Understanding and Choosing File Formats in Photoshop CS and Illustrator CS

This chapter is from the book 

Photoshop's File Formats

Each file format that can be generated from Photoshop has its specific capabilities and intended purpose. Most offer one or more options as you save the file. Typically, a supplemental dialog box opens after you have selected a filename, format, and location in Save As.

Format Capabilities: The Overview

Different file formats have various capabilities. Table 3.1 shows a summary of the features of the most important file formats, along with those of some less-common formats for comparison.

The primary formats are discussed individually in the sections that follow.

¹ - In Photoshop, PSB is also referred to as Large Document Format.
² - Clipping masks are stored but not used by most other programs. Adobe InDesign does read the clipping path.
³ - Type and shape layers are rasterized if the file is reopened in Photoshop.
⁴ - Transparent background can be maintained in GIF.
⁵ - Files are saved as Photoshop Raw, but Photoshop CS is capable of opening Camera Raw files from most digital cameras.
⁶ - TIFF advanced mode preserves layers.

Photoshop CS ships with optional plug-ins for a variety of file formats that are not needed by the vast majority of Photoshop users (see Figure 3.11). Should you need to work with one or more of these formats, drag the plug-in into the File Formats folder of Photoshop's Plug-ins folder. The next time you start Photoshop, the file format will be available.

Photoshop (.psd)

Photoshop's native file format supports all the program's capabilities. Files in this format can be placed in the latest versions of InDesign and GoLive as smart objects and can be opened in Adobe Illustrator. This is the default file format. Most workflows (but not all) benefit from maintaining a file in Photoshop format until it's time to create a final TIFF, EPS, JPEG, or other file. It's also usually a good idea to maintain the original image, with editable type and layers, for future use.

Figure 3.11 Look for these plug-ins in the Optional Plug-ins folder, inside the Goodies folder, on the Photoshop CD.

When saving in Photoshop format, you can see that the Save As dialog box indicates which features are used in the image (see Figure 3.12). If a check box is grayed out, that feature is unused. If you uncheck a box, the file is saved as a copy. If you disable a feature (layers, for example) and save the image as a copy, when the image is reopened, that feature is gone—no more layers.

Figure 3.12 Selecting the As a Copy check box without changing other options simply appends the word copy to the original filename.

Large Document Format (.psb)

New in Photoshop CS is the capability of saving truly huge images. In earlier versions of Photoshop, the maximum image size was 30,000 pixels by 30,000 pixels. The new PSB file format can handle images as large as 300,000x300,000 pixels. Other than the maximum image size, the PSB file format is functionally similar to the PSD (Photoshop standard) file format.

Photoshop will remind you that the PSB file format is compatible only with Photoshop CS (see Figure 3.13). Other programs and earlier versions of Photoshop will not recognize the format.

Figure 3.13 Any file can be saved in the PSB format, but it won't be readable in any program other than Photoshop CS. That's another good reason to have this reminder.

CompuServe GIF

GIF is a common Web file format, suitable for illustrations and other images with large areas of solid color and no or few gradients or blends. Many logos and cartoons, as well as Web navigation items, such as banners and buttons, are appropriate for GIF. This file format is not appropriate for most photographs and other continuous-tone images because it can record a maximum of only 256 distinct colors. (A photograph might have thousands or millions of individual colors.) Although very small and adequately detailed thumbnails of photographs can be created as GIFs, a continuous-tone image typically suffers from posterization. When areas of similar (but not identical) color are forced into a single tint, the image quality can suffer severely.

The color table (the list of specific colors included in the image) is controlled through the Indexed Color dialog box. This box opens after you select the CompuServe GIF file format, a name, and a location for the file and then click OK.

For full information on the color options available for GIF files, see "Indexed Color," in Chapter 5, "Color, in Theory and in Practice."

Among GIF's capabilities are interlacing and animation. When a GIF image is interlaced, a Web browser displays every other line of pixels, first, and then fills in the balance of the image. This allows the Web page's visitor to see the image loading, giving the impression that progress is being made and that the image is loading faster. Animated GIFs are created in ImageReady.

Animated GIFs are discussed in Part IV, "ImageReady." See Chapter 23, "Creating Rollovers and Adding Animation."

