Working Edits In Both Lr & Ps

[smoke665]

@Ysarex to summarize the issue - your quote from the other thread.

• You believe "that an empty clone layer is still a raster layer and if you do a clone operation to remove something from the sky and then later change the color of the sky in the layer below the clone work on the empty layer will show as a different color. The empty raster layer is still a pothole in that you'll either have to re-do it or separately apply a change to it directly."

I think I know where you're going off track and I'll explain that later, but for now let's start with the basics. Here we have a DNG file pretty much SOOC, exported from Lr. I picked it because I thought the sky would show any color shifts better. Note the big antenna.
#1

Next photo. I created a "Virtual Copy" of my DNG file in LR. Note my original is still intact same as above. I right clicked on the copy and sent it to Ps. where I cloned out the top of the tower, and my original DNG file is still safely tucked away.
#2

Next photo - with the image still in Ps, I created a "Virtual Layer" above the clone layer using the Shft + Ctrl + Alt + e short cut. This created an image layer which contained all the changes below. I then ran camera raw filter on this layer and changed the white balance way over to the blue. Note the cloned areas were changed along with everything else, and my original DNG file is still back safely in the file.
#3

Next Photo - With image #3 still in Ps, I decided I didn't like that color and I wanted to think about it. So I deleted the Virtual Layer, then saved it to Lr, which is actually the image I show as #2, so no potholes yet. Now that I'm in Lr, I settled on a color, so using the develop panel in Lr I really warmed it up, but then I decided there was more editing I wanted to do in Ps so Rt click edit with Lr edits and I'm back in Ps where I exported image #4 from. The cloning I did in #2 reflects the temperature change.

#4

If I save them to Lr, each of these edits can then be used to create another virtual copy to edit further. You don't have to but sometimes it's nice to have a stop gap.

As I stated at the beginning I think I figured out where you're off. I know nothing about how Capture 1 works, other then a few things I've read. I believe it was a couple years ago that they also went to a layer structure similar to Ps. Not sure about that, but in any case the layer structure used in Ps is top down, not bottom up editing. If you create a raster image layer then several layers later, you go down and change that first layer the edit doesn't travel up the stack. In the case of the examples you were in essence pulling a layer out of context from one image to another, by changing the WB on a lower level layer and using the clone layer from another stack where it was higher up. As you pointed out that won't work, it has less to do with non destructive editing and more to do with using the program incorrectly. I'm sure there are things in Capture 1 that would be equally off if used incorrectly. Granted there are advantages to parametric editing but sometimes it takes more. It's like trying to drive a railroad spike do you need a tack hammer or a sledge hammer, and we've already talked about the inherent slow down that comes with track all those instructions. I use both smart objects and layers frequently because it allows me that flexibility to make changes later.

I have no first hand knowledge of Capture 1. From what I hear it's a good software, but I'm not sure that it's any more price competitive then Adobe. Their subscription price is $20, twice the price of Lr, Ps, and Bridge combined. I can pay the monthly subscription to Adobe for 30 months for the price of C1 to purchase, during which time I get every update that comes out, whereas it's out of pocket for updates to Capture 1. I also thought it was interesting, does Capture 1 still fail to recognized other camera manufacturers in the medium format range? Finally add to that the fact that there are a ton of 3rd party suppliers offering everything from presets, actions, templates, textures, training.......for Lr and Ps. Think I'll stick to what I have for awhile.

 

[Ysarex]

@Ysarex to summarize the issue - your quote from the other thread.

• You believe "that an empty clone layer is still a raster layer and if you do a clone operation to remove something from the sky and then later change the color of the sky in the layer below the clone work on the empty layer will show as a different color. The empty raster layer is still a pothole in that you'll either have to re-do it or separately apply a change to it directly."

I don't believe that I know that fact.

I think I know where you're going off track and I'll explain that later, but for now let's start with the basics. Here we have a DNG file pretty much SOOC, exported from Lr. I picked it because I thought the sky would show any color shifts better. Note the big antenna.
#1
View attachment 169438

Next photo. I created a "Virtual Copy" of my DNG file in LR. Note my original is still intact same as above. I right clicked on the copy and sent it to Ps. where I cloned out the top of the tower, and my original DNG file is still safely tucked away.
#2
View attachment 169439

Next photo - with the image still in Ps, I created a "Virtual Layer" above the clone layer using the Shft + Ctrl + Alt + e short cut. This created an image layer which contained all the changes below. I then ran camera raw filter on this layer and changed the white balance way over to the blue.

Nope! You color adjusted an RGB image that already had the WB burned in. I don't want to do that. I want to change the raw file WB. Before you send the virtual copy from LR to PS look in LR and note the color temperature value. When you're in PS add 2450 to that existing value. If you can't do that you're not doing what I can do and what I want to do. Using Camera Raw filter to color adjust an RGB image does not produce the same result and is unacceptable.

