options, so you can see how an effect will look in the image window. But thanks to the sheer quantity of unfriendly options offered by the two commands, I suggest that you use them on only an occasional basis. The Calculations command can be a handy way to combine masks and layer transparencies to create precise selection outlines. Apply Image is good for compositing images in different color models, such as RGB and Lab (as I explain in the “Mixing images in different color modes” section later in this chapter). But if your time is limited and you want to concentrate your efforts on learning Photoshop’s most essential features, feel free to skip Apply Image and Calculations. I assure you, you won’t be missing much. The Apply Image command Channel operations work by taking one or more channels from an image, called the source, and duplicating them to another image, called the target. When you use the Apply Image command, the foreground image is always the target, and you can select only one source image. Photoshop then takes the source and target, mixes them together, and puts the result in the target image. Therefore, the target image is the only image that the command actually changes. The source image remains unaffected. When you choose Image ➪ Apply Image, Photoshop displays the dialog box shown in Figure 13-24. Notice that you can select from a pop-up menu of images to specify the Source, but the Target item — listed just above the Blending box — is fixed. This is the active layer in the foreground image. If this sounds a little dense, think of it this way: The source image is the floating selection and the target is the underlying original. Meanwhile, the Blending options are the blend modes pop-up menu and the Opacity value in the Layers palette. Using the Apply Image command is a five-step process. You can always simply choose the command and hope for the best, but you’ll get the most use out of it if you do the following.
Chapter 13 ✦ The Wonders of Blend Modes 641 Figure 13-24: The Apply Image command lets you mix one source image with a target image and make the result the new target. STEPS: Applying the Apply Image Command 1. Open the two images that you want to mix. If you want to mix the image with itself to create some effect, just open the one image. 2. Make sure that the two images are exactly the same size, down to the last pixel. Use the crop tool and Image Size command as necessary. (You don’t have to worry about this step when mixing an image with itself.) 3. Inside the target image, switch to the channel and layer that you want to edit. If you want to edit all channels, press Ctrl+tilde (~) to remain in the com- posite view. Tip Even when you’re editing a single channel, display all channels on-screen. For example, after pressing Ctrl+1 to switch to the red channel, click in front of the RGB item in the Channels palette to display the eyeball icon and show all channels. Only one channel is active, but all are visible. This way, you can see how your edits inside the Apply Image dialog box affect the entire image, not just the one channel. 4. Select the portion of the target image that you want to edit. If you want to affect the entire image, don’t select anything. 5. Choose Image ➪ Apply Image and have at it.
642 Part IV ✦ Layers, Objects, and Text Obviously, that last step is a little more difficult than it sounds. That’s why the fol- lowing list explains how all those options in the Apply Image dialog box work: ✦ Source: The Source pop-up menu contains the name of the foreground image as well as any other images that are both open and exactly the same size as the foreground image. If the image you want to merge is not available, you must not have been paying much attention to Step 2. Press Escape to cancel, resize and crop as needed, choose Image ➪ Apply Image, and try again. ✦ Layer: This pop-up menu lists all layers in the selected source image. If the image doesn’t have any layers, Background is your only option. Otherwise, select the layer that contains the prospective source image. Select Merged to mix all visible layers in the source image with the target image. ✦ Channel: Select the channels that you want to mix from this pop-up menu. Both composite views and individual color and mask channels are included. Keep in mind that you’ll be mixing these channels with the channels that you made available in the target image before choosing the command. For example, if the target image is an RGB image shown in the full-color com- posite view, and you choose RGB from the Channel pop-up menu in the Apply Image dialog box, Photoshop mixes the red, green, and blue channels in the source image with the corresponding red, green, and blue channels in the tar- get image. However, if you switched to the red channel before choosing Apply Image and then selected the RGB option, the program mixes a composite grayscale version of the RGB source image with the red channel in the target and leaves the other target channels unaffected. ✦ Selection, Transparency, and Layer Mask: If a portion of the source image is selected, the Channel pop-up menu offers a Selection option, which lets you apply the selection outline as if it were a grayscale image, just like a selection viewed in the quick mask mode. If you selected a specific layer from the Layer pop-up menu, you’ll find a Transparency option that represents the trans- parency mask. If the layer includes its own layer mask, a Layer Mask option also appears. None of the three options is particularly useful when you work in the compos- ite view of the target image; you’ll usually want to apply the Selection, Trans- parency, or Layer Mask option only to a single channel, as described in “The Calculations command” section toward the end of this chapter. (For an excep- tion, see the upcoming tip.) ✦ Invert: Select this check box to invert the contents of the source image before compositing it with the target image. This option permits you to experiment with different effects. The lower-left example in Color Plate 13-6, for example, shows one use for the Invert check box. I inverted the b channel before com- positing it with the RGB image to create an early dawn effect. ✦ Target: You can’t change this item. It merely shows which image, which chan- nels, and which layers are being affected by the command.
