Relationship to Planet Size and Starting Elevation

[D.A. Bentley (SuddenPlanet)]

Thought this was an interesting find.  I was wondering why the default planet had a default Elevation Value (Y) range hovering around -7m under the camera with a constant color fed into it.

I figured out the planet size is what determines the starting elevation (maybe other factors too?).  Incidentally the ground level accuracy seems to fade as you sample (mouse hover) the terrain further away from the camera.

Default Planet Size:  6.378e+006

Near Perfect Zero Elevation Planet (ie Ground Level near Y=0mm):  6.378000001e+006


Why I used a Constant Color fed into the Planet node was because even with just the Default Base colours (Power fractal shader v3) node fed into the Planet I was still seeing some slight displacements (Displacement Tab "Apply displacement" unchecked/off).

I disabled the default Fractal Terrain, Warp, Compute Terrain nodes, and even disconnected graph.  I couldn't understand where the displacements were coming from, so I tried the Constant Color.  I'm still on TG 4.4.67 for reference.

Anyway, I thought that was interesting so I hope it tickles your curiosity. 

Derek

[D.A. Bentley (SuddenPlanet)]

Actually I just realized Terragen seems to round off the Planet Size, so after I entered 6.378000001e+006 it just rounded back to 6.378e+006.

Also I figured out that the Default Planet does have a zero starting elevation.  I must of had a corrupt Terragen session running, because I can't reproduce what I posted previously.

[WAS]

Changing the Simple Shape fuzzy zone can easily disrupt the X:0-Z:0 area. Maybe that's what happened. Because the default terrain is multi-scale modulated for variation method, which will go into the negative values easily. I try to actually used Unclamped Multifractal as it tends to build from the underlying displacement better without cutting into it.

[D.A. Bentley (SuddenPlanet)]

Changing the Simple Shape fuzzy zone can easily disrupt the X:0-Z:0 area. Maybe that's what happened. Because the default terrain is multi-scale modulated for variation method, which will go into the negative values easily. I try to actually used Unclamped Multifractal as it tends to build from the underlying displacement better without cutting into it.

I'll keep that in mind.  I'm not sure that was it though because I had everything disconnected from the planet node except a Constant color.  I even disconnected that and was sampling heights directly from the planet.  I think my session just got messed up, or maybe I was seeing mm and thinking it was still displaying meters or km.

[Dune]

First of all; the simple shape uses a smooth edge of 90%, and I think this will raise or lower the elevation at the pole slightly. If all above the base color is disconnected it should be at zero.
And Y is the altitude above 0/0/0, and the further away from the pole, the curve of the planet will dive under that Y=zero, as Y is taken from a flat 'plane' at 0/0/0, kind of.
The altitudes above a 'zero surface' in say the surface shader can be 'real' if Y is unchecked, if you get my point.

[WAS]

First of all; the simple shape uses a smooth edge of 90%, and I think this will raise or lower the elevation at the pole slightly. If all above the base color is disconnected it should be at zero.

Basically 10% of the 10000m will be at the planets base elevation 0f 0 at 0,0,0. So probably 500m around 0,0,0 is unaffected before fuzzy zone starts effecting it, though maybe it's immediate from just 1m away from 0,0,0 (the fuzzy zone of the 10% to the rest of the fade)