*** From the Archives ***

This article is from November 25, 2005, and is no longer current.

Photoshop How-To: Get Smart About Smart Objects

One of the most powerful features of Photoshop (and most other image editors) is not a specific tool, but the fact that your image remains malleable throughout the entire editing process. In a wet darkroom, once you’ve timed your processing in a particular way, you’re stuck with the results. With Photoshop, in addition to the Undo feature, the History palette, and the option to save multiple iterations of your documents, you’ve got Adjustment Layers, which let you perform certain edits in a non-destructive manner.
When you shoot Raw, though, there’s an extra wrinkle in your workflow. Raw files must first be processed with a Raw converter, such as Adobe’s own Camera Raw, before you can edit them further. Once you start editing a Raw file in Photoshop — adding adjustment layers, say, or creating duplicate layers with special blending modes — if you later decide your initial Raw conversion parameters weren’t quite right, you have to throw out your Photoshop work and start over in the Raw converter.
Unless you’re using Photoshop CS2.
With CS2, Adobe introduced Smart Objects, special layers that can contain references to vector or raster image data. You treat these layers almost like any other layer, but at any time you can alter the source data for the layer using Camera Raw. Those alterations get passed through to the Smart Object in Photoshop. Because Smart Objects can contain Raw files, they allow for a Raw workflow that lets you re-edit your original Camera Raw settings at any time.
Basic Raw Workflow
When you shoot a Raw file, your camera stores all the original unprocessed sensor data that it’s captured. When you process that Raw file with Camera Raw, the original Raw data isn’t changed. Instead, Camera Raw processes your data according to your specified conversion parameters and either writes out a new file or opens the results in a new document in Photoshop. The conversion parameters you specified are stored as metadata. You can go back at any time and edit those parameters and thus re-process your original file in a different way.
In a typical Raw Photoshop CS2 workflow, you select the images you want to process in Bridge, set their Raw conversion parameters to your liking, and then tell Bridge or Photoshop to process all the images accordingly. The result will be either a batch of new files or a bunch of new Photoshop documents. If you don’t like a conversion, you need to throw out the converted file, re-configure your parameters, and start over.
Fortunately, Bridge and Photoshop work very well together, so the process of editing Raw parameters and converting Raw files is very simple. However, there may be times when you should choose a slightly different workflow.
This Must Be the Place
Photoshop CS2’s Place command lets you place a Raw file into a Photoshop document as a Smart Object. To do so, create a new document of any size, then select File > Place and select a Raw file for placement.
Photoshop then presents the standard Camera Raw dialog, in which you set or tweak your Raw conversion parameters. When you’ve finished adjusting, press Open to continue the Place command. Camera Raw processes your image accordingly and places it in a new layer as a Smart Object. The Raw file is placed at the full width of your document and initially provides control handles so you can scale the document. Double-click or hit Return to accept the document as is. You now have a Raw file Smart Object.
If your document and the Raw file don’t have the same aspect ratio, the canvas may have some extra space in it. Use Photoshop’s Trim command (Image > Trim) to cut the canvas size down to match your image.
In the Layers palette, you should see a new layer with a Smart Object icon (Figure 1).


Figure 1. Note the unusual icon that indicates a Smart Object.

 
How Smart Is It?
A Smart Object is kind of like an Alias on the Macintosh or a Shortcut in Windows. It’s a reference to some original image data. In the case of Smart Objects, the original data is copied into your Photoshop document, but it’s kept there in its original form, and you can’t do anything to alter that original data as long as it remains a Smart Object.
Because all of the original Raw data is contained within the Smart Object, Photoshop can return to that data at any time and re-process it. For example, when you initially created your Photoshop document, you might not have created it at the full resolution of your placed Raw file. Photoshop automatically scaled the file to fit inside your document, but if your document settings are too low, you may not be getting the full resolution of your Raw file.
You can use Photoshop’s Image Size command to change your document to your Raw file’s original pixel dimensions, and Photoshop automatically re-processes and re-rasterizes the Raw image to fit those new dimensions.
The best thing about Raw Smart Objects, though, is that when you double-click on the Smart Object layer in the Layers palette, Photoshop calls up the Camera Raw dialog box. You can now change your settings and hit Open, and the image will re-process according to your new settings. For complex edits involving multiple layers, this facility for returning to your original Raw conversion parameters can be very powerful.
Consider the photo in Figure 2. I wanted to bring more attention to the three people in the foreground. Because the busy background is distracting, I decided to lower the saturation of the background scenery and edit the blonde woman’s hair.


