Lessons

Lesson Info

Workflow Suggestions

I want to go through a couple of images to show you if I was working on an image like this how I might start it from beginning to end. In a kind of condensed, cause I'm not gonna go through every, you know detail work. It's kind of a broad brush. So if they're involved, for example, selections, I'm not going spend the twenty minutes I normally would take in the perfect selection, I'm gonna kind of speed that up, we'll just kind of see the process behind something. So let's make sure this is, so last year, I was in Iceland which was fantastic, except every day we were there the sky looked gray. And we were in a van of like five of us, I think. Five or six, and we had this rule that the tour guide was taking us to the next location but if anyone saw something interesting, we just say stop, and we all get out of the van. And on this day five people went stop there's a pony with a rainbow, come on! (laughter) Like how can you not stop and get out for that. And it was funny because our van ...

of five people and then every car behind us was like, and all these people were stopping, and these wild horses were just in fields, and they're so friendly they just come ambling over like hey how's it going, and we're like wait stay there, there's a rainbow, don't move. So, we didn't' have a lot of time, so it's not like, wait let me get the perfect settings here. It was like quick take a picture while you can. So this is what I got. So there's a few things about it that I'm not, that I want to edit, improve upon, etc. So, it's a raw file, so I start in here. And I would look globally first, and say is there anything I want to do globally to this. And one of the things I usually do first is check and see there's a function here called lens correction, which means it will attempt to determine from the data in this image what lens you used and if it needs to do any correction for things like distortion, stuff like that. Now, I'll tell you ahead of time, sometimes when I do that I don't like the results, even though it's technically correct, it doesn't actually help. What it does sometimes is kind of hard, you can barely see it on this if you look at the very top right hand corner above that little blurry thing, see there's a little dark area here, sometimes lenses will do that, they just add that kind of effect, so turning this on enable profile, it says this is the lens you use, so this is how I would adjust it for you right away. So here's the original, here's now. See it took a little bit of distortion, it took a little bit of that out. So I would always try that first to make sure, sometimes you'll notice no difference at all. Then I'd look it over now. As soon as I look at this photo, my first thought is ultimately I'm gonna have to do something to this guy separately because the grass and the horse I'm okay with, the sky doesn't look as nice as I'd like it to be so I'm gonna do that independently, which means selectively, which to me rings the bell do that in Photoshop later. Even though I could try to do it here, so instead I'm just going to look at things like the opening of the shadow that I want to get a bit more detail on the horse. I'm not worried about this stuff, I can get that Photoshop later, as well as this thing up here. Might try overall, so the difference between vibrance and saturation, saturation makes every color more saturated. Vibrance, if there's people in it, tries to not make people look bright orange and things like that. So this case it's not going to make as much of a difference. But if I go too far, the rainbow looks better but now the horse looks too much. So I'm just going to go a little bit, maybe something like that. And then open the object. So, did my initial part in raw because it's easier, but it's still smart, so I could always go back and change it. Yes. What are your thoughts on just using the auto tabs, does it come close? Well, I have mixed feeling about that because I would say the auto tabs have got better, but my concern is if I click auto and I don't like it, now what do I do? So I guess there's nothing wrong with clicking auto to see if you like it or not, but I would say more often then not, I'm like eh, I wouldn't have done it that way. So, it's like anything, where some people use a white balance tool because they say if I balance it correctly, well sometimes the white balance, while it's technically correct, visually you might not like it. So I would certainly say there's nothing wrong with hitting auto and seeing what it does, I would say a lot of the time I end up doing my own just cause I prefer doing that. But it's certainly a personal choice. Now, let's, line of horizon's bugging me a little bit. So let's see if we can't straighten this. So I'll take the crop, you'll notice this one I'm doing in Photoshop. You can straighten in camera raw, but it's a little harder to do what I want to do, which is not lose too much information. I'll show you what I mean as I do this. So I drag across what I want to be straight. And says well I'll have to crop it that way, but I feel like I'm losing some information. I mean that's what it does, it just crops it to the existing pixels. But I would like to have a little more of this way, I'm thinking a little more this way. And down here just a bit more. So, you're seeing just little bits of transparency when I crop it, it means the image won't fill up the whole thing, but that's okay, because I know that I have technology that can help me with that. There we go, hit enter. So now it's straighter, but now I've got missing pieces. This is where I have to kind of think through the process of Photoshop. Cause most people or many people's first thought is ooh, content aware fill. Good idea, except for one problem, content aware fill does not work on a smart object. So because I want to preserve the ability to go back to camera