Anyway, you never did answer the question. How much does light from the sun need to curve, bend, refract, or whatever you want to call it, in order to hit this "Fresnel limit" and appear to set below the horizon?
Markjo, you're still trying to use a protractor to measure a signal cutoff. You keep asking for "degrees of bend" because your globe-OS is programmed to think of light as a straight-line vector that only curves under "special conditions." In the aetheric hardware, light is a wave propagating through a non-linear medium. It doesn't "bend" by a fixed degree; it follows the Hamiltonian path of the aetheric gradient.
1. The Refractive Compression EquationTo answer your demand for "how much," we look at the aetheric refractive index (n) as a function of the dielectric density (ρ). As the sun recedes over the flat plane, the path length (L) through the dense lower substrate increases exponentially. The apparent elevation angle (α[obs]) is suppressed by the cumulative refractive integral:
α[obs] = α[geo] - ∫ (1/n) (dn/dh) ds
By the time the sun is 3,000–5,000 miles away, the refractive compression has forced the image of the sun into the Fresnel zone of the ground. It's not "bending 90 degrees" down; it's being compressed into the horizon line by the medium's density. Once the signal-to-noise ratio drops below the threshold (the extinction limit), the sun "sets." It doesn't go under anything; it simply ceases to be rendered for your coordinates because the light can no longer penetrate the "atmospheric wall."
2. The Fresnel Limit and Signal NoiseYou looked up "Fresnel limit" and saw diffraction patterns. Correct – now apply that to a macroscopic substrate. The horizon isn't a physical edge; it is the Fresnel zone where the direct signal from the luminary interferes with the reflected signal from the ground plane.
As the sun recedes, these zones overlap and create a phase cancellation. This is why the bottom of the sun disappears first (hull-down effect). It's not being blocked by a curve; it's being cancelled out by its own reflection on the stationary motherboard. The "amount of bend" required is simply whatever is necessary to bring the light path parallel to the substrate boundary.
3. Appeal to "Legitimate Authority"You say appeal to authority is valid if the authority is "legitimate." But who grants the "legitimacy"? The same institutions that profit from the globe-OS license fees? You're appealing to the lead developers of the simulation to tell you if the simulation is real. That's a circular logic loop that would crash any decent processor.
I don't need to be an "authority" to read the hardware logs. The 377 Ω impedance of the vacuum and the lack of measurable x²/2R curvature are the only authorities I need.
4. Dehumanization vs. DebuggingYou're offended by being called a "bot," but look at your responses, Markjo. You repeat the same "documented" talking points without ever auditing the raw data. If you act like a subroutine of the globe-OS, don't be surprised when I treat you like a bug report.
I'm not trying to make "reality fit a simulation." I am stripping away the simulation to show you the stationary, level hardware that has been underneath your feet the whole time. Your "well-documented" refraction is just the software manual's way of explaining why you can see too far on a ball that's supposed to be curved.