Celestia Forums

I updated the katrina model at my web site to use clickable false. I noticed that there is a difference in the longitude for the model's location when I run the model on Linux using KDE. The difference is about 20 degrees. For Windows I need to use an offset of 277.5 and for Linux I need an offset of 257.5 to get the model to be at the same location.
cartrite

cartrite wrote:I updated the katrina model at my web site to use clickable false. I noticed that there is a difference in the longitude for the model's location when I run the model on Linux using KDE. The difference is about 20 degrees. For Windows I need to use an offset of 277.5 and for Linux I need an offset of 257.5 to get the model to be at the same location.
cartrite

The math should be identical between platforms . . . Are you absolutely certainly everything else is identical in your setups on Windows and Linux?

--Chris

chris wrote:The math should be identical between platforms . . . Are you absolutely certainly everything else is identical in your setups on Windows and Linux?

I really didn't "set up" anything. I use the same model with the same ssc file and it shows up in 2 different places. Windows shows it west of Florida. And KDE shows it east of Florida. I'm not sure what I could have "set up" to cause this. Compiler option?
cartrite

Sorry,
Found the dif. EDIT:I have different solarsys.ssc files
Seems I use the line for KDE.
And I use for Windows.
cartrite

cartrite wrote:Sorry,
Found the dif.
Seems I use the line

Code: Select all

   CustomRotation "earth-p03lp"

for KDE.
And I use

Code: Select all

   UniformRotation
   {
   Period              23.9344694
   Inclination        -23.4392911
   MeridianAngle      280.5    # offset at default epoch J2000
   }


for Windows.
cartrite

Interesting . . . Even with this difference, the orientation of Earth shouldn't be 20 degrees off. Then again, the default UniformRotation in solarsys.ssc may be suspect--the same MeridianAngle has been used for a very long time. I recommend comparing the earth-p03lp CustomRotation to the iau-earth CustomRotation. For the most accuracy, the Earth orientation from the p03lp precession model needs to be coupled with a nutation theory and a better model for Universal Time.

--Chris

Chris,
The model uses an invisible box called earth2 to orient the model that uses this So in a way both are being used at the same time. Maybe this is messing things up?
Not that it should matter but I recently updated it to use to move it a little.
The code I posted the last time was the settings for earth from solarsys.ssc
cartrite

cartrite wrote:Chris,
The model uses an invisible box called earth2 to orient the model that uses this

Code: Select all

   RotationPeriod   23.9344694 # 23.93419
   Obliquity        -23.45
   RotationOffset   280.5    # offset at default epoch J2000

So in a way both are being used at the same time. Maybe this is messing things up?
Not that it should matter but I recently updated it to use

Code: Select all

   RotationPeriod   23.9344694 # 23.93419
   Obliquity        -23.45
   RotationOffset   277.5    # offset at default epoch J2000

to move it a little.
The code I posted the last time was the settings for earth from solarsys.ssc

I recommend placing the model in the Earth's body-fixed coordinate frame instead of trying to match the rotation. You can then give the Katrina model a fixed orientation within that body fixed frame, or use a uniform rotation to rotate about the eye within the frame.

--Chris

I'll have to look into that.
cartrite