raw, that's a smart object, I can't content aware fill. So I have to come up with an alternative which to me would be the patch tool. Cause we saw before I can add a blank layer and take the patch tool, and do the equivalent of content aware fill just takes a couple more steps. So I select this area, I'll say, fill that. And this part. That part. A bit there. Here. Now you see, what's happening, that wasn't a good. Moved to the wrong spot there. Is often it's doing a pretty good job, but you'll see how a couple times I'm having to over, do it again. And that's just a combination of me missing a spot, and it's just not knowing exactly where the best place to fill from. I'm gonna try and do as much as I can with these broad strokes, and notice I haven't yet zoomed in, because all I'm trying to do is overall make a little better. There's just one piece there that just doesn't want to fill, no matter what I do. So it's like okay, try something else. And up here, I see a little bit. Now of course, all of this is going on this blank layer. Which means we have to think about, even though I've chosen to do the patch tool on it's own layer, because I prefer that, it does mean now, I've in a sense eliminating the possibility of going back to camera raw to edit it, because if I do, now those patches won't match up. So that's kind of the game you have to play a little bit. But that's okay, in this case I'm fine with that for my purposes. So let's go a little bit bigger and see if we can't figure out why this little guy is deciding to be so uncooperative. It's just saying nope I don't want there to be grass there, I don't care what you say. Lets try the spot healing brush instead. Ha, take that. Okay. So I did my initial patching fit in a window view because it's quicker, but now I'll go in, as I do I can see that looks like a bit of a repeating pattern to me that was generated by my patches and if' I'm not sure, I can turn that on and say yeah, that's pretty much what's happening there, so I go back in and do a little bit of quick stuff here, just to eliminate. Because what you're trying to do is fool the eye into not seeing, wait that same little twig is in four different places, because that is kind of what gives it away. So if this was real project, I would be going though this whole thing and saying okay, that part there, I don't like. But I'm not going to do all of that but you get the idea, you kind of move around. So it's kind of the work flow for me is camera raw bring it in while still in fit and window view do as much as I can in that smaller view cause it's quicker, then get closer in and see if there's any obvious areas that I need to improve upon. Alright, so I want to also edit the sky independently. Which means I need to make a selection, because it needs to be selected right? So quick selection tool. In this case I also want to turn on this little check box that say sample all layers. Because if I only click on this layer, I'd miss out on all this patchy stuff I did. So now I can sample all layers and I'll look across the whole thing, and despite quick drag, it's done a pretty good job of selecting everything. Now, add adjustment layer, probably vibrant, so I can do more saturation in here. So that's okay, but I want some more. So here's a bit of a question, I want to add a second adjustment layer that uses the same mass that I just had. So how do I figure that out. Well there's a couple of possibilities. One of them is, anytime you can see the mask, you can always put it back to a selection, by command or control clicking on it again. So now I've made another selection, so I can go back, and say what if I add a photo filter adjustment layer and change it to cooling, so it's a little more bluish, so now I've got two adjustment layers, both with the same mask. So that's certainly a possibility. Those certainly work. Now one thing I'm gonna go back, cause I want to show you another option that I meant to show you earlier with smart filters. You can see that there's choice here, and that is, I keep saying, don't use these adjustment settings, because they're destructive. There is one way to use those and make them non-destructive. So, this is okay. I want to take a deep cleansing breath before I say this. So I have a camera raw smart object to which I've added a layer. I want to edit those two as if they're one, but I don't want to merge them, because I don't want to lose either bit, so I take the smart object and the other layer and convert that to a smart object. So it is possible to take a smart object, and if necessary, put it inside a different smart object for the purpose of editing. Which sounds crazy, but it happens. So, part of the reason I wanted to do this, is to show you that now this is a smart object for the whole thing, if I do something like levels, it still brings up this levels dialogue box, but remember before when I clicked okay, I was done. Now when I click okay, it comes down here and says levels. It's actually the rough equivalent of an adjustment layer. And the reason this might be better, is, actually let me not use levels, show you, same thing I just did. So let's use vibrance, push the saturation. Sorry, I still need to make a selection, that's what I was missing. So I still need to make a selection, now, that's what I meant to do. If I do vibrance push the saturation, it's just saturating that part, and now it's says vibrance, but a minute ago I had to make a whole nother layer with a layer mask. If I want to do a second one now, and so something like photo filter, before we said one of the problems with smart filters, they only have one mask. In this case, it's actually a benefit. Cause now I have one mask and I can apply these filters, which are actually different adjustments. So instead of having five adjustment layers, each with he same mask over and over. Now I have one mask with different adjustments. So