I was able to reproduce the original masking issue and logged it for our devs as (Report ID=CMAX-1600).
Before we make any changes, I would like to make sure it makes sense. This may sound like a silly question, but why is the "adding the main forest object generates a mask" solution better than the "adding source objects generates masks for them"? What if a user wants to render different masks for different tree types?
In a perfect world both ways is great. But when you have a lot of scatters, that use a lot of geometry to make up those scatters, and all you want to do is color correct say the trees or grass you don't have to find all the geometry used in those. It could have grass, moss, and each of those could have 20 individual pieces of geo each. Let alone when you use Forest pack library (Or 3rd party), they come in hidden and frozen, and some times not at 0,0,0 so you have to then find isolate and select. Like it literally adds so much time when you just want to grab the FP object and say ID 01 please. Also if you are not the artist setting up the scatter and geo but are tasked with rendering and lighting, you then have to find all that crap instead of just the FP object.
The reasons list goes on but you get the point.
But I still do want to be able to select and mask individual objects as it is useful, but want to more so be able to select just the FP object.
