Johan, thank you. Yes, you and Davesscm describe well what happens. Sorry if I haven't expressed myself clearly: My curiosity is about why. Put crudely (and please understand, I’m no technician!)… : if pixels to interact with, then Blend; if no pixels, then Cover. My question (my puzzlement) is why, when defining Brush Blend Mode behaviour, was it decided that, where there are no pixels to interact with, it should be Cover rather than Do Nothing ? To give a simplistic example… (and leaving aside use of masks)… We have on a transparent layer a fishing net. We want to brushstroke a change of hue across a part or parts of that net. As things stand, a brush loaded with Blue, and with Blend Mode set to Hue, stroked across would paint all areas between the mesh of that net Blue. Who would want that? To me, that behaviour seems (always has) incomprehensible. Counter-intuitive. Of no sense at all.. Why not have it that those blank, no-pixel-content, areas are left unaffected? Perhaps I could ask it this way: what advantage or benefit is there to the current behaviour - blank areas being covered? What disadvantage would there be to a behaviour that left blank areas blank? If somebody could offer an example, I would sincerely be interested to know.
... View more