that's a different strategy that is kind of interesting. The other thing that we can do, and this is a really interesting, now remember this is kind of interesting, this started life in camera raw. I made global adjustments, but now that I look at it, I want to take my quick selection tool again, and I'd like to apply de-haze only to the sky. Well there's no such thing as de-haze as an option in Photoshop, well not directly. But there kind of is. Cause if I go to the filter menu, one of the filters called the camera raw filter. So that means that I can apply the filter, now this part that's a little weird at first, it looks like it's applying it to everything, cause it is, kinda. But remember I have a mask on there, so I can take de-haze and go like this, let's be overly dramatic, click okay. And now it only does it to the selected area because of the mask. So, camera raw as a filter is interesting because in camera raw itself, to try and de-haze only the sky, would be frankly somewhat of a nightmare, trying to use the adjustment brush in camera raw. But in Photoshop, you saw how easy it was to take the quick selection tool, make a selection so now I'm using the camera raw filter inside Photoshop, because it's easier, gives me more control. Okay? So, at a certain point, I would of course save this as a PSD file, continue, so I'll just pretend that this for now is what I want to do. Save this. And I would do other things, obviously to it. But that kind of gives a rough idea of an idea of this particular work flow. Where even in something as relatively simple as this, there's still times where the smart object filter's gonna get me more benefit. So now if I were to open this weeks from now, I could edit all the pieces, and I'd still have that camera raw smart filter, I can go back, double click on this and go back to the individual pieces that I used to create it, etc, etc. The part that throws people off a little bit is before it looked like it was cropped, now it doesn't. Because in the contents window it's showing me everything, so it has to kind of make that window and make some extra transparency, but ultimately once I close that, it goes back to this kind of look again. So here's the other project that I have. I have taken these photos and as I look at them, you see they're fairly similar, but I like her pose in the first one, but I cut off her foot. So in the second one, there's her foot, but I don't like the pose. So I want to see if I can't use them both in some way. So they're both raw files. So I want to make sure that I edit them equally, because eventually when I combine them together, I want them to look the same. Now these were taken moments apart, so the camera settings should be virtually identical. So if I select both of them, one of the things we haven't taught before is you can edit more than one image at the same time in camera raw which is interesting. Just go up here in this tiny little menu and choose select all, so that means that whatever I do to the one I'm seeing, it will also apply to the other one. So I'm just gonna make couple of minor tweaks in here it's already been adjusted a little bit. Like that it looks actually pretty good. But instead of opening them, I'm actually just wanting to make sure they were edited the same. I'm just actually gonna click done, so whenever you do that just means, now they've been edited, it's just kind of preserving that information. If you look here in bridge, right here at the top of each of them there's a little symbol in the top corner which is an indication that you have made some editing to this in your raw file. But what I want to do, is I want to get them into Photoshop. So instead of opening both of them, now I could do it this way, I could open each one, have two separate files, drag one on top of the other to combine them together, which certainly would work. When you know already you want to combine two files together the reason I like using bridge, they're both selected, I go to the tools menu, Photoshop, and then load files into Photoshop layers. So that means it will say combine both of these images into one document automatically. So I don't have to do all that work. What it does do though is it removes the camera raw smart object part. Which is okay in this case for what I want to do. So now, I've got the two poses that I want, but what I really want, this one on top is easier to see. So the way that I typically do this, let's zoom a little closer, lowering the transparency or opacity, here's my one, this is the foot that I actually want. So I'm doing all this for a foot, believe it or not, but that' just sometimes the way it is. So I'm gonna lower the opacity of this one, so now I take my move tool, I can try and line things up. So you can tell, she wasn't really moving too much, except for a slight difference in pose. So it's actually not that hard to reposition to kind of match up. So lowering the opacity will be quite helpful to see when you're trying to, just kind of look for a reference point like the leg of a chair, something that's not moving, to match them up. Then I put the opacity back up again. Now something we haven't talked about before but comes up in case like this, so I took that top layer and I dragged it to the left so that I could get the rest of her foot, but you see it's still not there, because the canvas stopped there. Remember we talked about canvas size, that doesn't change, you could move a layer four inches over to the left but it's still going to be cut off because our canvas isn't big enough. And rather than me guess, well how big do I need to make this canvas, I use a command under the image menu called reveal all. And that will make the canvas big enough to reveal the missing part okay? Can we put that music back on again? (chuckling) so now, I have the part that I want, but I don't remember that's it's, I want this pose, but I want