Chapter 13 ✦ The Wonders of Blend Modes 643 ✦ Blending: This pop-up menu offers access to 12 of the blend modes I dis- cussed in “The remaining 17 blend modes” section earlier in this chapter. The Dissolve, Hue, Saturation, Color, and Luminosity options are missing. Two additional options, Add and Subtract, are discussed in the “Add and Subtract” section later in this chapter. ✦ Opacity: I gather you’re well aware of how this one works. ✦ Preserve Transparency: When you’re editing a layer in the target image — that is, you activated a specific layer before choosing Image ➪ Apply Image — the Preserve Transparency check box becomes available. Select it to protect transparent portions of the layer from any compositing, much as if the trans- parent portions were not selected and are therefore masked. ✦ Mask: Select this option to mask off a portion of the source image. I already mentioned that you can specify the exact portion of the target image you want to edit by selecting that portion before choosing the Apply Image command. But you can also control which portion of the source image is composited on top of the target through the use of a mask. When you select the Mask check box, three new pop-up menus and an Invert check box appear at the bottom of the Apply Image dialog box. For complete information on these options, see the upcoming “Compositing with a mask” section. Mixing images in different color modes Tip Throughout my laborious explanations of all those options in the Apply Image dialog box, I’ve been eagerly waiting to share with you the command’s one truly unique capability. Image ➪ Apply Image is the only way to composite images in dif- ferent color modes without setting the modes to match. For example, you could mix the lightness channel from a Lab image with each of the channels from an RGB image, or mix the green channel from an RGB image with each of the channels in a CMYK image. By contrast, if you were to simply drag and drop a Lab image into an RGB image, Photoshop would automatically convert the image to the RGB color space, which would result in a very different effect. To help make things a little more clear, Color Plate 13-6 shows four examples of an image composited onto itself using the Hard Light blend mode. The first example shows the result of selecting the RGB image as both source and target. As always, this exaggerates the colors in the image and enhances contrast but retains the same basic color composition as before. The other examples in the color plate show what happened when I duplicated the image by choosing Image ➪ Duplicate, converted the duplicate to the Lab mode (Image ➪ Mode ➪ Lab Color), and then composited the Lab and RGB images together. To do this, I switched to the RGB image, chose Image ➪ Apply Image, and selected the Lab image from the Source pop-up menu. In the top-right example, I chose Lightness from the Channel pop-up menu, which mixed the lightness channel with all three RGB channels. And in the bottom-left image, I chose b from the Channel pop-up menu and
644 Part IV ✦ Layers, Objects, and Text selected the Invert check box, which inverted the b channel before applying it. For the final, moderately psychedelic effect in Color Plate 13-6, I used the Lab image as the destination. I switched to the Lab image, chose the Apply Image command, selected the RGB image as the Source, and selected Blue from the Channel pop-up menu. Compositing with a mask The Mask option in the Apply Image dialog box provides a method for you to import only a selected portion of the source image into the target image. Select the Mask check box and choose the image that contains the mask from the pop-up menu on the immediate right. As with the Source pop-up menu, the Mask menu lists only those images that are open and happen to be the exact same size as the target image. If necessary, select the layer on which the mask appears from the Layer pop-up menu. Then select the specific mask channel from the final pop-up menu. This doesn’t have to be a mask channel; you can use any color channel as a mask. After you select all the necessary options, the mask works like so: Where the mask is white, the source image shows through and mixes in with the target image, just as if it were a selected portion of the floating image. Where the mask is black, the source image is absent. Gray values in the mask mix the source and target with pro- gressive emphasis on the target as the grays darken. If you prefer to swap the masked and unmasked areas of the source image, select the Invert check box at the bottom of the dialog box. Now, where the mask is black, you see the source image; where the mask is white, you don’t. The first example in Color Plate 13-7 shows a mask viewed as a rubylith overlay with the image. To make the mask, I selected the background with the Color Range command, inversed the selection, and saved the result as a separate channel. In the other examples in the color plate, I again composited the RGB and Lab versions of the image — as in the previous section — using Photoshop’s most outrageous blend modes. No matter how dramatically the Apply Image command affected the thinker, his background remained unscathed, thanks to the mask. If the Mask option had not been turned on, the background would have changed with the mode, turning light blue for Color Dodge, black for Color Burn, and deep red for Difference. Tip You can even use a selection outline or layer as a mask. If you select some portion of the source image before switching to the target image and choosing Image ➪ Apply Image, you can access the selection by choosing Selection from the Channel pop-up menu at the very bottom of the dialog box. Those pixels from the source image that fall inside the selection remain visible; those that do not are transparent. Use the Invert check box to inverse the selection outline. To use the boundaries of a layer selected from the Layer pop-up menu as a mask, choose the Transparency option from the Channel menu. Where the layer is opaque, the source image is opaque (assuming that the Opacity option is set to 100 percent, of course); where the layer is transparent, so too is the source image.
