P: Add Layers to Lightroom

[bcw99]

I've seen a plugin that adds layers to LR which would save a lot of to-ing and fro-ing to Photoshop. The plugin is actually stand-alon, but also integrates with LR to some extent. It allows many of the layer options found in Photoshop. Not tried it but seems like a cracking idea! 🙂

Making LR more of an editor could make Photoshop redundant for pure photographic work

[TheDigitalDog]

Is there a difference in the resulting data using raw+parametric edit versus altering the values of existing RGB pixels?

[hillrg]

There could be.

[areohbee]

Perhaps the distinctions are not so important for the purposes of this thread.

I think we both agree that Layers could be added to Lightroom, with enough work. Whether they will be or not, only time will tell...

[FeedbackCommunityMember]

Andrew, you wrote "The term Non Destructive edit SHOULD tell us about what we just edited, not what didn't get touched, wouldn’t you agree? ".
I don't agree. The term "non-destructive" commonly refers to the editing paradigm, as opposed to whether or not the sum of all adjustments are information preserving. It may make sense to care about information preservation, but only when you don't use a "non-destructive" approach (see below), and I am not aware of a commonly accepted term for your notion of information preserving editing.

You asked "Any application that has a Save As command, or multiple levels of Undo is therefore non destructive?"
No, it isn't. I explained the difference many times before. But I'll try again:

Let's assume you have an image and adjust the black level. Then you remove some dust spots, and finally, you adjust the white balance.

Now, looking at the final version of the image, you notice that you went too extreme with the black level adjustment; that you are losing detail in the shadows. Hence you want to revise your "black level adjustment" decision.

If you have "Undo" functionality, you can undo the white balance and all the dust spot edits. Now you choose a less radical black level setting. Next you have to manually redo all the spot removal work and do the white balance correction again. You cannot automatically "redo" these last two steps, because you lose "redo" steps from the history once you change the image.

The same holds true for "Save As". Even if you saved intermediate working copies of your image+black level adjustment, image+black level+spot removal adjustments, etc., you can only go back, losing all edits that you performed after your black level adjustment.

So in a workflow that only supports "Undo" or "Save As", you are forced to redo editing steps if you need to revisit earlier editing decisions. Why? Because the edits have been destructive. When you make the black level adjustment to remove some details in the shadows, those details are destroyed for this version of the image because in a destructive workflow the old pixel values are overridden with new pixel values. You may still have an original image that you "Saved As" or you may "Undo" back to the original version, but these are different versions/copies of the image. The copy that is available for any subsequent editing steps has lost detail in the shadows. The detail is lost forever for this version/copy of the image. If you later just pull up the shadows, say after you've done the white balance adjustment, the detail cannot be revealed again.

Now consider the non-destructive case:
You have done your black level adjustment, spot removals, and then a white balance correction. You change your mind about the black level and make it less extreme to bring details in the shadows back. As you know from LR, you can just do that and still retain your spot removal work and white balance correction. Why? Because all editing steps are just transformation steps that can be "redone" even if their source pixels have changed. When you change the black level, a new version of the original is created that reflects the new black level adjustment. This is now fed to the "spot removal" transformation step that you have (implicitly) created earlier. Spots are copied/healed again, this time using different source pixels. The result of this step is a new intermediate image that is fed to the "white balance" transformation step that you have implicitly created earlier. (It doesn't work exactly like that in LR, because there is a fixed image pipeline, i.e., white balancing always happens early in the transformation chain, independently of when you decide to influence the respective parameters. However, this is not important to define what "non-destructive editing" is.).

You can think of a non-destructive editor as one that is capable of "Redo"-ing an editing action, even if the image that is the basis for a redo step isn't the same that was in place when the editing action was originally made.

Destructive editors cannot really "redo" anything. They can only offer you a version that you created earlier. There are dumb and clever ways to implement thist but in any event, a "redo" in a destructive editor is just undoing an "undo" step).

In summary, non-destructive editing is characterised by allowing a user to revisit editing decisions without causing the loss of any editing decisions that followed the one to be revised.

This is the principle that LR implements. Adjustment layer in PS follow the same principle (and I'm assuming they have been introduced to PS to equip PS with at least some level of non-destructive editing).

Destructive editing is easier to implement and more performant (because there is no need to retrigger transformations). But that's all. Destructive editing does not enable anything that cannot be achieved in a non-destructive paradigm.

If all the above is suitable to kill the "pixel-pushing-power" myth then it was worth it.

P.S.: Some may argue that my definition of "non-destructive editing" really describes "parametric editing" and that "non-destructive" is something less than that. These people may refer to Picasa as implementing non-destructive editing, since Picasa never writes over your original images either. However, all Picasa (at least the version I last used) does is not save over the original image file and keep track of image operations. Hence you can only undo, but not revise an earlier editing actions without losing all subsequent ones. If someone wants to call that "non-destructive" and refer to my definition as "parametric editing" that's fine with me. I don't really see the need for reserving a term for something as simple as Picasa does.

[TheDigitalDog]

First off TK, thanks for taking the time to explain this concept as you've done. I don't know I agree with the use of the term, but have a better idea of how you are defining the concept. At least the commonly referred paradigm just yet...