If when you went from LR to PS you sent PS a smart object then you would be able to dbl clk the smart object icon on the smart object layer and in fact open ACR and change the WB. Unfortunately you can't do that with a virtual layer. And if you do open ACR using a smart object and there's a raster clone layer above it you're potholed.

Note the cloned areas were changed along with everything else, and my original DNG file is still back safely in the file.
#3
View attachment 169440

Next Photo - With image #3 still in Ps, I decided I didn't like that color and I wanted to think about it. So I deleted the Virtual Layer, then saved it to Lr,

Yikes! You mean you saved a gigantic 100+ mb RGB image in addition to your raw file. I don't want to do that. You're saving an intermediary RGB image file and your editing is no longer non-linearly re-editable.

which is actually the image I show as #2, so no potholes yet.

Of course there's a pothole. You can't go back to your raw file or virtual copy of your raw file and change the WB.

Now that I'm in Lr, I settled on a color, so using the develop panel in Lr I really warmed it up, but then I decided there was more editing I wanted to do in Ps so Rt click edit with Lr edits and I'm back in Ps where I exported image #4 from. The cloning I did in #2 reflects the temperature change.

#4
View attachment 169441

If I save them to Lr, each of these edits can then be used to create another virtual copy to edit further. You don't have to but sometimes it's nice to have a stop gap.

As I stated at the beginning I think I figured out where you're off.

I'm not off.

I know nothing about how Capture 1 works, other then a few things I've read. I believe it was a couple years ago that they also went to a layer structure similar to Ps. Not sure about that, but in any case the layer structure used in Ps is top down, not bottom up editing.

It needs to be both. The layer structure in C1 is top down, bottom up, middle up, middle down, all directions from any point. And that is part of the point: non-destructive & non-linearly re-editable.

If you create a raster image layer then several layers later, you go down and change that first layer the edit doesn't travel up the stack.

pothole.

In the case of the examples you were in essence pulling a layer out of context from one image to another, by changing the WB on a lower level layer and using the clone layer from another stack where it was higher up.

I did that because that's what you said you could do: "Even if it doesn't it's a moot point because I still have my original in Lr. In a matter of mins I can create a new virtual copy change the color temperature, open it in Ps, select and copy the layers from the old edit, and paste them on the new copy. I guess I'm not understanding your concern."

As you pointed out that won't work, it has less to do with non destructive editing and more to do with using the program incorrectly. I'm sure there are things in Capture 1 that would be equally off if used incorrectly. Granted there are advantages to parametric editing but sometimes it takes more. It's like trying to drive a railroad spike do you need a tack hammer or a sledge hammer, and we've already talked about the inherent slow down that comes with track all those instructions. I use both smart objects and layers frequently because it allows me that flexibility to make changes later.

I have no first hand knowledge of Capture 1. From what I hear it's a good software, but I'm not sure that it's any more price competitive then Adobe. Their subscription price is $20, twice the price of Lr, Ps, and Bridge combined. I can pay the monthly subscription to Adobe for 30 months for the price of C1 to purchase, during which time I get every update that comes out, whereas it's out of pocket for updates to Capture 1. I also thought it was interesting, does Capture 1 still fail to recognized other camera manufacturers in the medium format range? Finally add to that the fact that there are a ton of 3rd party suppliers offering everything from presets, actions, templates, textures, training.......for Lr and Ps. Think I'll stick to what I have for awhile.

I only used C1 because I'm most familiar with C1 since I use it daily. Forget I said C1 and let's just keep it LR in place of C1. Here's what I want: non-destructive & non-linearly re-editable. I can get that in LR. You can't get that if you add PS into your workflow. You have to settle for potholes.

Go back to your original photo (and I know that bridge! That's the Mississippi River bridge at Cape). In LR before you send it to PS go to the FX tab and add some grain simulation. Set values of 40, 20, 40. Now send your image over to PS and clone out the tower. Do whatever else you want to do. Create a virtual layer if you like but then change your mind and turn off the grain effect and you're potholed.

Joe

 

[smoke665]

@Ysarex you do realize that two Missouri mules arguing a point is like wrestling a pig in the mud, at some point you realize the pig's enjoying it. I'll study up on the points you posed and get back to you. May be a bit as sadly I learned that a friend in Cape passed away yesterday. So looks like I'll be headed to Missouri.

 

[Ysarex]

@Ysarex you do realize that two Missouri mules arguing a point is like wrestling a pig in the mud, at some point you realize the pig's enjoying it. I'll study up on the points you posed and get back to you. May be a bit as sadly I learned that a friend in Cape passed away yesterday. So looks like I'll be headed to Missouri.

Sorry for your loss. Take Care.

Joe