Adobe Creative Suite 4 Printing Guide
64
Path. The selected path is then used as a vector silhouette for the layer, much like a clipping path.
As with vector text and Shape Layers, images with Vector Masks should be saved as Photoshop
PDFs to retain a crisp, vector edge when placed in other applications, such as InDesign or
Illustrator.
Smart Objects:
Vector art placed as a Smart Object in a Photoshop image receives special
handling. While it displays as pixels, the source vector art is embedded within the Photoshop
file as vector. If a Smart Object is scaled in Photoshop, the content is re-rendered from the
embedded vector information. Thus, no fidelity is lost from interpolation during cumulative
transformations. Raster art can also be designated as a Smart Object, with the same benefits. But
vector Smart Objects differ in an important way from the vector components discussed earlier:
Smart Objects display and image as pixels, regardless of how the file is saved. That is, saving as a
Photoshop PDF does not result in sharp vector edges; Smart Objects are rendered at the resolu-
tion of the image.
Smart Objects do not retain any link to the original file; any edits performed to the original file
won’t be reflected in the Smart Object embedded in the Photoshop image. The Smart Object data
is part of the Photoshop file with no external links.
To edit a Smart Object, double-click its thumbnail in the Layers palette. If the Smart Object is
vector, Illustrator CS4 will launch, and the vector art will open as an Illustrator file. Edit as
necessary, then choose File > Save; the edited data will be written back into the Photoshop
file, and the Smart Object will be updated accordingly. If the Smart Object is a pixel image,
double-clicking will open the image data in Photoshop for editing and will write the corrected
image data back into the parent image when saved. Note that Camera Raw images can be placed
as Smart Objects, and double-clicking to edit will call up the Camera Raw interface. Vector
Smart Objects can be endlessly transformed without losing data (because each transform is
re-rendered from the embedded vector data), but raster Smart Objects are subject to some of the
same restrictions as any raster image:
•
Rotating the Smart Object will cause some loss of detail, although subsequent rota-
tions are each re-rendered from the original pixel data; thus, the transformations are
not cumulative.
•
Scaling down does not destroy data, although the reduced image will display less
detail because of reduced size; scale the Smart Object back up to its original size, and
the original data is re-rendered without additional interpolation.
•
Scaling up a raster Smart Object past its original size requires interpolation, with a
resulting loss of detail. However, Smart Objects offer the advantage of re-rendering
the image data fresh with each transformation, rather than causing cumulative data
loss with multiple transformations.
Clipping paths
One common method for silhouetting an object in Photoshop is to draw a vector path with the
Pen tool. Traditionally, it was necessary to designate the path as an official clipping path (by
choosing Clipping Path from the Paths palette menu), and then save the file as a Photoshop
EPS. This is still viable, although it’s no longer necessary to designate a path as a clipping path;
InDesign provides options for using any saved path within a Photoshop PSD or TIFF file. Saved
“regular” paths offer more flexibility than clipping paths; the user can choose from multiple
saved paths for multiple uses of a single image—with different appearance—by changing the use
of clipping paths within a page layout. Such paths can also be edited in InDesign; the original
path is unchanged, but a user-modified instance of the path is used by InDesign to customize the
silhouette. Such flexibility is limited for clipping paths; InDesign can modify the path to trim
out visible parts of the image, but cannot reveal anything that falls outside the clipping path. A
regular Photoshop path (not designated as a clipping path) doesn’t have such a limitation; it can
be freely edited, to hide or reveal any part of the image.
Flatness, expressed in device resolution pixels, governs how a device interprets curved vector
paths, using tiny, straight segments. The lower the flatness value, the more straight segments
are used to draw the curved path, and thus render it more faithfully. The higher the number of
straight segments that must be generated when imaging the curve, the more processing power
is required. But it’s not necessary (or advisable) for you to make a decision about appropriate
flatness settings for clipping paths. When designating a path as a clipping path, leave the field