The GIF format supports transparency in a limited way: A specific pixel can be opaque or transparent. You cannot have any pixels that are partially transparent in a GIF image. If the image is on a transparent background, you need only select the Transparency check box in the Indexed Color dialog box (see Figure 3.14).

You can also specify a color to be made transparent by using the Matte options in the Indexed Color dialog box. All pixels of the specified color then become transparent, regardless of location in the image. Dithering allows an image in Indexed Color mode to simulate more colors than are actually contained in the color table by interspersing dots of two or more colors.

GIFs can also be created with Save for Web. See Chapter 22, "Save for Web and Image Optimization."

The GIF standard is maintained by CompuServe, and software that creates or displays GIF images must be licensed by Unisys. (The licensing fee was paid by Adobe, and part of Photoshop's purchase price goes to the licensing.)

Figure 3.14 The Indexed Color dialog box opens automatically when you use Save As to create a GIF.

Photoshop EPS

PostScript is a page description language developed by Adobe, and it was at the heart of the desktop publishing revolution of the 1990s. An Encapsulated PostScript (EPS) file can contain any combination of text, graphics, and images and is designed to be included (encapsulated) in a PostScript document. EPS files contain the description of a single page or an element of a page.

EPS is typically used for elements to be included in a page layout or PDF document. Because PDF files can be designed for onscreen display as well as print, EPS supports the RGB color mode in addition to the CMYK and Grayscale modes. EPS files cannot be displayed by Web browsers (although they can be incorporated into PDF files, which can be shown through a browser plug-in).

One of the greatest advantages of EPS as a file format is the capability of including both raster and vector data and artwork.

To learn more about the difference between raster and vector images, see Chapter 4, "Pixels, Vectors, and Resolutions."

EPS supports all color modes, except Multichannel, and it does not support layers or alpha and spot channels. (The DCS variation of EPS supports both spot colors and masks and is discussed separately, later in this chapter.) The EPS Options dialog box, shown in Figure 3.15, offers several options.

The Preview option determines which image—if any—will be shown in the page layout program where you're placing the file. The preview doesn't normally have any effect on the printed image and is used for onscreen display only. Windows offers only None and TIFF as preview options. Here are the options offered by other platforms:

• None—No image preview is saved. Although Photoshop's File Browser can still generate a thumbnail, files placed into a page layout document may show on the screen as a box with diagonal lines.

• TIFF (1 bit/pixel)—A black-and-white (not grayscale) version of the image is shown in a page layout document.

• TIFF (8 bits/pixel)—A color version of the image is displayed onscreen in page layout programs. This option provides a reasonably accurate preview and is cross-platform.

• Macintosh (1 bit/pixel)—A black-and-white PICT preview is included with the EPS file. The PICT file format is designed for use on Macintosh systems but can be used for Windows programs as well.

• Macintosh (8 bits/pixel)—A color PICT preview is prepared.

• Macintosh (JPEG)—A JPEG preview is used, which can reduce the file size but also the compatibility.

Figure 3.15 Note in the Save As dialog box that the warning triangles indicate which features are not supported by the EPS file format.

The Encoding options include ASCII, Binary, and JPEG. ASCII is cross-platform and compatible with virtually all page layout programs. Binary is a Macintosh encoding system, and it can produce smaller files. JPEG encoding produces comparatively tiny files but uses lossy compression (which can result in image degradation) and is not compatible with all page layout programs. In addition, JPEG encoding can prevent an image from properly separating to individual color plates. ASCII is certainly the safest choice when the image will be sent to a printer or service bureau.

Halftone screens and transfer functions are established through the Print Preview dialog box. Screen frequency, angle, and shape can be established for each ink. These criteria determine the distribution and appearance of the ink droplets applied to the paper. The transfer function compensates for a miscalibrated imagesetter.

PostScript Color Management is available only for CMYK images outputting to devices with PostScript Level 3. This option should not be selected if the image will be placed into another document that itself will be color managed.

When the image contains unrasterized type and/or vector artwork or paths, select the Include Vector Data check box. When you're outputting to a PostScript device (such as a laser printer or imagesetter), the vector outlines will be preserved and the artwork will print more precisely. You need not include vector data when outputting to an inkjet printer.

