The bending of light to cause sinking is pretty common. This road effect also gives the same result when you look at it from different positions and angles near the surface.
VID snipped
Again, I haven't posted a video or other example of anything "sinking", nor of a "road effect"
Again, you're posting a near-surface video, whereas I keep asking you to address the examples I posted, where the observation sightline is 210m away from the surface. Looking at Rowbotham's drawings, and mine, they both suggest that if the observer at 210m looks past the top of a 210m object, that sightline should reach a higher object beyond at the 210m level, IF IF IF the Earth is flat. But it does not.
It passes clear above the hill, missing this point by a clear 200m (to the top of the hill) and more (which hasn't been measured, it's merely the amount of sky above the hill).
As can be seen from my scale model above, the sightline over 2-unit objects on a flat surface always meets a 4-unit object at the 2-unit level. It does this regardless of the distance to the camera, and the tilt of the camera. The sightline can be anywhere in the frame.
If you place the same three objects on a curved surface, the sightline between the two 2-unit objects, when continued to the 4-unit one, will hit a point ABOVE 2 units. The farther they are apart, the higher this sightline will go.