RT controlling highlights, differences between various tools

I am away from the computer at the moment. Thing is that I can work around the issue in various ways but there are advantages to masking that is hard to really solve with other modules.

Im wondering if the masked* modules cant get to the pre exposure data by necessity or if is a bug or oversight. Alternatively if there is a way in say local adjustments to get to the data that I have missed.

Highlight recovery works quite well but has limitations as it cant be masked. Dynamic range compression is also quite good but sometimes looks odd in a way I’ve not managed to get around. Graduated filter has obvious geometric issues unless your scene is a horizon.

* I’m aware the term masked is incorrect but I used it as a shorthand for shape/area control