>>The term "non-destructive" commonly refers to the editing paradigm, as opposed to whether or not the sum of all adjustments are information preserving. It may make sense to care about information preservation, but only when you don't use a "non-destructive" approach (see below), and I am not aware of a commonly accepted term for your notion of "information preserving editing".

I care about both (obviously or I would not have kept referring to the edited data). I'm rallying around the idea that the commonly referred editing paradigm needs a lot more work since it only deals with one aspect of the editing (it ignores the effects of the edits).

>>Let's assume you have an image and adjust the black level. Then you remove some dust spots, and finally, you adjust the white balance.

Let's do that and go back to the differences between a pixel editor ala Photoshop and a parmatric editor that renders data since this is the crux of the discussion (If all the above is suitable to kill the "pixel-pushing-power" myth then it was worth it.) Let's look at both the original and the iteration too. And the differences in the two Adobe applications under discussion.

Clearly one level of undo or Save As isn't the same as your well defined explanation of 'non destructive editing' certainly in terms of flexibility. You can step back through edits and change your mind, you haven't permanently altered anything in the original document if you do the work correctly. Let's do this with your analogy of the edits in both LR and PS.

>>Now, looking the final version of the image, you notice that you went too extreme with the black level adjustment; that you are losing detail in the shadows. Hence you want to revise your "black level adjustment" decision.

In Photoshop, you have these edits, you must have the spot (cloning) done on a layer so that you do not stamp this on the underlying bkgnd layer. You can turn on or off the adjustment layers or alter the opacity, blend mode and so forth at any time. In your well written explination of non destructive editing, this would work out in terms of the original document data. But the cloning in PS would not unless you applied them on an adjustment layer since going back in History (which is limited and only hangs around during the one PS session) would lose those edits along with that edit to black level which happened historically prior to the cloning. Even using a History brush for a non linear history edit would only work if you didn't quit that PS session and you had enough steps to go back to. Otherwise you'd have to have the cloning on a layer and you'd have to just remove some of those edits from the layer with say an eraser. If doing so, you could (in your definition) call this non destructive. Even after you save and quit Photoshop. Of course this has no bearing on the result of the edits, we seem to agree that there will be data loss unless you do nothing but look at the PS image.

Now we are in LR. We don't have layers, we do have unlimited history that never leaves the scene. While we are viewing the edited preview (which may be partially processed raw data), we have instructions plus original raw data that is never altered so in the above definition this is non destructive editing.

The difference as I see it is that the non destrustive edit obviously applies to the source (raw) but also the iteration. Hence my question several times about the difference in the resulting data from the rendering versus the Photoshop data. Unlike Photoshop, where edits have to be applied to the source data (remember the idea of printing the data or doing anything other than just looking at the layers in PS?) the result is data loss to the iteration. In LR neither the original data nor the NEW data has undergone data loss. Is that fair? The reasons are (in my mind) due to the crux of our initial disagreement about the term 'non destructive' editing upon the final data that IS edited depending on if we have existing pixels OR instructions to create those pixels. I was envisioning only the results of the edited data, others were envisioning the results upon the original data. Both are important!

In LR we use the same order of editing you propose, we have a history too. But no layers. Yet since we can at any time revisit the clone stamp and delete it, no problems. It was never rendered out, no pixels were ever created with this edit. It does follow the above definition of 'non destructive' editing. But better, the result of rendering the raw+instructions produce a non destructive end result of the EDITED data. Is that fair? Is this then not more a non destructive process or workflow if we examine the result of the edited data?

You say you wrote a raw processor so going full circle back to "If all the above is suitable to kill the "pixel-pushing-power" myth then it was worth it", what's the difference here? It seems to me, the LR approach is super non destructive. Both the original is untouched and as importantly to me, the new data has also undergone no data loss (or significantly less data loss than the Photoshop model if you will). In my mind, Photoshop has existing pixel values and altering them causes data loss. LR doesn't suffer this problem because while the results are edited pixels, because the parametric instructions and the lack of actual RGB pixels (from raw) have to be created together, the LR rendered data is truly non destructive in your definition (original data untoched) and mine (the new data). If this is true, then a new term needs to describe the differences in the two processes. Non destructive editing in this case applies to both the original and new data. Is this not information preserving editing? Or there's no difference here in terms of the final data?

Lastly, how Photoshop stores the number of limited history steps and how LR stores unlimited history seems to be in large part due to the parametric edit model versus the edit on existing pixels, all of which need to be sucked into RAM instead of being actual textual instructions plus a rendering step. The former is JIT alteration of pixels even with a original rendered image. Better still is LR applies the edits in the best processing order rather than the user order. Photoshop doesn't and probably can't. In both cases, pixels are changed, otherwise why use either tool? But the result of the edit on the data and how that edit is applied (and when) is vastly different in my mind. I associate this due to a parametric editing process.

[FeedbackCommunityMember]

It'd be awesome to have Layers in Adobe Lightroom. Every-time you click/make an adjustment it creates a "history" but it'd be really cool if that history also served as a layer where you can adjust opacity, blend modes, masking, etc. 

