fill layers vs filling a layer

Yes - with a colour fill layer you can click on it, to open the colour picker, and change the colour anytime.

As c.pfaffembichler says - the aim is to make changes non-destructive which means keeping the ability to alter them later.

Dave

Are you aware of the concept of Non-Destructiveness in image editing?

A number of issues can be considered under the header »non-destructive workflow« in Photoshop:

• Performing Touch-up (Clone Stamping, Spot Healing etc.) on separate Layers is advisable to keep the original image-data intact

• Layer Masks are preferable to deleting and erasing – but what can be achieved with a Clipping Mask to an existing Layer should not be done with a Layer Mask

• (Adjustment) Layers should be masked as a Group instead of individually if they are to affect the same region

• Finished Layer Masks of meaningful units should preferably not be mutilated but additional Layer Masks applied on Groups

• Adjustment Layers are preferable to applying an Adjustment directly whenever possible, but they should be kept to as small a number as possible

• For flat or patterned areas Solid Color- and Pattern Layers are preferable to regular Layers

• Transformations, Puppet Warps and Filter applications are best done on Smart Objects

• Clipping meaningful units from a Smart Object is better done within it than in the containing document so that the Layer Mask is created with the Smart Object’s full resolution and does not deteriorate with repeated transformations

• Multiple instances of elements (more intricate than achievable with a Brush Preset) in one file should be done with Smart Object-copies if their number and size don’t bloat the file to impracticability

• Color conversions are detrimental to image quality and should be kept to the possible minimum – this includes conversions to L*a*b

All these provisions should help keep as many of the changes performed on an image edit-able and remove-able as possible.

And while the use of Smart Objects can cause significantly increased file sizes and saving times the advantages can be significant.

A Pattern Fill Layer allows for later edits (for example changing the Layer Mask or the Pattern itself, scaling it, …) while applying the Pattern destructively is kind of a »dead end«.

Edit: And even for a Solid Fill Layer changing the Color would be easier that having to select the pixel Layer content and filling it again.