Most photographers I know try to follow some sort of “non-destructive” workflow. You probably aren’t “destroying” your image, so it might be more appropriately called a “flexible” workflow. The ultimate purpose is typically to give you the ability to easily redo an earlier step in the edit, without having to redo all the work that came after it. Camera RAW Smart Objects give you the ability to change the earliest parts of an edit, the RAW processing. Another great use is as a “rough” workflow to allow you to work quickly to initially evaluate a file, and then really dial it in later. Both purposes have merit, but you need to use Camera RAW Smart Objects the right way to get the most out of them.
When I teach Photoshop, I routinely run into 3 common misconceptions that photographers have about Camera RAW Smart Objects. Understanding these three points can help you improve the quality of your finished work.
Misconception #1: Smart Objects do not provide better quality than exporting a TIF from Lightroom.
An exported TIF, Camera RAW Smart Object, or rasterized Smart Object are all going to give you the same results (assuming you used the same RAW settings to create each of them). As far as the image is concerned, the Smart Object is exactly the same as a flat layer. The only difference is that you can make changes to recreate that flat layer. But the algorithm used in Lightroom or in Adobe Camera RAW (inside the smart object) is exactly the same and produces the same results.
So if don’t need the flexibility to make changes later, you’re probably better off skipping the Smart Objects to save a bit of file space.
Misconception #2: Not all copies are created equal
“New layer via copy” (<cmd/ctrl>-J) or “Duplicate” creates a new layer which use the SAME processing settings as the original Smart Object. If you change the contents of either (by double-clicking into one of them), you change both of them. This approach is helpful if you want to apply different filters or different filter masks to the same image. For example, you might apply Nik Color Efex to extract more detail from buildings, and then Nik Dfine to reduce noise in the sky above.
“New Smart Object via Copy” (right-click the layer) creates an independent layer which uses DIFFERENT processing settings. This approach is helpful when you want to use different RAW processing on different parts of the image. For example, you might use stronger noise reduction on on version of the image, and then use a layer mask to selectively apply that noise reduction to shadow areas of the image.
The Smart Object icon on the layers is the same either way, so be careful. The only way to tell them apart once they are created is to change one and see if any others change as well.
Misconception #3: The Camera RAW Filter does not use your RAW data!
Photoshop essentially rasterizes your Smart Object before applying any Smart Filter. Which means that the outside of your Smart Object is just like a flat TIF file. The RAW data is processed before any filter can be applied to it. In other words, using the Camera RAW Filter outside the Smart Object prevents the algorithm from working on the best possible data. The difference of these two approaches can be pretty significant for recovery of extreme details and color.
So you should never apply the Camera RAW filter to your Camera RAW Smart Object. Instead, you should double-click the layer’s thumbnail to open the Camera RAW interface. This will let you make all your RAW edits on the actual RAW data. These two approaches look nearly the same, and a good reminder is to always take a quick look at the white balance temperature slider. If it is in Kelvin (a scale from 2-50,000), you are working on the RAW data. If you get a +/-100 scale, you are not getting the benefit of the RAW data.
What’s the point of the Camera RAW filter then? It’s still a great way to apply basic processing to anything other than RAW data. For example, you could use it to apply clarity to a final layered image.