Bug or Something I'm Doing Wrong

Report bugs, bug fixes and workarounds here.
Topic author
sfo2phx
Posts: 7
Joined: 24.03.2006
With us: 18 years 7 months

Bug or Something I'm Doing Wrong

Post #1by sfo2phx » 24.03.2006, 04:50

I'm admittedly a newbie to Celestia, but I'm building my first hypothetical system and I'm running into a display problem with my orbit lines. Moon orbits (it seems to be only moons) are producing strange, double-broken arc lines when displayed:

Image

I notice this doesn't happen in our (prebuilt) solar system, so I'm wondering if I have some parameters incorrect somewhere. In addition, the planetary orbit lines do not run through the planets, but rather way off to one side. Are these related?

I've attempted to search the archives, but so far have come up empty-handed.

Here's the code from this particular planet and its moons:

Code: Select all

#Planet 9
"Olyxaa" "HD 34085"
{
   HazeColor [ 0.4 0.45 0.5 ]
   HazeDensity 0.3
   Texture "olyxaa.jpg"
   Radius       125000
   Oblateness 0.0863

   EllipticalOrbit {
   Period               22561.3276
   SemiMajorAxis        798.4415
   Eccentricity         0.0
   Inclination          1.09
   }
   RotationPeriod      12.364
   Obliquity      0.0
   RotationOffset      330.6
   Albedo         0.64
}

"Olyxam" "HD 34085/Olyxaa"
{
   Texture "olyxam.jpg"
   BumpMap "olyxambump.jpg"
   BumpHeight 3.0
   Radius      11236
   Oblateness   0.003

   EllipticalOrbit
   {
   Period          11.6069
   SemiMajorAxis    1309586
   Eccentricity    0.0135
   Inclination     0.297
   }

   Period      31.114
   Obliquity      9.04
   EquatorAscendingNode   186.6
   RotationOffset      135.4
   Albedo                    0.63
}

"Olyxan" "HD 34085/Olyxaa"
{
   Texture "olyxan.jpg"
   Radius       10110
   Oblateness 0.0046
   EllipticalOrbit
   {
   Period          16.5812
   SemiMajorAxis    1870838
   Eccentricity    0.0276
   Inclination     0.32
   }

   Period      30.114
   Obliquity      11.4
   EquatorAscendingNode   186.6
   RotationOffset      270.3

   Albedo                    0.55
}

"Olyxao" "HD 34085/Olyxaa"
{
   Texture "olyxao.jpg"
   BumpMap "olyxaobump.jpg"
   BumpHeight 2.5
   Radius       9580
   Oblateness 0.0029
   SpecularTexture "olyxaospec.jpg"
      Color [ 0.85 0.85 1.0 ]
      SpecularColor [ 0.5 0.5 0.55 ]
      SpecularPower 50.0
      HazeColor [ 1 1 1 ]
      HazeDensity 0.3
      Radius 2600
   Atmosphere {
      Height 98
      Lower [ 0.5 0.5 0.65 ]
      Upper [ 0.3 0.3 0.6 ]
      Sky [ 0.3 0.6 0.9 ]
      CloudHeight 9
      CloudSpeed 90
      CloudMap "olyxasclouds.png"
   }
   EllipticalOrbit
   {
   Period          26.5300
   SemiMajorAxis    2993340.8
   Eccentricity    0.0292
   Inclination     0.33
   }

   Obliquity        0.6
   EquatorAscendingNode   186.6
   RotationOffset      318.9

   Albedo                    0.43
}


Thanks in advance for any help...

Avatar
selden
Developer
Posts: 10192
Joined: 04.09.2002
With us: 22 years 2 months
Location: NY, USA

Post #2by selden » 24.03.2006, 10:14

Orbit paths usually don't go through the centers of objects because only a few points are calculated along the orbit and straight lines are drawn between them. (This approximation is used in order to minimize the amount of calculation needed.) If the object is not near one of the calculated locations, then it will not be near the line. You can edit celestia.cfg to increase the number of orbital points that Celestia uses.

The double line seems to be a bug in Celestia which is a side effect of the method used to draw the orbit paths in front of objects when that's appropriate.
Selden

Topic author
sfo2phx
Posts: 7
Joined: 24.03.2006
With us: 18 years 7 months

Post #3by sfo2phx » 24.03.2006, 13:38

Thanks, Selden! Glad to know the orbit thing wasn't something I was doing.

bdm
Posts: 461
Joined: 22.07.2005
With us: 19 years 4 months
Location: Australia

Post #4by bdm » 24.03.2006, 23:18

Here is another image of the orbit bug.
Image

I spent a few minutes studying this and I have come to the following conclusions.
* The full-orbit will jump around randomly, possibly due to floating point jitter.
* The half-orbit appears to be drawn correctly. While the full orbit is jumping around, the half orbit is steady as a rock.
* The magnitude of the error appears to be proportional to the orbit radius of the parent body, and inversely proportional to the orbit radius of the moon.

EDIT: This is Celestia 1.4.1pre1 on Windows.

Topic author
sfo2phx
Posts: 7
Joined: 24.03.2006
With us: 18 years 7 months

Post #5by sfo2phx » 26.03.2006, 16:13

That makes sense since my planets are orbiting at huge distances from the primary sun. I guess there's no workaround other than to simply turn orbits off. That's unfortunate because it makes finding things visually so much easier. Oh well! It's still an awesome program!

Avatar
selden
Developer
Posts: 10192
Joined: 04.09.2002
With us: 22 years 2 months
Location: NY, USA

Post #6by selden » 26.03.2006, 16:44

You might try drastically increasing the number of points used to draw the orbits. Edit celestia.cfg to do that. It's a text file with lots of comments.
Selden

Topic author
sfo2phx
Posts: 7
Joined: 24.03.2006
With us: 18 years 7 months

Post #7by sfo2phx » 26.03.2006, 17:02

When I increase the OrbitPathSamplePoints value to anything over the default of 100, I now get this on the orbits:

Image

The number of extra lines is dependent on the value of OrbitPathSamplePoints. In this case it was 1000, but even changing it to 150 causes two or three extra lines to be drawn.

This seems to be video-card related. It happens on my desktop system (NVidia) but not my laptop (ATI).

Avatar
selden
Developer
Posts: 10192
Joined: 04.09.2002
With us: 22 years 2 months
Location: NY, USA

Post #8by selden » 26.03.2006, 18:18

If you haven't already, you should install the current video drivers, which can be downloaded for free from Nvidia.
Selden

Topic author
sfo2phx
Posts: 7
Joined: 24.03.2006
With us: 18 years 7 months

Post #9by sfo2phx » 27.03.2006, 02:58

I did that right after I posted...and it solved the multiple line problem. Thanks!

bdm
Posts: 461
Joined: 22.07.2005
With us: 19 years 4 months
Location: Australia

Post #10by bdm » 27.03.2006, 08:44

I have discovered another related bug with the orbits of asteroid moons. The orbit of Dactyl around the asteroid Ida is drawn incorrectly. Only half the orbit is drawn, and the half that is drawn passes in front of Ida when it should pass behind. This can easily be demonstrated by making the orbit pass in front of Ida. When Dactyl follows this orbit it disappears behind Ida. When Dactyl passes in front of Ida it is not on the orbit.

Here is a screenshot. It shows Dactyl passing in front of the top half of Ida, with the orbit drawn on the bottom half of Ida.
Image


Return to “Bugs”