Look at the hills in the middle of your screenshot. On the top, the left half of the hills is shadowed. The same part in the bottom picture doesn't have any shadowing at all, so it shouldn't be just more contrast.
That's because the left half is facing away from the sun. Simple gouraud shading
would also give this effect, and it has nothing to do with self-shadowing.
There are a few other hints that the terrain isn't self-shadowing (take a look at all the areas that are in shadow and compare them to the sun angle - if the terrain slope is facing away from and steeper than the sun angle, it should be shaded anyway), but one area in particular stands out:
That little bit technically shouldn't be lit at all since it would be in the shadow of the taller hill.
I'm respectfully asking- honest, no sarcasm- how can you tell it's high contrast shading... and... what the hell is high contrast shading!?
I'm just referring to the fact that there's more shading contrast than what we currently have.
The key term here is shading
, which in computer graphics is a way of lighting an object based on the angle of the face relative to the sun. If the face is perpendicular to the sun, it is fully lit, and if it's parallel or facing away from the sun it's fully shaded. It doesn't take into account any objects which may be blocking the sun. It looks identical to self-shadowing in all cases except at very low sun angles and very rugged terrain.Self-shadowing
, on the other hand, projects a shadow bitmap against the terrain from the sun's point of view and is much more cpu/graphics intensive. In the screenshot I posted, and in Remon's screenshot, shading
is used because it's still pretty effective (ie. it fooled two people on this thread into thinking it was self-shadowing). [created with Blender]
Of course, any of this is an improvement over the existing graphics engine where there's hardly any shading at all. The greater variation in tone in DCS World 2.0 is what makes the terrain 'pop' and I'm pretty excited to see it in game.