Kinks in moon and earth orbits

[alexibu]

Ive noticed that there are significant kinks in the earths and moons orbits.
Next to their current position, when viewing from a position perpendicular to the plane of the orbit.
It was even more evident when I increase the number of samples (nSamples) in the orbits to 10,000 from 100.
Does anyone know what this is caused by

[JackHiggins]

It's because the earth's orbit is slightly different every time it goes around the sun. There are gravitational effects from the other planets, the moon, and... um... (!?) The moon's orbit is somewhat egg-shaped, because the sun has an influence on its orbit also. Since celestia only draws the line for one orbit ahead of the current time, there is a bit of a kink.

What version are you using though? There have been 1000 points/orbit for a few versions now...

[alexibu]

Hi,
Now that I understand the reason for the kinks and how celestia samples the orbits: Would anyone be philosophically opposed to us moving the kink to the other side of the orbit ? i.e. sampling the orbit from -0.5 to +0.5 years instead of from 0 to 1 years.
This would help because you notice the kink most when zoomed in on the planet in question, and if the kink was on the other side of it's orbit then it would be less likely to be seen.
Alex

[pm]

Alexibu,
I believe that issue has been resolved in the latest CVS code.
pm

[alexibu]

Hi

Alexibu,
I believe that issue has been resolved in the latest CVS code.
pm

Noticed this when i was having a look at the fix for kinks in orbits


Orbit lines now go through the center of bodies which looks much better - however this makes the fact that the orbits are not depth tested even more apparent
Alex

[HankR]

Alex,

From your screenshot, it looks like there's just one too many sample points in the orbit. Shouldn't be hard to fix. (Famous last words...)

- Hank

[alexibu]

Hank,
I had nSamples set to 1000 instead of 100 to get better quality orbits - don't know if this would cause the above or not.
Alex