New Galaxies

[FriQenstein]

Hello all (again),

You know how you can zoom out fully from the MilkyWay and see some of the other galaxies in the background?
Is it possible to create a new galaxy that you can edit and zoom into like the MilkyWay default?
So say you want to make a demo of a travel from our sun, thru our solar system, and then all the way out of our galaxy, and then soar over to M81 and down into a new fictional solarsystem?

Or perhaps even be able to zoom into different parts of the MilkyWay? Say the "other side" or journey to the center?

I supposed one would need to create an entirely new starmap for the remaining unmapped portion of the galaxy before anything else eh?
Or would it require a complete re-write of the program code first?

Just curious.
Tnx

[selden]

The current version of Celestia cannot reliably display stars that are more than about 16KLY from the sun. Please read the "Preliminary User's FAQ", which is a sticky near the top of the Celestia Users Forum, especially Q/A #12.

[ajtribick]

Unfortunately the code used in Celestia only allows stars out to a distance of 16,000 light years, which isn't far enough to populate the whole of our galaxy with stars, let alone other galaxies. Beyond this distance, the floating-point format used is too inaccurate to represent star positions, and distances to such distant objects have large errors associated with them anyway.

There have been proposed workarounds to this, but (as far as I am aware) no code for it. There are several issues to sort out before the 16,000 light year limit is overcome.

You can create new galaxies by putting a .dsc file in extras (take a look at the deepsky.dsc file for an idea of the syntax). You won't be able to add star systems to them though.

[FriQenstein]

Doh !!
I was reading thru the FAQ and lots of other threads while posting and came to my answer soon after... I know I know... launch me at the sun now I guess.

But thanks for the reply none-the-less.

[Reiko]

At first I didn't even know I could fly out of the galaxy and then I did it by accident.

It would be fun if it was possible to have solar systems in other galaxies.

[Starshipwright]

Could Celestia be made to "emulate" a different galaxy by using different data files? If you replaced Stars.dat, solarsys.ssc, etc.? Why any one would want to do this I can't think, but could it be done?

[Chuft-Captain]

I think this code which I added as an experiment to nearstars.stc demonstrates the accuracy problem described by chaos_syndrome:

Code: Select all

"Chuft-Captains Star"
{
      RA            0.1214
      Dec          27.6786
      Distance   3.376e+08  # method: T-F
   SpectralType "G2V"
   AbsMag 4.83
   RotationPeriod      609.12  # 25.38 days
}



This is using the Right Ascension, Declination, and Distance copied from the Galaxy NGC 2, but ends up several Mparsecs away, instead of at the center of NGC 2.

[Reiko]

So when you place a galaxy that far out it's off place too?

[Chuft-Captain]

So when you place a galaxy that far out it's off place too?

Actually, I'm not entirely sure. This is the only time I have tried this (prompted by the speculation about extra-galactic solar systems).
From what I understand Celestia's precision is sub-arcsecond out to about 16,000 ly from the Sun (which is still well within our own galaxy).

The amount of error I get in this experiment could be due to a mistake by me, or a difference between the way coordinates are handled in nearstars.stc and deepsky.dsc.

See this comment by Chris, the original creator of Celestia: http://celestiaproject.net/forum/viewtopic.php ... ght=#78831

I'm not an expert on these "distant" matters as mostly what I do is model spacecraft and O'Neill style colonies/habitats in our own solar-system. This only involves distances in metres and kilometres rather than mega-parsecs.

[Reiko]

I think this code which I added as an experiment to nearstars.stc demonstrates the accuracy problem described by chaos_syndrome:

Code: Select all

"Chuft-Captains Star"
{
      RA            0.1214
      Dec          27.6786
      Distance   3.376e+08  # method: T-F
   SpectralType "G2V"
   AbsMag 4.83
   RotationPeriod      609.12  # 25.38 days
}



This is using the Right Ascension, Declination, and Distance copied from the Galaxy NGC 2, but ends up several Mparsecs away, instead of at the center of NGC 2.

I added this into the deepsky.dsc and labeled it as a galaxy. It put it right where it was supposed to be but instead of being a star it became a tiny galaxy 1 light year wide. lol

[selden]

I'm not quite sure why someone didn't put this together before and make a little noise about it...

Perhaps some "minor" modifications to the DSC Galaxy code would allow the placement of bodies that look like stars at arbitrary distances. A problem is that there seem to be quite a few assumptions built into the code that would make it rather awkward. For example, one wouldn't be able to orbit planets around these objects. But one can't have everything. Yet.

I think the minimal set of changes would include

+ Define a Type which draws a single opaque, glowing sphere instead of random fuzzy blobs
+ Allow Radius to be smaller than 1LY
+ Allow a Color to be specified.

Doing it right would include adding the code necessary to interpret all of the STC parameters, of course, which would be a larger project for another time.

Fridger,

What do you think of this perversion?

[t00fri]

I'm not quite sure why someone didn't put this together before and make a little noise about it...