this part here, so I need to kind of mask them together. So in my first thought was, I'll just a mask on her foot, but then I'd be missing all this area right here, cause there's not enough information, so what I need to do is zoom out a bit, take my rectangular marquee tool, and get just enough, something like that, try that and see, and click on the add layer mask button. Now, it's close, but you can see how there are some aspects to it that didn't quite match up. And the colors are slightly different. So with that in mind, I still need all this information up here, but where her leg is, I'm trying to use as little of this one as possible, so even though I made the mask, this way, by using that initial selection, I'm gonna take my paint brush, do my little checklist, 100 percent opacity, everything looks good. Black is my forerun color, and basically hide as much of this other one as I can get away with, to keep the original one. And up here doesn't really worry me quite as much. So it's starting to look pretty good there. If I now wanted to do any kind of filter, which I probably don't but just for the purpose of demonstration. This is where we have to start putting our thinking cap on because if I say, well let's sharpen this whole thing, so I can print it. Well how do I sharpen two separate layers? I mean I could, but I end up with one layer sharp, and another one layer sharpened. So my thought would be well why not make my life simpler by taking both these layers and convert to a smart object so then it acts as if it's one layer eventually, and then I can do whatever filter I want like sharpening. Knowing I still have those separate pieces, so even after I've sharpened it, or whatever filter I've done, if I look close and go, I can still see a bit of a difference in the color tones on the wallpaper, I can still go back and edit it if I need to. So part of my purpose in showing you this one is that as odd as it sound sometimes that's what you need to do. Cause all I really wanted to make this photo look the way I wanted, was really just needed the rest of her shoe. But in order to do that, it took a few extra steps to get to that point. Now I look at this on my monitor, and it looks like, I can kind of still see the edge of this wall. So I might want to go back here to the original. See yeah, there's just a bit of a difference here. So looks to me like this one, if I add a levels adjustment layer, let's zoom in here a little where we can see where the problem is and just lighten this just a hair. Looks pretty good to me. There we go. Now, save that. And I can now, when this is finished saving, these are pretty big files, so it's taking a little longer for everything. But this is kind of the real world of what we're gonna do when we're working on a project like this, then I go back to this one again, apply filters, add type, whatever I want. So just another example of how you want to use, still all using layers, still using masks, it's just finding different ways to combine them, that's gonna give me the best result. Now some people would say, for something like that why wouldn't you just cut and paste and merge, and I could, but i still in the back of my head I'm going, just in case I see something later, I still want all those pieces. Keep in mind what this does, this file now is much bigger than it would have been otherwise. But how much is much bigger? I don't know, I can go back and find out if you really want. But I've yet, thankfully, to ever run into a situation where I tried to open something, and it's so big that is just doesn't work. It might slow down a little bit, but I'm always balancing slow down a little bit, vs the level to edit this one any level that I want. So she looks and says oh, I like it but, can you do whatever. I know I don't have to answer no to anything. Unless I just don't want to do it. But in terms of technically I can. So I personally live with the fact that file sizes get a little bit bigger so I can preserve this non- destructive way of working. Okay? And that's kind of the recurring theme that I want to end with, one of the things you find about Photoshop really early on when you start exploring things, is there's always various ways to do the same end result. And Adobe has chosen to leave functions in that frankly are, in my opinion, not the best choice. So as a result, people use them cause they're there. But if anytime you're doing something, I would encourage you even at an early stage, to go, is there a way I could do this with a layer, or an adjustment layer, or make it a smart filter, or and adjustment layer or a mask. Or something, it might even seem unnecessary, at the moment, but the more you do that on every project, then I hope that you never have this issue, but if there be times later on, and you're like, oh no, I need to edit this, and you still can. As I've said numerous times, there's nothing worse to me than that feeling of you open something and try and edit and you've got nothing. You've got like a jpeg period and there's nothing to work with. I'd much rather, if it's my work, have, open something and go oh, cool, smart object, layer mask, smart object, smart filter, cause it reminds me how I did it, it helps me be more creative and experimental, and in the long run it just makes it easier. It also means that even though there's who knows how many overall, I think I counted once, at the time, and I think that was probably a few versions ago, there were 610 menu items. If you look under every menu and counted, there was certainly over 600. I probably use 20 of those on a regular basis? I mean literally. So there's all these choices, and people are like how do I ever learn? You don't learn all those, you find the ones that work for you, and hopefully in this method that makes your life simpler because you're using methods you can repurpose and understand and be more flexible with.