Checking Image Interpolation allows a low-resolution image to be upsampled (resampled to larger pixel dimensions) when printed to a PostScript Level 2 or 3 printer. Increasing the resolution during printing helps smooth edges and improve the general appearance of low-resolution images, but this could introduce a slight out-of-focus appearance. If the output device isn't PostScript Level 2 or higher, this option has no effect.

For more information on upsampling, see "Two Types of Resampling," in Chapter 15, "Image Cropping, Resizing, and Sharpening."

JPEG

Joint Photographic Experts Group (JPEG) is technically a file compression algorithm rather than a file format. The actual file format is JFIF (JPEG File Interchange Format), although JPEG is more commonly used. JPEG supports Grayscale, RGB, and CMYK color modes and can be used with files more than four times as large as Photoshop's 30,000x30,000 pixel maximum. JPEG does not support transparency, alpha channels, spot colors, and layers. Paths can be saved with a JPEG file, including clipping paths (although most programs can't use the clipping path, with the notable exception of InDesign). Type is rasterized when a file is saved as JPEG.

JPEG is commonly used on the Web for photographs and other continuous-tone images in which one color blends seamlessly into another. Because of its support for 24-bit color (in RGB mode), JPEG is far better than GIF for displaying the subtle shifts in color that occur in nature. (Rather than GIF's maximum of 256 distinct colors, JPEG files can include more than 16.7 million colors.) JPEG is the most common file format for digital cameras because of the outstanding file size reduction capabilities.

When you choose JPEG as the file format and click OK, Photoshop's Save As dialog box is replaced with the JPEG Options dialog box (see Figure 3.16). You can select a matte color, a level of compression, and one of three options for Web browser display.

JPEG files can also be created with the Save for Web option. See Chapter 22, "Save for Web and Image Optimization."

Figure 3.16 The bottom of the dialog box shows the estimate of the amount of time it will take, under optimal conditions, for the file to download at a given modem speed at the selected quality setting.

Simulating Transparency in JPEG Files

Although JPEG does not support transparency, you can choose a matte color that will help simulate transparency in your image. From the Matte pop-up menu in the JPEG Options dialog box, choose a color that matches the background of the Web page into which the image will be placed. By replacing transparent areas of the image with the selected color (rather than just flattening to white), you cause those areas to match the Web page's background (see Figure 3.17).

Figure 3.17 The Photoshop image is shown at the top. To the left, it has been saved as a JPEG with a red matte and placed into a GoLive document with a red background. On the right is the image as it appears in a Web browser—a JPEG duck surrounded by "transparency."

Optimization Options

The cornerstone of the JPEG format is image optimization. The JPEG Options dialog box presents you with the tools to strike a compromise between file size and image quality. The Preview check box enables you to monitor changes to the image's appearance. The Image Options slider ranges from 0 (very small file, possibly substantial quality loss) to 12 (large file, very little—if any—damage due to file compression). The Quality pop-up menu offers Low (3), Medium (5), High (8), and Maximum (10) settings. Figure 3.18 shows a comparison of file sizes and file quality, using the duck image from the previous figure. The chart shows file format, compression type or level, and file size as reported by Mac OS X's Get Info command. (Different images might produce substantially

Figure 3.18 From left to right, the three images show JPEG quality settings of 12, 5, and 0. Note the substantial image quality degradation at the lowest quality setting.

Format Options

The Format Options section offers three JPEG format versions from which to choose:

• Baseline ("Standard")—This is the most widely compatible version of JPEG, which can be displayed by all graphics-capable Web browsers. It's also the version to use for non-Web JPEG files.

• Baseline Optimized—Optimized JPEG can produce better color fidelity but is not compatible with all older Web browsers.

• Progressive—Like interlaced GIF files, progressive JPEG files appear in stages in a Web browser, providing visual feedback and the illusion of speed to the Web page's visitor. You can elect to have the image appear fully in three, four, or five passes. (Three passes is usually adequate.) Progressive JPEG files are typically (but not always) slightly larger than baseline (standard) versions of the same files.

