The orbits management system is suitable for some improvements

The representation of orbit paths for non elliptical or irregular orbits is not perfect. Some improvements are possible.

Description:

Actually the orbit paths are simulated sampling about 100 points during a complete revolution of the body at the start of the simulation.

The interpolation between the sampled points for the orbit of bodies is of linear type. This implementation has some ugly drawback. First causes that going in the future or in the past or accelerating the time, the body averts and accosts from the orbit path (specially for planets). Very ugly and disorienting. Look at the time in the following figure.

Should be advisable that at least the orbit path for the selected object should be resampled each frame. It is CPU expensive but using another type of interpolation should be implemented easier. A possible alternative technique is to use the Bezier interpolation that is already supported in OpenGL.

Applying this technique at extreme consequences should allow to calculate 4 nodes and 8 control points (or less) for each orbit path: 12 calculations instead of 100. But the real number of points that could be used to draw the orbit path could be even greater than 100. The Bezier interpolation could display different level of detail accordingly from the distance of the observer.

Second. For non regular orbits the line that joins the first and the last point of the path looks fictional and forced. Look at the following figure where I've evidenced the nodes and the real path of Callisto when completes the first orbit and starts the second.

Perhaps should be better if the path remains opened instead of forcing the closure. If the path could be resampled as described before should be easier to solve this problem.

When the time versus is changed (J key) the path should be resampled avoiding the discontinuity problem

Third. Specially in closer views and with accelerated time the orbit path of the currently selected object looks somewhat unstable blinking fastidiously. Probably this effect is caused by rounding errors. Probably resampling each frame the path should solve this problem.

Required/Suggested:

  (...)

Further related improvements:

  (...)

Related Newsgroup Threads:

  (...)

Reference Celestia Version:

  1.3.0

Status:

  (05-07-03) Submitted to forum attention.

  (24-08-03) Waiting.

Subscribers (Subscribe this item)

  Paolo Angeli, John J. Szucs.

Back to Group 3.
Back to Feature Requests.
Back to My Celestia