Figure 2. The people in this photograph don’t stand out enough from the busy background.

 
Using the Place command, I placed two copies of this Raw file into a Photoshop document. For the first image, I set Camera Raw’s Saturation slider to -60. For the second placement, I used a normal Saturation setting of 0. I then added a Layer Mask to the upper Smart Object layer and painted my mask to knock out the background details of the upper layer, revealing the less saturated layer below (Figure 3).


Figure 3. Because the background is now less saturated than the three people walking, the people command more attention.

 
So far, I haven’t achieved anything that couldn’t be done with a normal image and a Hue/Saturation adjustment layer. However, I now have the full power of Camera Raw’s editing controls if I want to tweak my layers further.
For example, to add a little more warmth to the blonde woman’s hair, I can return to the Raw conversion parameters for the upper layer and use the Temperature control to define a new white balance for that image. In addition, I can use the Camera Raw Curve adjustment to improve the contrast in her hair (Figure 4).


Figure 4. I return to Camera Raw to add warmth and contrast to the woman’s long, blond hair.

 
While using Raw files placed as Smart Objects doesn’t necessarily let you do anything unique, it gives you an additional set of tools that might be easier for certain tasks.
Sharpening and Re-sharpening
Because Camera Raw 3 provides such a robust tool set, on many images you may find that you need to use Photoshop itself for little more than printing. However, printing usually requires a sharpening step, and Smart Objects provide a very flexible mechanism for sharpening.
Under the Detail tab, Camera Raw provides a sharpening slider that lets you apply a standard Unsharp Mask effect to your image. Because this is a Raw parameter like any other, once you’ve placed your image as a Smart Object, you can go back at any time and adjust the Sharpening setting.
Note that if you see “Sharpness (Preview Only)” rather than “Sharpness,” you must change your Camera Raw Preferences to Apply Sharpening to All Images.
What You Can’t Do
Unfortunately, a Smart Object doesn’t share all the behaviors of a regular layer. Because it contains a reference to a set of image data, rather than normal pixel data, you can’t use Photoshop’s painting tools, filters, or anything else that would alter the Smart Object’s pixel data. To perform touch-ups, such as dodging, burning, cloning, or painting, you first need to convert your Smart Object to a normal layer by choosing Layer > Smart Object > Convert to Layer. Once you take this step, you no longer have access to the original Raw parameters.
For complex multi-layered edits, you can work around a bit of this limitation by first duplicating your Smart Object before converting it to a normal Layer. Then, if you need to re-tune your Raw parameters, you can throw out your rendered layer and re-convert the original Smart Object. You’ll lose any painting edits you’ve made, but other layers and masks in your composition remain intact.
Another option is to perform as many of your painting operations as possible using non-destructive techniques. For example, if you need to Dodge or Burn, try adding a new layer above your Smart Object. Set the layer’s Blending Mode to Overlay. If you paint into the new layer with a color that’s darker than 50% gray, you’ll darken the underlying Smart Object. A tone that’s lighter than 50% gray will lighten the Smart Object.
Similarly, if you need to tint or paint into an image, perform these operations on a separate layer.
Streamline Your Smart Object Workflow
For quickly creating Smart Objects from Raw files, Adobe Bridge provides its own Place command, located within the File menu. This places the currently selected file into the front-most Photoshop document. However, you still need to create your document at the appropriate size, or re-size it later.
Adobe’s Russell Brown has written a set of scripts you can execute from within Bridge that let you select a Raw file and place it in Photoshop without having to first create a document or worry about document size. You can download the scripts and see QuickTime tutorials on how to use them at https://www.russellbrown.com/tips_tech.html.
Smart Objects are not a complete non-destructive editing solution, but they do provide handy non-destructive flexibility when you’re working with Raw files.

  • anonymous says:

    Excellent article that explained how useful Smart Objects can be. What’s interesting is doing a color space conversion (like to CMYK). You can keep the original data if you wish while still seeing the doc in that new color space. Cool, didn’t think about this. So you’re working in CMYK but you can go back to the RGB data in ACR and further tweak.

  • anonymous says:

    More of a question for Ben than a comment. I assume all of the steps with both Adobe Camera Raw and Photoshop are being done with 16 bit mode files. Correct? Is there any reason why you might want to work in 8 bit? Advantages? Disadvantages?
    Keith Dannemiller

  • >