Chapter 13 ✦ The Wonders of Blend Modes 645 Add and Subtract The Add and Subtract blend modes found in the Apply Image dialog box (and also in the Calculations dialog box) work a bit like the Custom filter that I discuss in Chapter A on the CD-ROM that accompanies this book. However, instead of multi- plying brightness values by matrix numbers and calculating a sum, as the Custom filter does, these modes add and subtract the brightness values of pixels in differ- ent channels. The Add option adds the brightness value of each pixel in the source image to that of its corresponding pixel in the target image. The Subtract option takes the bright- ness value of each pixel in the target image and subtracts the brightness value of its corresponding pixel in the source image. When you select either Add or Subtract, the Apply Image dialog box offers two additional option boxes, Scale and Offset. Photoshop divides the sum or difference of the Add or Subtract mode, respectively, by the Scale value (from 1.000 to 2.000), and then adds the Offset value (from nega- tive to positive 255). If equations help, here’s the equation for the Add blend mode: Resulting brightness value = (Target + Source) ÷ Scale + Offset And here’s the equation for the Subtract mode: Resulting brightness value = (Target – Source) ÷ Scale + Offset If equations only confuse you, just remember this: The Add option results in a desti- nation image that is lighter than either source; the Subtract option results in a des- tination image that is darker than either source. If you want to darken the image further, raise the Scale value. To darken each pixel in the target image by a constant amount, which is useful when applying the Add option, enter a negative Offset value. If you want to lighten each pixel, as when applying the Subtract option, enter a positive Offset value. Applying the Add command The best way to demonstrate how these commands work is to offer an example. To create the effects shown in Figures 13-25 and 13-26, I began with the thinker and sunset shown way back in Figure 13-3. After switching to the Sunset image and choosing Image ➪ Apply Image, I selected the Thinker image from the Source pop-up menu. (I happened to be working with flat, grayscale images, so I didn’t have to worry about the Layer and Channel options.) I selected the Add option from the Blending pop-up menu and accepted the default Scale and Offset values of 1 and 0, respectively, to achieve the first example in Figure 13-25. The thinker went blindingly white, much lighter than he would under any other blend mode, even Color Dodge. To improve the quality and detail of the image,
646 Part IV ✦ Layers, Objects, and Text I changed the Scale value to 1.2 to slightly downplay the brightness values and entered an Offset value of –60 to darken the colors uniformly. The result of this opera- tion is the more satisfactory image shown in the second example of the figure. Add, Scale: 1, Offset: 0 Add, Scale: 1.2, Offset: -60 Figure 13-25: Two applications of the Add blend mode from the Apply Image command, each subject to different Scale and Offset values. Applying the Subtract command To create the first example in Figure 13-26, I selected the Subtract option from the Blending pop-up menu, once again accepting the default Scale and Offset values of 1 and 0, respectively. This time, the thinker turned pitch black because I subtracted the light values of his face from the light values in the sky, leaving no brightness value at all. Meanwhile, the thinker’s hair had virtually no effect on the sunset because the hair pixels were very dark and in some cases black. Subtracting black from a color is like subtracting 0 from a number — it leaves the value unchanged.
Chapter 13 ✦ The Wonders of Blend Modes 647 Subtract, Scale: 1, Offset: 0 Subtract, Scale: 1.2, Offset: 180 Figure 13-26: Two applications of the Subtract command on the images from Figure 13-25, one subject to Scale and Offset values of 1 and 0 (top) and the other subject to values of 1.2 and 180 (bottom). The result struck me as too dark, so I lightened it by raising the Scale and Offset val- ues. To create the second image in Figure 13-26, I upped the Scale value to 1.2, just as in the second Add example, which actually darkened the image slightly. Then I raised the Offset value to 180, thus adding 180 points of brightness value to each pixel. This second image is more likely to survive reproduction with all detail intact. The Calculations command Although its options are nearly identical, the Calculations command performs a slightly different function than Apply Image. Rather than compositing a source image on top of the current target image, Image ➪ Calculations combines two source chan- nels and puts the result in a target channel. You can use a single image for both
648 Part IV ✦ Layers, Objects, and Text sources, a source and the target, or all three (both sources and the target). The tar- get doesn’t have to be the foreground image (although Photoshop previews the effect in the foreground image window). And the target can even be a new image. But the biggest difference is that instead of affecting entire full-color images, the Calculations command affects individual color channels only. Only one channel changes as a result of this command. Choosing Image ➪ Calculations displays the dialog box shown in Figure 13-27. Rather than explaining this dialog box option by option — I’d just end up wasting 35 pages and repeating myself every other sentence — I attack the topic in a less structured but more expedient fashion. Figure 13-27: Use the Calculations command to mix two source channels and place them inside a new or an existing target channel. When you arrive inside the dialog box, you select your source images from the Source 1 and Source 2 pop-up menus. As with Apply Image, the images have to be exactly the same size. You can composite individual layers using the Layer menus.