Perhaps some "minor" modifications to the DSC Galaxy code would allow the placement of bodies that look like stars at arbitrary distances. A problem is that there seem to be quite a few assumptions built into the code that would make it rather awkward. For example, one wouldn't be able to orbit planets around these objects. But one can't have everything. Yet.

I think the minimal set of changes would include

+ Define a Type which draws a single opaque, glowing sphere instead of random fuzzy blobs
+ Allow Radius to be smaller than 1LY
+ Allow a Color to be specified.

Doing it right would include adding the code necessary to interpret all of the STC parameters, of course, which would be a larger project for another time.

Fridger,

What do you think of this perversion?

Selden,

honestly not all that much, at least for now...

Well, the reason is that such "perversions"( <-Selden ) will necessarily concern individual attachments of objects in galaxies, while the clearcut underlying philosophy of the DSO project is MASS rendering of thousands of deepsky objects without human interference! I.e. all data read in from scientific published catalogs

Second, there is a clearcut workflow ahead, namely the mass rendering of clusters (globular and open ones), again directly from catalog data.

Third, the changes are way more than "trivial" and thus need detailed re-testing of many galaxies, the rendering of which might get (erroneously) affected by such changes. As you suspected, the present galaxy code is pretty much in a "complex balance" so to speak...


On the other hand, such experiments are fun, of course. So I would definitely encourage people to go on trying. Perhaps eventually something will be coming out of it that everyone wants to have How about using a Lua hook for this, just for trying? The underlying philosophy is anyhow much closer to our general add-on philosophy.

Bye Fridger

[heath_rezabek]

This may be a strange place to say so, but as a new and awestruck enthusiast of Celestia, I for one adore the glowey-blob galaxies. I distinctly remember the first time I saw this method, a few years ago in a digital planetarium, and I thought it was brilliant, beautiful, elegant and actually quite 'realistic'-seeming.

Given our own meager limits, i see some poetic justice in the fact that we cannot (yet) reliably detail objects that far out. Celestia still allows us to reach far further than we otherwise could... but not to the edges of infinity. Not just yet.

Feeling thankful to have found Celestia before 2007,

- Heath, Austin TX

[Reiko]

I placed two of them into the center of the small magellanic cloud and this is what it looked like

[Vincent]

On the other hand, such experiments are fun, of course. So I would definitely encourage people to go on trying. Perhaps eventually something will be coming out of it that everyone wants to have How about using a Lua hook for this, just for trying? The underlying philosophy is anyhow much closer to our general add-on philosophy.

I agree with Fridger, Lua is the way to go for this kind of experiment. I've been also thinking about adding a Lua Hook that would allow users to define new custom classes/types of objects. That would also be useful for Cham's magnetic field lines, for example... Defining keyboard shortcuts to toggle the rendering of these classes would be very easy using Lua...

[Reiko]

This is interesting. I took a G6 star that was about 90 lys away and had planets, changed the distance to 64.384Kpc and it expanded!
The star at 90 lys was 1.41 Rsun and at the extragalactic distance it expanded to 3322.20 Rsun. All of the planets that were orbiting it are now inside the star.
Not only that, you can only see the star if you look at it with the Milky Way in the background.





If you pan the camera to where you can't see the Milky Way, the star can't be seen.




I know you can't accurately place stars beyond 15,000 lys but I didn't expect them to expand to monsters when you did.

[selden]

My guess would be that the STC file contains an AppMag declaration. That specifies a star's apparent magnitude: how bright it appears to us when we look at it from here on the Earth.

In contrast, an AbsMag declaration would define a star's intrinsic brightness: it specifies how bright a star would be at a distance of 10 parsecs.

If two stars have the same apparent magnitude and one is more distant than the other, then the more distant star must have a greater intrinsic brightness (and be larger) than the nearby star.

[Reiko]

My guess would be that the STC file contains an AppMag declaration. That specifies a star's apparent magnitude: how bright it appears to us when we look at it from here on the Earth.

In contrast, an AbsMag declaration would define a star's intrinsic brightness: it specifies how bright a star would be at a distance of 10 parsecs.

If two stars have the same apparent magnitude and one is more distant than the other, then the more distant star must have a greater intrinsic brightness (and be larger) than the nearby star.

Yes it did have an AbsMag, I changed it around and you are right in that is what made it larger. But why does the star, and planets once I set them far enough to be outside the star, not show up? The markers are there but you cannot see the planets.

[selden]

Celestia can only draw stars and systems reliably if they're within 16KLY of the Sun. Chris is hoping to rewrite the code so that stars can be defined at larger distances, but this won't happen soon. For details, please read the Preliminary User's FAQ, Q/A # 12

[Reiko]

Celestia can only draw stars and systems reliably if they're within 16KLY of the Sun. Chris is hoping to rewrite the code so that stars can be defined at larger distances, but this won't happen soon. For details, please read the Preliminary User's FAQ, Q/A # 12

Oh sorry, I thought it could draw them at a distance but not place them accurately. My misunderstanding.