Photo666studio wrote:
Can someone provide some education on the workflow where a layer is clipped to the one below, and then you work on the one above.

If the lower layer is fully filled and without any mask it doesn't matter whether you "clip" the upper one to it or not.

I see this a lot with frequency separation but the tutorials don't fully explain.
They create the usual low/high split, then they duplicate each layer, and then clip the duplicated low to low and duplicated high to high and work on the clipped not doing much to the original underneath.

Can someone explain the advantages of doing that and what can be accomplished, methods, etc etc?

Nowadays almost everyone who records a screencast calls it a tutorial and everyone is some kind of teacher. Unfortunately most of them merely repeat mechanically what they have seen around the web. So don't expect any in depth explanations from those sources.

Generally it is a common (and wrong) practice used by many to simply duplicate a layer in order to have a "backup" in case they mess up something during working. People who do this rarely understand that duplicating a layer doubles the memory needed for the same thing and rely on things like "storage is cheap" and "i have a lot of memory". So they end up with 2Gb files with 20-30 layers and more - a "safety backup" which is rarely needed (if a tall). All that makes work slower, on whatever computer. Frequency separation itself is a way to use too much memory because of the duplicate data it creates. The benefits of using it are very limited compared to the problems it creates.

If you are looking for a good way to work efficiently: learn to use as few raster layers as possible and keep only the data you actually need in each layer.

Andrey Bautin wrote:
Beside a backup copy the real reason to use it is the ability to toggle the working high layer and see your progress while keeping low and high layer composited.

But you don't need to duplicate data for that. You can work on an empty layer on top and still toggle it. I.e. you are keeping only the difference (the actual work) which is much more memory efficient.

Also you could make a script to do it, assign a hotkey and remap it using autohotkey to anything you want. That way you don't even need to thing about it and toggle with just a press of a button.

You can still have that regardless of layer contents.

anchev wrote:
But you don't need to duplicate data for that. You can work on an empty layer on top and still toggle it. I.e. you are keeping only the difference (the actual work) which is much more memory efficient.

Not with the high layer though. You need to sample current layer in order to work on it, hence duplicating. But I agree in regards to low layers.

Andrey Bautin wrote:
Not with the high layer though. You need to sample current layer in order to work on it, hence duplicating. But I agree in regards to low layers.

For FS in particular you can also optimize the file by merging the split after you have finished and creating a diff with the underlying ones. The instead of 2 full layers you will have only 1 diff.

anchev wrote:

For FS in particular you can also optimize the file by merging the split after you have finished and creating a diff with the underlying ones. The instead of 2 full layers you will have only 1 diff.

Solid idea