Hi @RobH,
Thanks for the feedback. Can you clarify when it started tanking in this way? 5.0.5, or prior?
My hunch is that we're seeing some major performance hit with the code implemented to occlude mountains being seen through clouds at certain levels. The hit wasn't large on my end, but it's becoming evident this may not be working out. Originally we were reserved to thinking being able to even occlude the mountains was not going to be possible with what Laminar provides us and how volumetric clouds work in the depth buffer. Once we saw more reports of it crop up, we started discussing how to tackle this, that there had to be a way, and discovered there was a potential path forward but it would be a multi-month re-write for a projection method we hadn't thought of. Later that day, @sundog came up with a way to make it work that we were happy with, but that happiness is not working out I think.
For the most part, I think your observations on what's causing a performance hit or not are mostly conditional. How ray-casting works, certain scenarios cause different hits, all the way down to how low cloud levels are/distance from your aircraft.
Appreciate you taking the time to write!