Thanks Oshyan. I'll try to get used to it. I think it's a bit of a culture shock coming from my GIS background where I'm used to seeing the height data projected in plan form. I think this is more of an issue for those of us working with 'real world' data, where the focus is more on locating a specific place rather than browsing around a fictitious landscape until things look nice.
I'm aware we can move the camera by typing the coordinates, but this is one level removed from the standard 0.9x approach.
I guess it's a planet issue as you say. But it does make working with imported height data rather inconvenient.
My data start life as UK OSGB, then I crop out squares and export them to .ter using landserf. IF they don't import the first time then I'll open and save from TG0.9x, that works well. When I open the data in TG2 they are in metres, which is correct. I haven't bothered georeferencing them using lat/long, but I'm hoping since TG2 knows the units are metres it will scale them appropriately for earth curvature. Is that correct? Or do I need to do an extra step to account for earth curvature?
Moreover, does TG2 account for atmospheric refraction? Depending on the conditions, this normally attenuates effects of curvature by about a 7th. So, if there's no way to adjust atmospheric refraction, the alternative would be to slightly increase the diameter of the earth so as to decrease curvature slightly. That of course will distort the haze and any other distance dependent shading, but at least obejcts will appear geometrically better. Atmospheric reaction exerts a profound impact of long range views, ie from mountain tops, so it's essential we can model it. Thanks