At the bottom of the JPEG Options dialog box is an area that shows the size of the file that will be created with the selected options as well as how long such a file will take to download. The pop-up menu enables you to select modem speeds ranging from 14.4Kbps to 2MBps.

JPEG/JFIF became an ISO standard in 1990. Baseline JPEG is in the public domain, but many variations are patented and licensed. Variations of lossless and near-lossless JPEG are under development.

Resaving JPEG Images

Because JPEG is a lossy compression scheme (some data is thrown away during compression), it's generally understood that files should be saved in the format only once (if at all). Resaving a JPEG file in JPEG format sends the image through the compression process again, possibly resulting in damage to the image's appearance because of the second round of data loss.

With the booming popularity of digital cameras, more and more images are being captured as JPEGs. Completely eliminating JPEG as a file format for resaving these images (and JPEGs created for the Web) after editing is unfeasible. To best understand how to avoid damaging an image when resaving it as a JPEG, you should know a little about how JPEG compresses.

Lossy compression, such as JPEG, can produce far smaller file sizes than lossless compression. That's because some image information is actually thrown away during the compression process. When the image is reopened, the existing data is averaged to re-create the missing pixels. However, the cost is image quality (see Figure 3.19).

To the left is the original. In the middle, saving at the highest quality produces little degradation. On the right, a JPEG Quality setting of 1 greatly reduces the quality of the image. Zooming in to 800% shows one key aspect of how JPEG works. In the low-quality version of the image, JPEG's distinctive 8x8 blocks of pixels are visible (see Figure 3.20). Colors can be averaged or patterns determined based on the content of the blocks.

Because JPEG uses blocks of pixels to compress an image, cropping becomes important. If you must crop an image that has already been substantially compressed with JPEG, any cropping from the top and left should be in increments of 8 pixels. Two copies were made of the image saved at JPEG Quality 1 (see Figure 3.21). One copy (left) was cropped exactly 8 pixels from the top and 8 pixels from the left. The second copy (right) had 4 pixels cropped from the top and left. Both copies were then resaved at JPEG Quality 1.

Figure 3.19 As you can see, the image on the right (saved at Quality 1) has a much degraded appearance compared to the original (left) and the image saved at Quality 12 (middle). However, the lower-quality image has a file size about one-third that of the higher-quality image.

Figure 3.20 An image is broken down into these 8x8 pixel blocks, starting from the top-left corner.

Perhaps the most precise way to crop an image in increments of 8 pixels is to use the Canvas Size dialog box. It not only allows you to specify an exact pixel size, but the Relative check box even eliminates the need to do math—select the check box and enter a negative number to reduce the canvas size (see Figure 3.22).

Another technique for preserving image quality when resaving a JPEG file as a JPEG file is to use exactly the same compression setting as originally used. If the original quality/compression setting is known, and if the image is being recompressed in the same program, then this technique is valid. However, keep in mind that different programs use different "scales" of JPEG compression. Saving a JPEG file at Quality 4 in Photoshop is not the same as saving as Quality 4 in Illustrator, nor is it the same as using 40% in Save for Web. Also, there is no set formula to translate between compression settings from a digital camera and Photoshop.

Figure 3.21 The image that was cropped in increments of 8 pixels (left) bears far more resemblance to the original (top). On the right, the image that was cropped in increments of 4 pixels looks much blurrier.

Figure 3.22 The grid, called the proxy, determines from where the canvas will be reduced or enlarged. If you're cropping from the center, remember to work in increments of 16 pixels (8 pixels for each edge).

When minimizing additional image degradation is important, consider resaving JPEG files at High or Maximum Quality, especially if the image has been cropped. The file size might be somewhat larger than the original JPEG, but quality will be maintained.

If you use a digital camera, a little experimentation can help you determine which quality setting you should use. In the example shown in Figure 3.23, the image in the upper right was taken at SQ (Standard Quality), which captures 640x480 pixels for this camera. To the left is the same scene, shot at HQ (High Quality), which produces an image 1,600x1,200 pixels, compressed at approximately the equivalent of using JPEG Quality 7 in Photoshop. The image in the lower right was shot at SHQ (Super High Quality), 1,600x1,200 pixels, minimal compression.

