Celestia Forums

Sorry to bother anyone that might be bothered by this.
I downloaded Celestia yesterday because I am very interested in
Astronomy, but I couldn't help but think that such a great program
is possible for someone like me. Anyways, - my question is - is
everything in Celestia factual? I mean like every single possible thing
on the program, is it really true? Everything is to our best of knowledge?

Thank you all for your time.

Everything in Celestia is to the best of our knowledge AND to the limits that about 30 fps give to our computer hardware today. That means, the numerical equations aren't computed up to the 16th term.

So everything is quite exact as viewed from earth, but some orbits or spacecraft tracks may be quite ~1000km off reality. That's about an 0.0000...? % (who knows?) error.

maxim

PS: There are of course some fantasy addons available.

A few things worth noting:

- Some surface textures are extrapolated. Use '+' to toggle the Limit Of Knowledge surface textures (see Mercury for example).
- Most small bodies are represented by the same asteroid model.
- Not all known Solar System bodies are included in the default installation (add ons for the asteroid belt are available).
- Light travel delay is not included by default in the simulation, for example transits or occultations do not happen at the correct time when observed from Earth, you need to press '-' to toggle the light travel delay correction.
- No relativisitic effects.
- No proper motion of stars / galaxies.

Christophe wrote:...
- Light travel delay is not included by default in the simulation, for example transits or occultations do not happen at the correct time when observed from Earth, you need to press '-' to toggle the light travel delay correction.
- No relativisitic effects.
...

Hi Christophe,

while I agree of course with your explanations, it might nevertheless be worth remembering that the mere existence of the LT delay is the most basic "relativistic effect" together with the (supposed) universality i.e. frame independence of the finite speed of light. From the finiteness of the latter then follow all the familiar formulae of special Relativity, i.e. length contraction and time dilatation...

Bye Fridger

t00fri wrote:Hi Christophe,

Hi Fridger, it's been a long time...

t00fri wrote:while I agree of course with your explanations, it might nevertheless be worth remembering that the mere existence of the LT delay is the most basic "relativistic effect" together with the (supposed) universality i.e. frame independence of the finite speed of light. From the finiteness of the latter then follow all the familiar formulae of special Relativity, i.e. length contraction and time dilatation...

I give you that, but you must admit that as currently implemented the LT delay is just a hack (useful and working, but a hack nonetheless).

By the way I'm on vacation at the end of next week, and I really intend to get back to some Celestia hacking this time. The new key mapping code is long overdue now, as is 1.3.2 -- a special Cassini-Huygens Edition would be nice, don't you think?

Christophe wrote:

t00fri wrote:Hi Christophe,

Hi Fridger, it's been a long time...

Yes, indeed...a pity of course!

Christophe wrote:

t00fri wrote:while I agree of course with your explanations, it might nevertheless be worth remembering that the mere existence of the LT delay is the most basic "relativistic effect" together with the (supposed) universality i.e. frame independence of the finite speed of light. From the finiteness of the latter then follow all the familiar formulae of special Relativity, i.e. length contraction and time dilatation...

I give you that, but you must admit that as currently implemented the LT delay is just a hack (useful and working, but a hack nonetheless).

hmm, a hack is not quite the right characterization, I think. If one of us would have had an idea how to implement the finite speed of light in a manner that is both correct and does not suck the CPU to "zero" fps, we would have clearly incorporated it. My LT formula is fast and could not be more correct, but it does strain the users intuition a little

Christophe wrote:
By the way I'm on vacation at the end of next week, and I really intend to get back to some Celestia hacking this time. The new key mapping code is long overdue now, as is 1.3.2 -- a special Cassini-Huygens Edition would be nice, don't you think?

Did you mean that your vacation will start end of next week?

There are still some buggies/weaknesses that should be fixed before 1.3.2. I think that's also what Chris feels.

I also wrote to you about some missing KDE call-back for some menue entries, but never heard of you.

Bye Fridger

Another limitation is that a computer screen can't come anywhere near reproducing the enormous range of brightnesses of things in space. The star display does a pretty good job of getting magnitude differences across, but the brightnesses of solid-modeled objects are somewhat arbitrary. Since the human eye can adjust to vast brightness variations in real objects (within limits!) and astronomical photos tend to be exposed and adjusted to bring out their subjects in all their glory, Celestia is arguably only doing what you'd do were you looking at the objects in question; but it's good to keep in mind.

...Also, there's the matter of star distances, which are notoriously hard to measure. Most of the stellar distances in Celestia's default catalog come from the parallax measurements taken by the Hipparcos satellite some years back, but astronomers have gotten significantly different distances for some stars by other methods. It's probably best to consider them rough order-of-magnitude estimates for stars that are more than a few light-years away.

Hmmmm, I seem to understand it so far...
I don't mind the computers limitations too much, because the program
is awesome in itself, but hey if it would be possible to fix it or make that,
that would be pretty lifelike . I have some more questions, first, to follow up on my first post - all the moons, sattellites, and spacecraft
are in their pretty much exact locations? Everything is where it really
is right now? So if I had a telescope and I see like the Hubble passes over us, then I could somehow scan the sky (night/day whatever), and actually find it? I will mention why I am asking this in a later post. Also, how do you figure out the distances between two objects, say a meteor and ummm Earth lets say, I haven't been able to figure out any distance.
The distance displayed in the upper left hand corner when you select
something shows, but I don't understand what that distance stands for.
Is it also possible to center on more than 1 object? What are some other kinds of things could I do on the program that aren't listed in the menus?

