Use case
To implement something similar to the markups ROI where the control points are always visible (interactive / render order > 0) while the ROI cube occluded by the geometry (render order = 0) it would be necessary with the current architecture to split the display pipeline in 2.
Allowing for multiple render order would allow developers to select and compose actors per layers without having to add complexity / changing depth properties per actors etc.
Making the render order dynamic would also make sense to allow some interactor to be visible on top (during interaction / hover / etc) and set them back to a lower render order if display properties are changed.
Use case
To implement something similar to the markups ROI where the control points are always visible (interactive / render order > 0) while the ROI cube occluded by the geometry (render order = 0) it would be necessary with the current architecture to split the display pipeline in 2.
Allowing for multiple render order would allow developers to select and compose actors per layers without having to add complexity / changing depth properties per actors etc.
Making the render order dynamic would also make sense to allow some interactor to be visible on top (during interaction / hover / etc) and set them back to a lower render order if display properties are changed.