The two larger JPEG images are cropped to about 469KB when open in Photoshop. The original images as captured were 480KB (left) and 956KB (lower right). With the minimal difference in image quality and the substantial difference in file size between HQ and SHQ, for most purposes the more practical setting for this camera is HQ. Your camera may differ.

Figure 3.23 The two lower images are zoomed to 300%. The upper-right image, because of the smaller pixel dimensions, is zoomed to 700%.

Photoshop at Work: Controlling JPEG Compression

JPEG is an important file format when you consider the number of such images on the Internet and those taken with digital cameras. The lossy compression algorithm that JPEG uses is worth exploring. To practice controlling JPEG compression, follow these steps:

1. Open an image, any image, in Photoshop.

2. Use the menu command File, Save As and then select JPEG as the file format.

3. Use an image quality value of 0 and save the file to your desktop.

4. Close the open file and use the menu command File, Open to open the file you just saved.

5. Click the leftmost button at the bottom of the History palette to create a copy of the open image.

6. Save this copy, appending "-1" to the name. Again, use JPEG quality 0. Close and reopen the –1 file.

7. Position the two windows next to each other and zoom in 500% on an area of detail. Compare the two. Although neither looks good, they should be nearly identical.

8. Use the menu command Image, Canvas Size. Click in the lower-right corner of the proxy (the 3x3 grid labeled Anchor). Switch the unit of measure to pixels and then input a height and width, each 4 pixels smaller than the original value. Click OK.

9. Save the file as JPEG again, using the lowest quality setting.

10. Close and reopen the image.

11. Zoom in 500% and compare this version to the original.

You should see that saving the same image at the same setting doesn't cause much harm, but using a low compression setting after cropping an image that has already been compressed with JPEG can cause severe quality reduction.

JPEG 2000 (.jpf)

JPEG 2000, an improved variation of the JPEG file format, is now supported in Photoshop. Although this format is used for both print and the Web, most Web browsers require a plug-in to display JPEG 2000 images. You'll need to install the JPEG 2000 plug-in from the Photoshop CS CD to use this format.

Photoshop opens files in the extended JPEG 2000 format (JPF), but it can also create JPF files that are compatible with the JP2 format for JPEG 2000. (The file extension remains .jpf, regardless of whether the file is JP2 compliant or not.) The JPF format is more flexible than JP2, but fewer Web browser plug-ins are available. JP2 files will be 1KB larger than their JPF counterparts.

Among the advantages of JPEG 2000 over the earlier version of JPEG are transparency and lossless compression. You also have some control over what metadata is included with the file and how the image will appear in a browser as it downloads.

The JPF format supports RGB, CMYK, and Grayscale color modes, in either 8-bit or 16-bit color. Alpha channels, spot color channels, and paths are retained. Layers are merged when saved as JPF.

To create a JPF file, select the file format in the Save As dialog box, name the file, and choose a location. If desired, check the boxes to include the document's ICC profile (mandatory if JP2 compatibility is required) and spot or alpha channels. Then click Save. The JPEG 2000 window will open (see Figure 3.24).

Figure 3.24 More than just a dialog box, the JPEG 2000 window is reminiscent of a less-complex Save for Web window. You can resize the window by dragging the lower-right corner.

The JPEG 2000 window consists primarily of a large preview area to the left and settings options to the right. Here are the principle features of the JPEG 2000 window:

• Standard Zoom and Hand tools are available in the upper-left corner.

• The lower-left corner offers a zoom pop-up menu, with 10 preset zoom factors, ranging from 3.125% to 1600%.

• The File Size field enables you to select a target size for the file. This option is not available if lossless compression is used.

• The Lossless option prevents degradation of the image caused by lossy compression. Selecting Lossless will produce larger file sizes.

• Fast Mode improves the speed of the JPF file encoding, without harming the image. The File Size option is grayed out when Fast Mode is selected, although the actual file size can still be changed using the Quality slider or entering a value in the field.

• The Quality slider is used to determine the amount of compression applied to the file. It is not available when the Lossless box is checked because it controls only the lossy compression scheme. Like Save for Web, the Quality slider's values range from 1 to 100.

• You can elect to include metadata with the file. Depending on the choices made in the Advanced Options, you can include the image's metadata as XML (JPEG 2000 specific), XMP (Photoshop's File Info format), and/or EXIF data (from a digital camera).

