[xray360]

This has been bugging me for a while, but is the horizon line always going to be approximately eye level even at high altitudes. I know from basic perspective anything above the horizon line you will not be able to see it's surface, but does this apply in a drawing/painting that takes place high in the sky??

[math]

i dont get ur question. i guess its cuz im not a native english speaker.

well maybe this helps u

[Sumaleth]

At sea level the horizon line will be in the center of the view, assuming you're looking AT the horizon.

Now imagine yourself floating straight upwards. Once you reach space there's no way the horizon line is still going to be in the center -- so you can summize that the higher you are the lower the horizon line will be, assuming that you are looking directly forward.

In art, the horizon line is rarely in the center of the view, unless you're specifically trying to make a point about it being in the center. For whatever reason, it doesn't feel like good composition to us humans.
_________________
Art Links Archive -- Artists and Tutorials

[xray360]

[jfrancis]

It matters not what is the altitude of the camera (or viewer) but the ATTITUDE of the camera. "Attitude" meaning "orientation."

On a flat, infinite plane, a LEVEL camera will always see a horizon dead in the center. No matter how far above the camera rises above that infinite plane, it will ALWAYS see a horizon line dead in the center, because the plane itself is infinite. (there are no infinite planes in the real world, but our planet seems like one for most practical purposes)

This is a little counter-intuitive, so it bears repeating: on an infinite plane, a LEVEL camera will ALWAYS see a horizon line dead in the center NO MATTER HOW FAR ABOVE the infinite plane it rises.

The ONLY way to raise or lower the horizon line in a normal, uncropped image is not to raise the camera (Y-translation) but to TILT the camera (camera X-rotation) so that it is no longer LEVEL but instead it looks a bit up or a bit down.

A high horizon ONLY happens when a camera is tilted so as to look somewhat downward. Therefore a high horizon always goes hand-in-hand with keystoning verticals that converge toward the bottom of the image and spread toward the top, like you are looking down into the grand canyon.

A low horizon ONLY happens when a camera is tilted so as to look somewhat upward. Therefore a low horizon always goes hand-in-hand with keystoning verticals that converge as they rise, like you are looking up at tall buildings.

You can take one of the above scenarios and CROP it so as to try and fool someone into thinking it is the other scenario, but that would be a confusing and not particularly illuminating or helpful thing to do -- with one exception:

It is generally considered not only acceptable, but creatively desireable to take a LEVEL CAMERA (horizon dead center, verticals pretty close to vertical) scenario and RECROP it so as to favor sky over earth (or vice versa). This happens all the time.

...

so xray, to answer your question, you need to ask yourself: is the horizon low because of camera orientation? Or is it low because of the use of simple cropping to force the issue one way or another? In the case of your painting, I'd say you are in a level camera situation, and regardless of how you have cropped it, the horizon is FUNCTIONALLY as if it were in dead center for the purpose of judging perspective, and it has been dropped down through the use of cropping. Anything above the ocean's horizon should not be seen (unless it is not level, but rather tipped and cheated to show its top surface to the viewer)

...

Another (less likely) interpretation is that your painting is correct, an invisible horizon is dead center, we should and DO see the land mass's upper surface, but the low ocean horizon doesn't reach the true invisible horizon because the whole thing is in a Truman Show-set, and the ocean is small, and the wall is a relatively close, painted backdrop -- although I don't think that was your intention.