Smart Objects: Quick Start
So here I have an image, just any image I've opened, there's nothing special about this particular image, and if you take this image, I'm gonna duplicate it, so I just go to the image menu, and I should be able to find in here, duplicate. And when I duplicate it, I'm gonna call this one not smart. So any time you see that as the file name, you'll see it at the top of the little tab above the file, you'll know if I'm working on a smart object, or if I'm not, 'cause this one I'm gonna leave alone. Then I can take these images, and they usually show up as tabs like this, and I could pull those tabs down to make them floating windows, or if I go up to the window menu, there's a choice here called arrange, I can say I wanna see two up, meaning two images side by side vertically, therefore I can see the two files side by side. Now, let's make the image on the right, that's the one that does not say not smart, let's click on that one, and we're gonna turn it into a smart object. I can do that...
in a few different ways, but the main method is to go to the layer menu, to choose smart objects, and say convert into smart object. Now when I do this, this picture's not gonna look any different. But, the place where we'll notice a change is in the Layers panel. In the Layers panel, look at the little thumbnail image, the miniature version of the picture that's sitting there, and look in the lower right corner of it. When I choose convert to smart object, do you see the little icon that appeared there? That's your only real indication that this layer is different from others. Anytime you see that icon in the lower right of a thumbnail, it means you're working on a smart object. Whenever you have a smart object, it's like taking whatever you had selected in your Layers panel, at the moment you created it, and isolating it into a protective bubble. You can think of it like a ZipLoc bag, you're just putting stuff in there and sealing it shut. And now, anything that I do to that layer cannot permanently affect the contents of what's inside. Instead, whatever I do to that layer will just be applied as settings attached to the layer. Settings that could be undone, so that if I scale the image, for instance, there's just a number attached to that layer that says display at 45 percent scale, and if I scaled it back up, all it would do was change that setting that's attached to the layer. These are settings we don't usually see, but it's how it's working behind the scenes. Whatever I do to this layer cannot affect what's inside the smart object. So let's take a look at how that changes the way we work. I'm gonna take the image on the left, that's the one that's not a smart object, and I'm just gonna do some things to it. For instance, I'm gonna go here to image size, and I'm gonna scale it down to a tiny size. Let's make it so it's four pixels wide. Click okay. Then let's zoom up on that, I'll type command zero, which means fit in window, that's all there is in the file. If I save that file and close it, that's all that's saved in the file. Nothing more. It does not remember where that came from, what it was made from. If I take the image on the right, and I do the same thing: I go to the image menu, I choose image size, and I also make it four pixels wide, it might look slightly different when I click okay, only because this was not a layer called background. When you have a layer called background, it's thought of going all the way to the edges of your document, and it can't have a soft edge on it. When I do it to a smart object, it's a normal layer, and if you scale a normal layer and do something weird like we've just done, you'll get a little bit of transparency showing up, but the image on the left would have looked the same, had I changed the name of the background layer to anything else, so it wasn't a background. Anyway, you see that they look similar, and if I save and close both images, and then reopen them, it wouldn't be any different from what I'm about to do. I'm gonna take the image on the left, and I'm gonna try to scale it back somewhere near normal size. Now I don't remember the original size, so I'm just gonna make up a number. All right, I'm gonna say I want this 1500 pixels wide. Now when I do that, all it can do with the image on the left, since it's not a smart object, is it can take what you currently see there, and scale it up, and there's no way it's going to invent fine details, 'cause they're simply not there. So the end result looks like this. The image on the right, though, we turned into a smart object before we scaled it. Whenever you turn something into a smart object, whatever you have selected in your Layers panel gets put into a protective bubble. And nothing can get into that bubble and change the original. When I scaled down the size of the document, all it did was use some text to say when I display this picture, scale it down to this size, but inside that smart object is still the full size original layers. So if I take the image on the right, and I choose image size, and I type in 1500, don't know if that's the original size or not, I click okay, do you see it looks a little bit better? And that's because it's acting on the contents that's inside the smart object, it's not actually changing what's in the smart object, but it's looking back to what is the actual contents there, and that's always what it's scaling. Now I'm going to undo this a few times, just get back to the original picture, make sure that one's a smart object, so we don't have the one on the left lookin' on, now let's try something else. Let's try scaling and rotating this time. Now to scale and rotate, when you have a background layer, that layer's locked, so usually you can't scale or rotate, but if you just clock, click the lock symbol, you'll unlock it. And now, I'll type command T, command T is simply a shortcut for image, or sorry, edit, and then free transform. So command T, and what I'll do here is I'll scale it down, and I'll rotate it. Like that. And maybe I type command T again. Now do you notice when I type command T a second time, it forgot about the original rotation, because it would have remembered it, that little rectangle would stay on the same angle as my other version, but it treats this second transformation I'm doing as if it never knew I ever transformed it before. I'll scale it down some more, and I'll rotate it. But if I type command T again, to decide I want it bigger again, and I don't want it rotated, command T, and I'm like, how can I get this back straight? Well I have to eyeball it, because there's no place here that's keeping track of how much it was rotated, it's as if it thought I scanned it at that angle, as if it never knew it was straight. And then I could try to scale it back up, but if I do, it's always using the end result of my last step, and therefore by the time I get this large, it's gonna start looking really soft, because it doesn't have enough information in it to make it still have quality. When I hit return or enter and I zoom up, we might be zoomed down enough, but you can see it's degraded, looks all kinda weird and jaggy. Now let's try the same thing with the smart object. If I type command T to transform, I bring this smaller and I rotate. I press return to say I'm done. Maybe I use my move tool, move it up. Okay, now, I'm gonna tell you command T a second time. Remember on the other one, it never remembered it was rotated, it put a brand new rectangle around it, but this time when I type command T, it remembers it was rotated, it's as if I never left the original transformation, and not only that, up here it remembers the angle it had been rotated, where as on this image, if I type command T a second time, those would always be zero, it wouldn't have any idea it was ever rotated. If I wanna get it back to normal, I just take this number and type in zero for the angle, and then here there's a percentage for the width and height, I just come in here and type 100 for both of 'em, and we're back to the exact same original size, and it's always calculated from the exact contents that's inside that smart object, it always goes back and calculates it based on the full size image that's in there. So a smart object kind of encapsulates whatever it is you have in your Layers panel, puts it in a protective bubble, and then we can do stuff to it that's not permanent, and it gives us many different advantages. So if I come in here and open another file, I'll see if I can find an old one, I'm not actually sure, well we'll work on a bunch of files in a day, and eventually, I'll find one that was how should I say it? That was made decades ago that I can update with new files, meaning that that image, when I scaled it and rotated it, I could just as easily say replace the contents that's here, with a brand new file and it would scale and rotate it and remember.