• The image's ICC profile will be included, provided all three of the following boxes are checked:

• Embed Color Profile (in the Save As dialog box)

• Include Color Settings (in the JPEG 2000 window)

• ICC Profile (in the Advanced Options dialog box of the JPEG 2000 window)

• The Include Transparency option is only available when the image is on a transparent background. Layers will be merged upon saving, but the image will not be flattened.

• JP2 compatibility increases the file size by 1KB, but a wider variety of plug-ins will be able to open the file.

• The options accessed by clicking the Advanced Options button provide you with a choice of what types of devices you want the file to be compatible with (currently there is only one choice), the type of lossy compression "wavelet" you want to use, Integer (more consistent throughout the image) or Float (a bit sharper), the Tile Size setting (generally 1,024x1,024 is appropriate), the metadata format, and the type of ICC profile(s) to include (standard or the restricted profiles intended for cell phones and wireless devices). The Float compression scheme, available only for lossy compression, is useful for images with lots of fine detail, but it can create artifacts along edges. The tile size, which determines the areas of each "block" used for compression, is generally best left large. However, images for wireless Web access may benefit from a smaller tile size. Also, although the JPEG 2000 standard implies that a "restricted" ICC profile should be embedded in every JPF file, the devices for which the restricted profiles are intended (cell phones and PDAs) are not generally color managed...yet.

• As shown in Figure 3.24, you can specify how you want the image to display as it downloads using the Order pop-up menu. The options are Growing Thumbnail (a small image appears, then larger and larger versions of the image appear until the image is fully loaded), Progressive (the image starts fuzzy and then sharpens as it downloads), and Color (a grayscale version appears, which changes to color when the image is finished downloading).

NOTE

Keep in mind that if your image is not intended for the Web, the only significant options are in the JPF Settings section of the window.

• If the file contains one or more alpha channels, a channel can be selected as the "region of interest." The alpha channel determines what areas of the image require the highest quality. Because the Region of Interest option doesn't increase file size, all improvements within the region of interest result in decreased quality outside the region.

• The Enhance slider determines how much priority is given to the region of influence. The higher the level, the greater the quality within the area designated by the alpha channel (and the lower the quality outside).

• The Download Rate pop-up menu enables you to choose an Internet connection speed that will be "previewed" in the window. You can see how long it will take your image (nominally) to download over a connection of that speed in the field to the left, below the preview area.

• The Preview button shows you how your selected optimization order looks. Click the button, and the image will appear in the preview area as it would load into a Web browser.

Photoshop PDF

Adobe's Portable Document Format (PDF) is a cross-platform format that can be opened and viewed in the free Acrobat Reader, available for most computer operating systems. PDF is, at heart, a PostScript file format. Photoshop breaks PDF into two categories: Photoshop PDF and Generic PDF. Both can be opened, but only the former can be created.

Photoshop PDF supports all of Photoshop's color modes, transparency, vector type and artwork, spot and alpha channels, and compression (JPEG or ZIP, except for bitmap images, which use CCITT-4 compression).

The PDF Options dialog box, shown in Figure 3.25, offers a choice of compression schemes as well as several other choices.

ZIP compression is most effective for images with large areas of solid color, and it is also more reliable than JPEG for PDFs that are destined for process printing. (PDFs compressed with JPEG might not separate correctly.) JPEG, however, can produce substantially smaller files (at the cost of image quality).

Figure 3.25 After you select a filename, location, and Photoshop PDF as the file format, clicking OK opens this dialog box.

Photoshop also enables you to use standard PDF security, such as that added to PDFs in Adobe Acrobat (see Figure 3.26). Passwords can be required for opening the document as well as for printing, selecting, changing, copying, or extracting the document's content.

Figure 3.26 Other check boxes in the PDF Security dialog box are made available if they pertain to the file being saved.

PNG

Developed as an alternative to GIF and JPEG for the Web, Portable Network Graphics (PNG) comes in both 8-bit (Indexed Color) and 24-bit (RGB) variations. JPEG's lossy compression and the licensing requirement associated with GIF's compression lead to the demand for an alternative. PNG-8 and PNG-24 are both now widely supported by Web browsers.

