Why Photoshop Smart Objects Are So Useful

There are a ton of ways to approach editing an image, with different methods having their own pros and cons. One method involves using Smart Objects, and this helpful video tutorial will guide through editing landscape images with them and show you why they are so useful. 

Coming to you from Michael Shainblum, this great video tutorial will show you how to use Photoshop's Smart Objects to edit landscape images. If you have not seen them before, a Smart Object is a special sort of layer that uses data from a raster or vector image without discarding the original information. This allows you to do things like apply resizing and geometric transformations nondestructively or use things like filters while retaining the ability to undo or go back and change their settings whenever you would like, which is particularly useful if you are applying a lot of overlapping adjustments that work in tandem to create the final image. On the other hand, you can't do anything that affects pixel-level data on a Smart Object (such as using the Clone Stamp); you would have to rasterize the layer first. Check out the video above to see them in action while Shainblum edits. 

Alex Cooke's picture

Alex Cooke is a Cleveland-based portrait, events, and landscape photographer. He holds an M.S. in Applied Mathematics and a doctorate in Music Composition. He is also an avid equestrian.

Log in or register to post comments
2 Comments

IMO it's kinda silly that anything has to be rasterized outside of the final export. If you watch any advanced compositing tutorial in Photoshop, you see people duplicating layers and masks everywhere, and at the end the entire image is flattened into a new layer and sent into some color grading plugin like the Adobe Raw filter or AlienSkin or whatever. If the client tells you they want some adjustment that has to be done early in the pipeline, you better remember everything that got copy/pasted and reflatten the image and send it through that plugin again if you want to incorporate those changes. This workflow sucks and reeks of poor software design.

The compositing software that photographers use is not particularly well suited for non-destructive workflows. You should be able to do composites by linking the input files and then constructing a computation graph, so you can reuse or reference the same mask/layer in many places and only have to change one occurrence of the mask to change it everywhere that it's used. Everything should be parametric and editable at any time, propagating your changes immediately into the final output. That's how most video compositing software like Fusion, Nuke, Natron, etc. do it.

I agree completely.

Although a node based compositing workflow is not available in Photoshop (PS) right now I work with smart objects in this manner like in the video for a long time. With one difference:
➡️ I take Lightroom (LR) out of the equation altogether. I just dont need it for that workflow.
(although it is ok for processing hundreds of images with small adjustments).

The RAW engine is exactly the same as in LR. Everything one does in LR is the same in PS Adobe Camera Raw (ACR).

I even retouch the basics of skin in the PS ACR because if the client asks for a warmer tone I just go in the smart object back to the raw file adjust it and boom - all the retouchings are warmer as well.

Of course at some point I have to work pixel based and that looked is baked in. But normally these are the last steps and easy to adjust.

But even though the file size increases and the workflow might get slower I try to stay non-destructive as long as I can.

From the final (almost non-destructively) edited image - lets say in 8256 x 5504 resolution in ProPhotoRGB or AdobeRGB - I make a copy in 3500 Pixels in sRGB that is linked also via smart object.

So when I make a change in the "big" one it will be updates in the smaller one.