[FeedbackCommunityMember]

LR is focused on organized, synching and practical options which are good but left behing the real photographic tool for digital photography. Adding layers is so important. Specially when you can select which is first, second etc. is changes the result differentely like PS. And Lightroom needs more photographic real filter from analog era. Like film glow, better grain engine, film stock engige and Luts. 

[Link 2 pc with lightroom]

Simple! I've my local adjustments with "pin" on my photo? Why don't hide
"pin" and create a level panel... So clicking the layer i can see my
mask, i can move levels up and down, i can overwrite my retouch without
clicking casually on another "pin".


Furthermore, it will be better insert color management into local adjustment panel... so if i've a orange floor and an pink/orange skin ... i can correct only the skin or the floor using color management but with mask of my brush.

[FeedbackCommunityMember]

> Specially when you can select which is first, second etc. is changes the result differentely like PS. And Lightroom needs more photographic real filter from analog era. Like film glow, better grain engine, film stock engige and Luts.>

So why not use Photoshop instead of turning Lightroom into another version of Photoshop? It comes with a subscription. I am not a huge fan of LR. I use it for some things, but I mainly have always used Photoshop because I must use layers for what I mostly do.

However, I understood adding Lightroom to Adobe's stable of apps, since there was a call for a relatively simple photo editing app that didn't take forever to learn and had a kind of guided workflow. I'm afraid I don't understand why Adobe should give up on that concept???  It's a program I can happily, and sincerely, recommend to my photographer friends who don't enjoy a lot of image editing or compositing, but want to be able to quickly prepare files without a steep learning curve and with the protection built into LR against destructive editing. As noted, filters can run within LR, so it's not as if there isn't access to most of the effects you might want to apply as a photographer.

[tomasnilsen.se]

Programs without this function fells really outdated, why not add it. And please fix the spot removal tool you have a superb one in Photoshop.

Not so intuitive in soon 2020 to need to jump into Photoshop to remove something and use layers

[Victoria Bampton LR Queen]

What do you imagine you'll do with these layers?

[FeedbackCommunityMember]

Since you get a full copy of Photoshop with the LR subscription it seems redundant to clog up LR with the overhead of Layers when a simple click to edit in PS gets me the full "layers" and other PS experience.

I do agree that the spot removal tool needs some improvement.

[JackFrusciante]

Aren't profiles and all the local adjustment tools in Lr already layers (or a kind of)?
How can you apply opacity without the concept of what is in front and what is behind?

[FeedbackCommunityMember]

Can we get the layerings for all the features available? This will make it a complete package for photographers.

[Jerry Syder]

Add your vote here - Lightroom: Add Layers to Lightroom | Adobe Photoshop Family

[edmundgall]

Layering is a good feature, but feels like another request to add something from Ps into Lr. If you merge the two, you may get the complete package for photographers, but at what price (money-, learning- & performance-wise)?

To use an analogy, Adobe is handling Ps & Lr development like F1 manufacturers with cars: the bleeding edge tech from F1 cars eventually gets transferred to their consumer market models. So, Jerry's advice above is sound: the more folks who vote for the features they want, the more feasible it may become for Adobe to approve the business case for copying them from Ps to Lr...

[SoyMarianoMarin]

Debería tener mas ajustes de control de iluminación, capas, Capture One actualmente es mejor revelador que Ligthroom, se debería empezar a plantear e igualar y mejorar a capture one, al igual que el precio final licencia de por vida son 149€  en adobe es carisimo

[John R Ellis]

Why conversations get merged: 

https://feedback.photoshop.com/conversations/lightroom-classic/lightroom-auto-import-to-folders-orga...

[Tatiana28101010sz9a]

I think Lightroom Mobile should have a simple layer system, where you can add new blank or transparent layer to the open workspace or even duplicate the picture layer, so you can edit some parts and then change the layer transparency to be able to choose the correct amount of editing that looks better. It would improve a lot the app!

[Daniel J24241165m679]

I would like to avoid going out to photoshop just to simply blend two or more images together.  My idea is that I select two or more images in the bottom bar, and then select Blend in the PHOTO menu (similar to the Enhance... or Photo Merge features).  A window opens that allows me to select a blending option such as multiply, burn, dodge, screen,...  By highlighting one of the selected images, I can set basic blending options such as opacity.   A preview is presented so that I can verify the blend effect.  Once the effect is adjusted, I can confirm the blend operation at which time the Blend window closes and the a new image is rendered and created in a stack with the others.

The main advantage to this feature is not having to go out to photoshop (with format conversions) and back to do a simple one step operation.  I can use lightroom to process my images with its tools that I am much more familiar with such as basic image processing and masking.

A specific example is to generate a true Orton Effect.  With blending in LrC, I can:  duplicate the image, adjust brightness and contrast, select only the brightest part of the image using a luminance range mask, blur the result, and then finally blend this with the original image.  The result is a true Orton Effect glow.

A similar workflow can be used to simulate a halation effect of old black and white and color films.

[Jeremias330945482cn5]

Vocês poderiam adicionar camadas para as fotos seria interessante.