The Save As dialog box makes no distinction between PNG-8 and PNG-24, instead using the file's existing color mode. If the image is RGB, PNG-24 is automatically created. If the image is Indexed Color, PNG-8 is used. Grayscale is also 8-bit. (Save for Web allows you to specify whether you want to create an 8-bit or 24-bit PNG file.) Because PNG does not support CMYK, it is not appropriate for commercial printing applications.

Like interlaced GIF and progressive JPEG, PNG files can also be displayed incrementally in a Web browser. After you select the PNG format, name, and location and then click OK, the PNG Options dialog box enables you to make the selection (see Figure 3.27).

Figure 3.27 Interlacing can add slightly to the file's size. It's not necessary with small interface items but is often appropriate for larger images.

Images in Indexed Color mode are comparable in size and appearance when saved as GIF and as PNG-8. PNG-24 cannot match the file size reduction available in JPEG; however, the format does offer transparency (which is not available at all in JPEG) and uses a lossless compression algorithm, thus preserving image quality.

Raw

In Photoshop, it's necessary to differentiate between Raw files saved from the program (Photoshop Raw) and digital camera files being opened in the program (Camera Raw). Many high-end cameras use proprietary versions of the Raw format. Previous releases of Photoshop could not natively open such files. The various camera manufacturers had their own software to open and process the files, from which you could save a TIFF file to open in Photoshop. In early 2003, Adobe released a Photoshop 7 plug-in called Camera Raw for use with the top cameras from Canon, Nikon, Minolta, Olympus, and FujiFilm. That plug-in has been incorporated into Photoshop CS and has some additional capabilities.

Photoshop Raw

The Photoshop Raw file format records pixel color and very little else. Each pixel is described in binary format by color. Because the file doesn't record such basic information as file dimensions and color mode, coordination and communication are important. If incorrect data is entered into the Raw Options dialog box when an image is opened, an unrecognizable mess is likely to result (see Figure 3.28).

When saving an image in the Raw file format in Macintosh, you can specify the four-character file type, the four-character creator code, how many bytes of information appear in a header before the image data begins (if any), and in what order to save the color information (see Figure 3.29). When interleaved, color information is recorded sequentially for each pixel—the first pixel's red, green, and blue values are followed by those three values for the second pixel, and so on. The sequence is recorded as R,G,B, R,G,B, R,G,B.... Noninterleaved order records the red values for all the pixels, then the green values for the pixels, and then the blue values. The sequence is then R,R,R... G,G,G... B,B,B....

Figure 3.28 In the upper left, the image was opened properly. To the right, the Interleaved option wasn't selected. Below, the dimensions were incorrect. If the number of channels or the bit depth is wrong, you'll see a warning that the file size doesn't match.

Figure 3.29 Before saving a file in the Raw format, make sure you know and understand the requirements for the program in which the file will be opened.

Camera RAW

When you select and open an image from a high-end digital camera, Photoshop will, if necessary, launch the Camera Raw plug-in. (Some Raw files can be opened directly in Photoshop.) The Camera Raw plug-in offers very powerful image-adjustment capabilities. Global color and tonal adjustments, sharpening, noise and moiré reduction, and compensation for color fringing caused by chromatic aberration are some of the features of Camera Raw.

TIFF (.tif)

Tagged Image File Format (TIFF) and EPS are the two most widely accepted image formats for commercial printing. TIFF files can be produced directly by most desktop scanners and many digital cameras. The format supports CMYK, RGB, Lab, Indexed Color, Grayscale, and Bitmap color modes. In Bitmap mode, alpha channels are not supported, but they are available in all other color modes. Spot channels are supported, and clipping paths can also be saved with TIFF images to denote areas of transparency.

Photoshop offers a variety of TIFF options, some of which are not supported in other programs. Background transparency, layers, and JPEG compression are three. Additional TIFF options are found in the TIFF Options dialog box, which appears following the Save As dialog box (see Figure 3.30).

Figure 3.30 Some options in this dialog box might be grayed out and unavailable, depending on the image's content.

