Sum0 wrote:Sorry to sound rude, but what does that mean?
The planetary textures are, I believe, specified as latitude-longitude grids. In other words, a distance on the texture in pixels corresponds to a certain number of degrees of latitude or longitude.
Now, approximating a planet as a sphere, an interval in latitude in the north-south direction always corresponds to the same distance
r * angle * (pi/180)
where r is the radius of the planet and the angle is the latitude difference in degrees.
But this is not the case for longitude. There is an extra factor. A difference in longitude in the east-west direction corresponds to a distance
r * angle * (pi/180) * cos (latitude)
where cos is the cosine, the thing that the cos button on a scientific calculator gives you.
Now the cosine is 1 at the equator, but it approaches 0 at the poles. So things get squnched together at the poles.
To compensate, you'd have to draw surface features stretched out by an inverse factor. How much is that? Well, a fancy name for the reciprocal of the cosine is the secant:
1/cos(latitude) = sec(latitude)
If your calculator doesn't have a "sec" button you can just as easily get it as 1 divided by the cosine.
If something is drawn on the texture as sec(latitude) times as wide as it is tall, then when mapped to the sphere it will look just as wide as it is tall.