Chapter 13 ✦ The Wonders of Blend Modes 649 Select the channels you want to mix together from the Channel options. In place of the full-color options — RGB, Lab, CMYK — each Channel menu offers a Gray option, which represents the grayscale composite of all channels in an image. The Blending pop-up menu offers the same 14 blend modes — including Add and Subtract — found in the Apply Image dialog box. However, it’s important to keep in mind how the Calculations dialog box organizes the source images when working with blend modes. The Source 1 image is equivalent to the source when using the Apply Image command (or the floating selection when compositing conventionally); the Source 2 image is equivalent to the target (or the underlying original). There- fore, choosing the Normal blend mode displays the Source 1 image. The Subtract command subtracts the Source 1 image from the Source 2 image. Half of the blend modes perform identically regardless of which of the two images is Source 1 and which is Source 2. The other half — including Normal, Overlay, Soft Light, and Hard Light — produce different results based on the image you assign to each spot. But as long as you keep in mind that Source 1 is the floater — hey, it’s at the top of the dialog box, right? — you should be okay. Tip The only mode that throws me off is Subtract, because I see Source 1 at the top of the dialog box and naturally assume that Photoshop subtracts Source 2, which is underneath it. Unfortunately, this is exactly opposite to the way it really works. If you find yourself similarly confused and set up the equation backwards, you can reverse it by selecting both Invert options. Source 2 minus Source 1 results in the same effect as an inverted Source 1 minus an inverted Source 2. After all, the equa- tion (255 – Source 1) – (255 – Source 2), which represents an inverted Source 1 minus an inverted Source 2, simplifies down to Source 2 – Source 1. If math isn’t your strong point, don’t worry. I was just showing my work. As you can in the Apply Image dialog box, you can specify a mask using the Mask options in the Calculations dialog box. The difference here is that the mask applies to the first source image and protects the second one. So where the mask is white, the two sources mix together normally. Where the mask is black, you see the sec- ond source image only. The Result option determines the target for the composited channels. If you select New Document from the Result pop-up menu, as in Figure 13-27, Photoshop creates a new grayscale image. Alternatively, you can stick the result of the composited channels in any channel inside any image that is the same size as the source images. Combining masks As described for the Apply Image command, the Channel pop-up menus may offer Selection, Transparency, and Layer Mask as options. But here they have more pur- pose. You can composite layer masks to form selection outlines, selection outlines to form masks, and all sorts of other pragmatic combinations.
650 Part IV ✦ Layers, Objects, and Text Figure 13-28 shows how the Calculations command sees selected areas. Whether you’re working with masks, selection outlines, transparency masks, or layer masks, the Calculations command sees the area as a grayscale image. So in Figure 13-28, the white areas are selected or opaque, and the black areas are deselected or transparent. Source 1 Source 2 Figure 13-28: Two selections expressed as grayscale images (a.k.a. masks). The left image is the first source, and the right image is the second. Assuming that I’ve chosen Image ➪ Calculations and selected the images using the Source 1 and Source 2 options, the only remaining step is to select the proper blend mode from the Blending pop-up menu. Screen, Multiply, and Difference are the best solutions. The top row in Figure 13-29 shows the common methods for combining selection outlines. In the first example, I added the two together using the Screen mode, just as in the preceding steps. In fact, screening masks and adding selection outlines are equivalents. To subtract the Source 1 selection from Source 2, I inverted the former (by selecting the Invert check box in the Source 1 area) and applied the Multiply blend mode. To find the intersection of the two masks, I simply applied Multiply without inverting. The Calculations command doesn’t stop at the standard three — add, subtract, and intersect. The bottom row of Figure 13-29 shows three methods of combining selec- tion outlines that are not possible using keyboard shortcuts. For example, if I invert the Source 1 mask and combine it with the Screen mode, I add the inverse of the elliptical selection and add it to the polygonal one. The Difference mode adds the portion of the elliptical selection that doesn’t intersect the polygonal one and sub- tracts the intersection. And inverting Source 1 and then applying Difference retains the intersection, subtracts the portion of the polygonal selection that is not inter- sected, and inverts the elliptical selection where it does not intersect. These may not be options you use every day, but they are extremely powerful if you can man- age to wrap your brain around them.