Smart Objects

Did I save the best for last? I believe I did. Smart Objects offer almost all of the advantages of groups and very few of their limitations. Like a group, Smart Objects (SOs) are containers that can hold all kinds of Photoshop layers, including groups. But they can also contain Illustrator graphics, raw image data, and other SOs. And very unlike layers or groups, you can nondestructively perform edits like filters and transformations that are destructive when performed on any other type of layer.

Creating Smart Objects

There are several ways to create a Smart Object: by converting one or more existing layers, or inserting graphics from other documents (or even some other applications).

Converting

I wish this were called “encasing.” Like Han Solo in carbonite or Tony Stark in his Iron Man suit, the contents of Smart Objects are protected from abuse. When I need to perform a potentially destructive action like using a filter (e.g., Gaussian Blur) or applying a transformation (like Puppet Warp or Free Transform), I usually do so after converting the subject to a Smart Object.

Simply right-click near the name of a layer and choose Convert to Smart Object. If you’d like to convert more than one layer, highlight each before right-clicking near the name of one of them. If the layers are stacked contiguously, click on the top one’s name, then shift-click on the last one’s name to highlight them all. You can do that in the other direction too. If the layers aren’t contiguous, click the first, then ⌘/Ctrl click the others.

The result startles the first time: it looks like you’ve merged the layers. There is a small but significant reassurance in the lower-right corner of the Smart Object’s thumbnail. That icon lets you know that all those layers you had are safely inside their armor.

Any filter that’s applied becomes a “Smart Filter” and can be edited or removed later. This is a powerful aid to building intricate visual effects. If you should transform the S.O.—for example, scaling it—you can reverse that easily when needed. These approaches are discussed at length in “5 Filters & Transforms” (page 294).

Warning: If you attempt to retouch or paint on the S.O., a potentially confusing warning message will appear. Clicking OK to Rasterize will destroy the contents of the Smart Object! The right answer is to choose Cancel, then create an empty layer above the Smart Object onto which to paint or put repairs.

The engineers are not dictators. The attitude is “well, if you really want to do this, we’ll let you.” For too many users, however, the term “rasterize” is rather opaque. It used to have a simple meaning: convert vector artwork into a raster (pixel-based) image. The term has been expanded beyond vector art to include anything that isn’t simple pixels. So, if you clicked OK in that warning dialog, you would have a single simple-pixel layer.

Placing

Sometimes I might have an entire file that I’d like to use as part of a Photoshop project. When that’s the case, I choose to “place” it into the project’s document.

To do this, go to File > Place Embedded… or Place Linked…. (The difference between the two is discussed below.) When you’ve chosen a file to place, it will appear on the canvas with transform handles so you can adjust its size immediately. If you look in the Options Bar, you’ll see how much it’s been scaled. You can choose to disallow any initial scaling in the General Preferences.

Placing other files allows us to perform Photoshop actions on Illustrator artwork, for example, while still retaining the ability to edit it in Illustrator. Even other Photoshop files can be used. For the sake of the space-time continuum, I’ve never tried to place a document into itself.

One of the more powerful options is placing raw files. There are several ways of doing this and many advantages. The methods and reasons are discussed both later in this Compendium (see “7 Editing Raw Files” on page 350) and in the Course (in the “Adobe Camera Raw” chapter, see “Processing Images with ACR” on page 98).

Pasting

When you paste content into a Photoshop file, it will become an embedded Smart Object. As is the case when placing, you are given transform handles to size it.

Linking vs Embedding

Whether linking or embedding has the advantage depends on the circumstance. For example, if you need to use an infographic in several Photoshop projects, especially if that graphic may be edited later, then linking is the clear winner. When that infographic is edited in Illustrator (perhaps because the information it illustrates has changed), you should note a warning in the Smart Object thumbnail when next you open any of the Photoshop documents that used it.

To update a Smart Object with modified content, right-click near its name in the Layers panel (not on the thumbnail), then choose Update Modified Content. Better, especially if there are several out-of-date Smart Objects, choose Update All Modified Content. In fact, I see no harm in choosing the latter every time.

When you need to provide your Photoshop project to someone else or move it to another computer, you will also need to supply any linked assets. This may sound like a time consuming task, especially if you have many linked SOs, but there is a command to do this for you! The process is called “packaging,” and the command is found at File > Package. This creates a folder whose name is based on the Photoshop file’s, and in which you’ll find a copy of that file and a folder called “Links” with copies of all the linked assets. InDesign and Illustrator both have this feature too.