Thank you all.

t00fri wrote:hmm, a hack is not quite the right characterization, I think. If one of us would have had an idea how to implement the finite speed of light in a manner that is both correct and does not suck the CPU to "zero" fps, we would have clearly incorporated it. My LT formula is fast and could not be more correct, but it does strain the users intuition a little :roll:

Of course, by hack I mean a simple solution to a complex problem ;-)

t00fri wrote:Did you mean that your vacation will start end of next week?

Yes, that's what I mean, although it actually starts tomorrow night, but I'll be in Brittany for a week.

t00fri wrote:There are still some buggies/weaknesses that should be fixed before 1.3.2. I think that's also what Chris feels.

Maybe we shouldn't aim for total perfection.

t00fri wrote:I also wrote to you about some missing KDE call-back for some menue entries, but never heard of you.

I digged that one out, I suppose that's what you're refering to

t00fri wrote:while the individual settings for orbit rendering and labels of spacecrafts and other objects are indeed done in KDE-Celestia's
Settings->configure Celestia -> Orbits/Labels, these settings cannot be saved! Also the callback of the 'Show Markers' is reversed as to the Flush text: when I deactivate the Markers in the dialog, the flush text says 'Markers enabled' and vice versa...

I'll have a look at the orbit settings, but I can't reproduce the Show Marker bug. You're talking about the one in the configure dialog, right?

Xcron wrote:I have some more questions, first, to follow up on my first post - all the moons, sattellites, and spacecraft
are in their pretty much exact locations?

Some objects (the planets and some of the major moons) are defined to follow orbits which are defined by polynomials called "VSOP87 Theory". Their paths are much more accurate than simple ellipses, since gravity and other forces cause orbits to change with time.

Most objects in Celestia are defined to follow Keplerian elliptical orbits. As a result, those orbits are not as accurate as they might be. Elliiptical orbit parameters are only accurate at a specific time, called their Epoch. If you need a precise location for a body, you need to provide elliptical orbit parameters that are valid at that time.

Everything is where it really
is right now?

Pretty much.

So if I had a telescope and I see like the Hubble passes over us, then I could somehow scan the sky (night/day whatever), and actually find it?

You'll be able to find it, but it won't be in exactly the direction that Celestia shows you.

Even when you have the precise elliptical orbit parameters for a satellite at the time you want to observe it, Celestia won't be able to provide exactly the right viewpoint.

Celestia models the Earth as a nearly perfect sphere or oblate spheroid.
The actual shape of the Earth is much more complicated. The position in space of your latitude and longitude on a sphere is not the same as the position in space of that same latitude and longitude on the real Earth.

I will mention why I am asking this in a later post. Also, how do you figure out the distances between two objects, say a meteor and ummm Earth lets say, I haven't been able to figure out any distance.
The distance displayed in the upper left hand corner when you select
something shows, but I don't understand what that distance stands for.

The distance shown by Celestia is the distance of your viewpoint from (approximately) the surface of the selected object. To determine the distance between two objects, you must place your viewpoint at one of them.

Is it also possible to center on more than 1 object?

I'm not sure what you mean by "center" on an object, but you can place your viewpoint relative to one object and cause another object to move to the center of the screen without changing the position of your viewpoint.

What are some other kinds of things could I do on the program that aren't listed in the menus?

You should read Frank Gregorio's manual for Celestia. It's available in several languages. See http://shatters.net/celestia/documentation.html

I hope this helps.

Christophe wrote:

t00fri wrote:hmm, a hack is not quite the right characterization, I think. If one of us would have had an idea how to implement the finite speed of light in a manner that is both correct and does not suck the CPU to "zero" fps, we would have clearly incorporated it. My LT formula is fast and could not be more correct, but it does strain the users intuition a little

Of course, by hack I mean a simple solution to a complex problem

I can easily agree with THAT definition;-)

Look at this glorious "success report":

Chris wrote:Using the light time delay feature, the timing in Celestia was perfect! Just a few more minutes of burn until SOI is complete, then I'm keeping my fingers crossed for the descending crossing of the ring plane.

--Chris

Christophe wrote:

t00fri wrote:Did you mean that your vacation will start end of next week?

Yes, that's what I mean, although it actually starts tomorrow night, but I'll be in Brittany for a week.

Great! I'll be in China (Bejing) for a week, but not for vacations... Curious whether the rental prices for bikes have gone up since my last visit;-)...

Christophe wrote:

t00fri wrote:There are still some buggies/weaknesses that should be fixed before 1.3.2. I think that's also what Chris feels.

Maybe we shouldn't aim for total perfection.

We also don't want to copy the KDE philosophy, do we?;-).

Christophe wrote:

t00fri wrote:I also wrote to you about some missing KDE call-back for some menue entries, but never heard of you.

I digged that one out, I suppose that's what you're refering to

t00fri wrote:while the individual settings for orbit rendering and labels of spacecrafts and other objects are indeed done in KDE-Celestia's
Settings->configure Celestia -> Orbits/Labels, these settings cannot be saved! Also the callback of the 'Show Markers' is reversed as to the Flush text: when I deactivate the Markers in the dialog, the flush text says 'Markers enabled' and vice versa...

I'll have a look at the orbit settings, but I can't reproduce the Show Marker bug. You're talking about the one in the configure dialog, right?

Hmm, I am sure these bugs are real. I bet you will confirm them soon or later;-) It's just not so easy to describe the conditions for their occurrence precisely...

Have a nice holiday,
Bye Fridger

Thank you all for your help.
If I have any more questions, I will be sure to post them up on this forum.
So far, all I have learned has turned out to be very beneficial for me.

Thank you, once again.