sure, i get that
but i think thats missing my point (which would be cos i haven't made it so well)
if we consider the half-moon shadow line (for want of a better term) it remains parallel to the horizon at the equator for it's entire journey across the nights sky (correct me if i'm wrong)
likewise its orientation is different at other latitudes, but remains consistent viewed from the same spot as it traverses the sky, (of course it is flipped over like a pancake, but you know what i mean)
the angular displacement between the equator and poles is exactly 90 degrees and 180 from pole to pole.
in FE i get that the image flips due to point-of-view, but why doesn't the parallel-to-horizon half moon stay parellel to the horizon at various latitudes? and would the moon flip gradually, 180 degrees, between the 'poles'?
It never flips over like a pancake. It only ever presents one side (the red/green (or golden brown if you like) face) the dark side (the side of a pancake that looks a little bubbly) always faces away. It only rotates around the axis pointing directly towards us. The Moon is so very very freaking far away that we only ever are looking at the one side (and because it's tidally locked).
I think I know what your getting at - that the line separating light/dark side is parallel to FE surface but not to RE surface. And you are making a correct observation, however for that to make any difference it would have to intersect one of the viewing surface to to the North or South be visible. Since it does not intersect the surface in either model and it's an imaginary line, there's no way to determine what the shape of the viewing surface is.
In the East to West direction, the drawing in shows what it would look like from an observer standing East of the two observer in the drawing. Because the moon is so far away they still only see the underside, red/green side, but to them red would be on the right and green to the left. Someone to the West, like on the other side of your screen, would have to turn around to see the Moon. They have a different concept of left and right than you, it looks like red is on the left and green is on the right.
I don't have Visio available right now to draw with. But if you relabel look at the last drawing, relabel N and S as W and E, and change 'Equator' to 'Prime Meridian', you're now looking down at the North Pole and the people are now sanding on the Equator, looking E/W or W/E, respectively. The Moon, doesn't move, as it presents it underside to everyone at all times.
Hopefully, you starting to see how 'UP' is relative to the observer, not the shape of the surface. On RE everyone senses UP as above their heads, and can't tell other people's up is different. On FE everyone senses UP as above their heads, but can't tell other people's up is same.
I will repeat that I think the moon is the best proof of a round earth that is visible to everyone. During the arctic winter, the full moon goes around the viewer and is visible the entire night. If you are in Norway, Siberia or Canada, the full moon circles you with the same face visible all the time. How can it face all three viewers all the time on a flat model? If it were an automobile, one should see headlights, or tailights, side doors and undercarriage as it circles around the overhead track if it is circling the north pole. And then again we have to assume the people south of the equator are not as important and can be ignored.
I'm not arguing an FE model here. Only that
The inversion of the Moon's face is not an issue, inversion alone could be consistent with an FE model.However, to your question,
How can it face all three viewers all the time on a flat model?. The answer is the same way it does on a spherical model. By being both (a) 'tidally locked' and (b) very far away (way more than 3000 miles).
I've talked about how the Moon is not 3000 miles away in this thread:
https://forum.tfes.org/index.php?topic=8653.0