An embedded Smart Object can be converted into a linked one. Right-click near its name and choose Convert to Linked…. You’ll then be asked to choose where the new linked file should be. As with all linked Smart Objects, if the linked file is moved, the Smart Object’s content can not be edited. Attempting to do so results in a dialog box requiring you to locate the file.

Workflows & Limitations

Lightroom and Raw Workflow

For those who use Adobe Lightroom, I recommend doing as much as possible with your raw images in Lightroom before turning to Photoshop. That’s because all edits in Lightroom are nondestructive and can easily be revisited. Once an image is passed to Photoshop, a new TIFF file is generated that will be Photoshop’s responsibility.

So when I need to edit in Photoshop a raw image that I manage in Lightroom, I right-click it in Lightroom and choose Edit In > Open as Smart Object in Photoshop…. When finished editing in Photoshop, simply Save, and a TIFF file is created with “-Edit” appended to the name, located in the same folder as the original, and managed by Lightroom. Subsequent edits to that TIFF should then be done in Photoshop, however. To do so, right-click its Lightroom thumbnail and choose Edit In > Adobe Photoshop. In the dialog that appears, choose Edit Original, meaning the TIFF, rather than generating yet another copy. This approach allows us to maximize the features of each of these two programs with the fewest number of files.

You may also set Adobe Camera Raw to supply Smart Objects to Photoshop (see “ACR Workflow” on page 352). When you save the file that opens in Photoshop, you may choose any location for it.

In either case (Lightroom or ACR), the document you edit in Photoshop has a Smart Object whose content is raw data. Double-clicking its thumbnail will open Adobe Camera Raw to allow edits. You’ll see all the edits you performed in ACR or Lightroom before pushing that data to Photoshop, which you can now alter.

Note: When editing the Smart Object’s content, you aren’t editing the original raw file, but rather a copy embedded in Photoshop. So the raw edits you make won’t work their way back to the original. This is fine because other edits you’ve done to the Photoshop file, the S.O.’s context, are likely the motivation to edit the S.O..

Adobe Illustrator to Photoshop Workflow

You can File > Place Linked or Embedded or Edit > Paste Illustrator artwork into a Photoshop file. As discussed earlier, placing can either link or embed the original Illustrator document in its entirety. Pasting always embeds, but does allow you to be more selective about the exact content of the Smart Object that’s created. In other words, you can copy just a small piece of a larger illustration to be pasted into Photoshop. Once embedded, the S.O. can be converted to a linked one.

When you double-click the Smart Object thumbnail, Adobe Illustrator will launch so you can edit its content. As always, simply save your edits, and the S.O. updates.

General Photoshop Use

In general Photoshop work, I convert one or more layers into Smart Objects if I need to apply a transformation or filter, since those are otherwise destructive processes. There are, however, a few transformations and filters that can’t yet be applied to SOs. With each version of Photoshop, the list of those shrinks. If the filter or transform you need is grayed out when a Smart Object is highlighted, you’ve found one! In those rare cases, you will need to apply that function to an ordinary layer.

Likely the most general way to do so is to first apply allowed filters and transforms to the S.O., if you need any. Then duplicate the Smart Object (⌘-J/Ctrl-J) and rasterize the copy (right-click near its name and choose Rasterize Layer). I also usually hide the S.O. by clicking its eye icon. I won’t delete it because I may need it again, especially if I regret the actions I perform on its rasterized copy.

Frame Layers

No, I didn’t forget to include this new kind of layer. I include it here because it’s extremely limited and use of the Frame tool creates Smart Objects. If a layer other than the Background layer is present, a frame drawn with the Frame tool will automatically contain (or “nest” in Photoshop parlance) the topmost layer, which is converted to a Smart Object. If you wish to frame a layer other than the top one, you will have to either lock all the layers above it (covered later in this chapter) or move the desired layer to the top. If you want to frame a Background, click its padlock to convert it to a pixel layer. If there was an empty frame when unlocking a Background, that frame will now contain the former Background.

As of this writing, I find the Frame tool frustrating and unwieldy in all but the very simplest (e.g., single layer) documents. Also, it is limited to rectangles and ellipses, although shapes drawn with the shape tools can be converted to frames. In that case, I’d prefer to use those shapes as vector masks, which are far more flexible and only slightly more difficult to manipulate. Read “Vector Masks” (page 291) to learn more about them.