For maximum compatibility, choose LZW as the compression scheme. ZIP and JPEG are not supported by many programs that use TIFF files. JPEG offers the same 0–12 scale of quality available in the JPEG Options dialog box.

The Byte Order option determines compatibility with Macintosh or IBM PC computer systems. Because the Mac has no trouble reading the IBM version of TIFF files, that is the safest choice.

Although Photoshop itself doesn't read a TIFF file's image pyramid, the multiresolution data can be saved for other programs, such as InDesign. Image pyramid refers to multiple versions of the same image being stored in one file, each at a different resolution. Photoshop reads only the highest- resolution version of a TIFF file.

If the file contains background transparency, you can elect to save that transparency. As with other advanced TIFF features, the transparency option is not widely supported for TIFF files; however, it is supported by InDesign 2.0. (Remember that this transparency refers to the entire file, not layers above a background layer.)

When a file is saved with layers, you have the option of determining how those layers will be compressed. Because layers can greatly increase file size, ZIP may produce substantially smaller files. You can also choose at this point to disable layers in the TIFF file and save as a copy. In terms of the resulting file, there is no difference between discarding layers (flattening the image) with this option and unchecking the Layers check box in the Save As dialog box.

Because not all other programs can read a TIFF file's layers, Photoshop saves a flattened version of the image as well. Saving layers in TIFF file format triggers a warning message from Photoshop (see Figure 3.31). Because a flattened version of the image is also saved in the TIFF file, the warning can be ignored.

Figure 3.31 If your image is to be used in a program that doesn't support layered TIFF files, discarding the layers can greatly reduce file size with no loss of functionality.

Photoshop DCS

Desktop Color Separations (DCS) is a version of EPS developed by Quark. (The file that's produced has the .eps extension.) The original DCS file format is now referred to as DCS 1.0, whereas an updated, more flexible version is called DCS 2.0.

Photoshop DCS 1.0

Version 1.0 should be used only with older programs that don't read the DCS 2.0 standard. DCS 1.0 supports CMYK and Multichannel modes and creates a separate file for each color channel, plus an optional 8-bit color or grayscale composite. The DCS 1.0 Format dialog box appears after you click OK in the Save As dialog box (see Figure 3.32).

Figure 3.32 These options determine compatibility with page layout programs.

You can elect to save no preview, a 1- or 8-bit TIFF preview, and in Macintosh, a 1- or 8-bit Macintosh (PICT) preview or a JPEG preview. The 8-bit TIFF preview is the most widely supported. In addition to the preview, which is used primarily in dialog boxes, you have the option of including a grayscale or color composite image or no composite at all. The composite is used to show the image when a page layout document is viewed onscreen.

The encoding options include Binary, ASCII, and JPEG. As with EPS files, ASCII is used in Windows and can be read by Macintosh programs, Binary produces smaller files but is Mac specific, and JPEG reduces both file size and compatibility. In addition, JPEG encoding can prevent an image from properly separating to individual color plates. ASCII is certainly the safest choice when the image will be sent to a printer or service bureau.

The Print Preview dialog box offers access to halftone screens and transfer functions for the file. These criteria determine the distribution and appearance of the ink droplets applied to the paper. The transfer function compensates for a miscalibrated imagesetter. Do not make changes to the screens or include a transfer function without explicit guidance from your printer or service bureau.

Photoshop DCS 2.0

DCS 2.0 is a more sophisticated version of Desktop Color Separations. The majority of its options are the same as those described for DCS 1.0. However, DCS 2.0 gives you several options when it comes to what file or files are actually produced (see Figure 3.33).

Figure 3.33 The six Preview options boil down to one file or one file for each channel and a choice of composite image.

If you choose any of the three Multiple File options (no composite, grayscale composite, color composite), one file is generated for each color channel. The four process channels (if present in the image) use the channel's letter for a file extension. Spot channel extensions are numbers, starting with 5. The composite, if generated, will have the .eps extension.

For example, the file "Target" with two spot channels could be saved as DCS 2.0, with multiple files and a color composite. The files generated would be Target.C, Target.M, Target.Y, Target.K, Target.5, Target.6, and Target.eps.

DCS 2.0 should be selected over DCS 1.0 in any circumstance where both are supported. The single file saves disk space and can simplify file exchange and handling.