Celestia Forums

I am new both to Celestia and newsgroups, so please bear with me.
I have several questions:
How accurate is the orbital prediction revealed in Celestia? If I want to see the exact position of the planets like 10 million years from now, can I do that?
I have downloaded Thomas Guilpani's asteroid belt add-on, and have been using it.
How can I visually represent the asteroid belt as spots of light, rather than a haze of names? How can I see the orbits of the asteroid belt as a whole? Is it possible to add and remove orbit representations for particular bodies? (i.e. pick and choose the orbits that you will see?)
Is there ways to increase the visibility of certain bodies? To caculate the changing distances between the planets of somehow visually represent it? Is it possible, in other words, to use the program as a kind of dynamic learning aid for mathematics?
Might it be possible to visually represent magnetospheres and things like that? The electromagnetic noise of the stars, gas giants, and earth? The gravitational interaction between orbiting bodies?
I am a plainly self-educated kind of person, so forgive me if I've missed anything that might explain to me how to do these things with Celestia--if you know of any sites or such explanations, please pass them along.
Thank you,
Baby Beluga

Baby Beluga,

Welcome to the joys of Celestia. And finding things that could be done better I think half the fun is in creating your own add-ons.

Celestia's accurate planetary orbital calculations (up to 500 terms in the expansions) only work until 4000AD. After that it reverts to simple Keplerian orbits. Celestia doesn't try to model the solar system dynamics due to gravitational interactions, so it can't predict what will happen in the far future. You'd need to rent time on a supercomputer to do that!

Probably the best way to see the asteroids is to edit Thomas' file. You can change them from being Class "asteroid" to be Class "moon". Then the orbits will be drawn for those you've changed. To see tham as points from a large distance, you'd have to change all their diameters to be large (~2000km?) and give them high albedos ( e.g. 0.9). These changes will reduce Celestia's frame rate, of course, as it spends more time figuring out what objects have to be drawn on the screen.

This increase in size and albedo (brightness) can be used to make any body more visible. Ce;estoa's data directory contains the file solarsys.ssc which defines the characteristics of many solar system bodies. At least one person changed it so they orbit closer together -- more like an orrery, with the planet sizes much larger compared to their orbital distances.

Numerically displaying distances between bodies can be done pair-wise: select and goto one object, then select a second. The distance from your viewoint to a selected object is shown in the "heads-up" display.

A lesson plan for introductory Astronomy is available on the Celestia Web site. Maybe it could be modified to do something like what you want.

Extended objects like magnetospheres could be added to Celestia by creating translucent 3D objects. Celestia can display objects which are defined in the '3DS" format exported by the software package 3D Studio MAX and other 3D design software packages. Rassilon, one of the regulars here, has managed to do something similar when he created a representation of Eta Carinae's Humunculus Nebula. (It's available on his Web page on Shatters.net) Learning to use 3D design software is somewhat time consuming, unfortunately.

At the moment, Celestia is primarily designed to display the motions of massless satellites orbiting around a central object. It doesn't model gravity. It draws objects which are following pre-defined courses specified in .SSC files. It can be used to draw any Keplerian eliptical orbit, for example, but more complicated gravitational effects, like two co-orbiting stars, would have to be faked (e.g. both orbiting an invisible 3rd object).

Unfortunately, there really isn't a lot of documentation about what can and cannot be done with Celestia. Chris (ths primary author) spends most of the time he has avaiable on improving the code. Some of us have spent some time looking at the code and writing some brief notes. I've accumulated links to some of those sites on my own Celestia Web page. You can visit it by selecting the WWW icon just below my .sig here.

I hope this helps a little.

selden wrote:Baby Beluga,

Probably the best way to see the asteroids is to edit Thomas' file. You can change them from being Class "asteroid" to be Class "moon". Then the orbits will be drawn for those you've changed. To see tham as points from a large distance, you'd have to change all their diameters to be large (~2000km?) and give them high albedos ( e.g. 0.9). These changes will reduce Celestia's frame rate, of course, as it spends more time figuring out what objects have to be drawn on the screen.

To make the asteroids visible as nice points is a 1 second action with the following /absolutely trivial/ Perl script:

#!/usr/bin/perl
while (<>) {
if (/Albedo/){
@line=split(" ");
$line[1]=$line[1]*1.E+6;
print " $line[0] $line[1]\n";
}else{
print $_;
}

}

It just increases the Albedo of every asteroid by a factor 10^6.
The result of this action is exposed at Bruckner's site...

Place the code into a file called asteroids.pl, say. Then execute it from the commandline as

asteroids.pl asteroid_IMPS.ssc

In UNIX the Perl script has to be made executable with chmod +x asteroids.pl, of course...

Bye Fridger

I think I can understand why field gravity would be computation-expensive--you'd have to calculate every point in the field, right? (Or something like that.)
So--and please bear with me--is it that Celestia's only connection to real-life stars and planetary bodies is the information in the ssc files, which is a bit of visual information and orbital elements?
Now, I don't mean that you would necessarily have to show the actual effects of gravity, but would it be possible to make a visual representation of this? Maybe with some kind of scalable-color force-line kind of thing--just drawing lines between the bodies? Or something like that.
Unfortunately, I am primarily self-educated, so what I know isn't very complete or practical, but I seem to remember in my physics course in high school that there was a simple way to calculate gavitation: mass times the square of the distance, or something like that--would it be possible to make it so that these lines would visually represent (perhaps with brightness of shading color) that relationship of mass and distance between any two bodies?
Baby Beluga

Baby Beluga,

Yes, all the information that Celestia has about the Universe is represented in the .SSC files. It doesn't really know *anything* about gravity or the masses of the various objects, let alone gravitational field maps. It has to be told "this object was there following that path." It then transforms that description from the 3D Keplerian orbital coordinate system into xy coordinates to determine which pixels need to be lit up on the screen.

Celestia does include some internal routines with high precision formulas for the orbits of the planets of the Solar System and many of the major moons (the "CustomOrbit" definitions), but that's really not what you're looking for.

Sorry.

Selden

If you do a search on n-body simulation, you will find a lot of information about how difficult it really is to perform those computational tasks.

Simulating two or three objects is fairly easy, but once you get beyond that, it becomes very difficult.

By visualizing the mass and distance of objects between each other, what exactally are you trying to represent? Maybe you could visually represent lagrange points between bodies?

I suppose it would be cool to shade the areas around bodies with varying colors based on the inverse square law. I think there is a guy that created a patch to celestia that calculated gravity pull between bodies somehow. It wasn't a full n-body simulator, but functioned in such a way that it acts only on the observer. I believe it is called "mostly harmless"

http://mostlyharmless.sourceforge.net/index.htm

baby_beluga wrote:I think I can understand why field gravity would be computation-expensive--you'd have to calculate every point in the field, right? (Or something like that.)

Unfortunately, I am primarily self-educated, so what I know isn't very complete or practical, but I seem to remember in my physics course in high school that there was a simple way to calculate gavitation: mass times the square of the distance, or something like that--would it be possible to make it so that these lines would visually represent (perhaps with brightness of shading color) that relationship of mass and distance between any two bodies?
Baby Beluga