Yes, if you always make a duplicate layer or work with adjustment layers your original (background copy) will be preserved. But it will only be preserved if you save your work as PSD. Once you flatten the image or save as jpeg; the layers are lost. The alternative is to keep a backup copy of your originals, from camera, perhaps on DVD or an external hard drive and then you don’t have to worry about destructive editing or large file sizes.
But it will only be preserved if you save your work as PSD.
I am confused about the circumstances in which to use layer adjustments or Enhance> Adjust lighting> levels, for instance. The same then applies for Hue/ saturation versus adjust colour > adjust colour curves..
It would seem that using the Enhance path does not use layers and therefore requires a copy of the image to be adjusted. So, if I want to then adjust colours, do I use a another copy of the image for that purpose? I will end up with a big file!
This is made all the more confusing when there seems to be adjustment tools in the Enhance menu which cannot be found in the adjustment layers; eg, Shadows/ highlights, colour curves, colour variations, etc.
I like to use adjustment layers for non destructive enhancements and I guess this keeps file sizes reasonable as well.
I shall appreciate some advice, please!
If your question is : why is there no adjustment layer for enhancements like shadow/highlights ?
The reason is that those kinds of enhancements do not work like the other kind of adjustments.
What are adjustment layers ? In this digital world, each pixel of the source layer has a specific value (color : R, G and B value). If you can apply the same recipee (a math formula) to each one of the pixels of the source layer to calculate the value of the resulting layer, you simply have to record and save the recipee ; that makes saving the formula much smaller than saving the results of the calculation to all pixels in the resulting layer.
Not all kinds of adjustments (enhancement) use the same recipee for each pixel. Typically, sharpening, blurring, enhancing local contrast, rely on calculations made relative to adjacent pixels on the same layer. That does not work like the above method, which fools you by showing a 'virtual' layer in the layer palette. The values of each pixel in the resulting adjustment layer are not saved, they are calculated on the fly.
By the way, the shadow/highlight adjustment is not a simple curves adjustement (available in the full Photoshop CS), it adds local contrast adjustments based on adjacent pixels in the same source layer.
Instead of using adjustment layers, you can use a procedural approach : saving each command with its parameters in an action or a script. Another way is to use the ACR plugin (or Lightroom) to save all your enhancements in a parametric approach : all the recipees (the slider values) are saved, but only the original values are saved, not the values of the final pixel.
I hope I have not made things too complicated...
Of course, you are right to prefer adjustment layers whenever it's possible.
Thanks Michel, that is more in line with what I want to know.
I find the Shadows/ highlight adjustment really good for a quick and effective adjustment when an image has a HDR and I have not got bracketed images which could yield a merged image. I cannot achieve such an adjustment with layers although I have not tried Dodge and Burn adjustments yet. Perhaps their lies the answer.
I am not sure I understand your explanation but I have begun to get a sense of what you are saying. I could not explain it to another person though. It seems that the non-layer method does change the recipe of the image. What is the effect on the image of having several versions of the original each with a non-layer adjustment sitting over the original? I presume that they will have to be merged or flattened to achieve the result sought?
Thanks for your help.
It's not easy to understand how an image is transformed via enhancements, layers or filters in the editor, or via sliders in ACR/Lightroom. For a photographer, what's important is to know what kind of change you want in luminosity, contrast, color, sharpness etc. Then you'll be able to find the right tool. This requires some education or training, for instance being able to tell if an image needs global adjustments like levels or local contrast (shadow/highlight, unsharp mask with high radius, clarity in ACR...) Understanding our visual perception helps, especially with HDR scenes or black and white conversion. Understanding the math behind the tools may be useful, but not necessary. That can explain why different approaches (workflows) require different tools, why some effects can't be achieved with adjustment layers.
My practical advice would be to concentrate on the effects of layers, blend modes and masks.
What is the effect on the image of having several versions of the original each with a non-layer adjustment sitting over the original? I presume that they will have to be merged or flattened to achieve the result sought?
The answer lies in blend modes, masks and opacity. The layer palette may show many layers in normal blend mode without masks. Each layer hides what is under it, and acts as an history step. Only the top layer is visible, but if you want to reprocess your edits, you can go back to a former step.
With adjustment layers or blend modes, you have various recipees to mix the pixel values of different layers. A few blend modes are easy to understand : lighter, darker. It's useful to know what the effect of soft light or similar may be ; a lot of others are rarely used.
A very useful trick to 'freeze' the effect of several layers into one resulting layer is to 'stamp visible' which is achieved by the following shortcut : Ctrl Alt Shift E. Remember that you can show or hide individual layers.
You can limit the effect of an adjustment layer to just one layer instead of applying it to the combined effect of all underlying layers : Ctrl G.
Appreciate your patient and informative reply.
I now have a clearer idea of how to use the adjustment alternatives and looking at some video tutorials (in some cases for the second or third time!) on this subject has helped as well.