Celestia Forums

dirkpitt wrote:Comoving distance is a concept that's supposed to take Hubble expansion into account for distances from an observer... am I correct? Right now, Celestia's objects do not recede from each other and Fridger wants that changed I think.

Hi DW,

at cosmological distances and times, general relativity becomes increasingly important. There are many implications of GR, but the concept of a 'distance' is about the first notion that has to be defined and used properly in concordance with GR in an expanding Universe. In parallel, one also has to implement a definite ansatz for the metric of space-time that unlike special relativity now becomes a /dynamical/ quantity. The most popular metric in a Big-Bang cosmology is the Friedmann-Robertson-Walker metric. It's the solution of Einstein's equations for a homogeneous and isotropic Big-Bang Universe.

In fact there are various differing definitions of a cosmological distance, like the luminosity distance, the angular-diameter distance ...and the comoving distance. The latter involves an integral over the expanding cosmological scale factor a(t), and thus --in turn-- depends in a calculable manner on the particular cosmology under consideration (this includes its geometry: flat| closed | open...and the fractions of luminous and dark matter as well as dark energy. We know these from WMAP and thus everything is ~ fixed).

The comoving distance is the one that is theoretically most appropriate, and actually it's the distance appearing in Hubble's law.

The respective code is in my celx script that displays the redshift of our Celestia galaxies.

People can also see the large differences in distance by using this online cosmology calculator by Ned Wright:

http://www.astro.ucla.edu/%7Ewright/CosmoCalc.html

On the same page, there is also a link to a nice and simple tutorial about cosmology by him.

Once Celestia is to be expanded to include cosmological distances and times, we first got to get the framework right that will be underlying the respective new visualization tasks.

Bye Fridger

chris wrote:I'm also quite fond of Mitaka's screen-aligned square that shows the object scale. It would be a great addition to Celestia, and should be easy to at least prototype with a script.

Chris,

Please, could you have a look at
http://celestiaproject.net/forum/viewtopic.php ... &start=168 ?

Celestia blows mitaka away easy.
Though I do like the galaxies in mitaka and the ort cloud surrounding the solar system.

Just my two cents, but Mitaka cannot hold a candle to Celestia in most regards. Although its few select galaxies and the Milky Way are beautiful, and I love the Oort cloud add-on, it is missing a great deal that makes Celestia so special. This includes:

* Navigation is grossly limited to zoom in/out, and a few pre-selected destinations. To change orientation, you have to drag, drag, drag on the mouse.

* User cannot fly anywhere at will and cannot visit stars or even other galaxies except the few targets in the database.

* The few galaxies that it draws are beautiful, if you can get close to them. You have to drag the mouse around painfully to align your path so you can then zoom in or out and hope a target galaxy passes close to you.

* You cannot change your landing point on Earth. It is stuck at Japan. As such, I cannot see the night sky from my latitude and longitude.

* The size of the stars is crazy. Mitaka seems to use scaled disks to a very unrealistic degree.

* You cannot set time to pass smoothly. You have to keep clicking the + button with your mouse.

* The Milky Way is too bright from the ground. Its haze is too dense to be realistic.

There are lots more things I don't like about it, but ...

Thanks to all the talented folks at Celestia for making Celestia by far the best in existence.

PS - Because of Celestia, Astronomy has become the most popular elective science in my High School in Virginia. A whopping 8 sections of juniors and seniors (about 240) have signed up for Astronomy next term, forcing my school to hire a new teacher just to cover the sections I cannot. I credit my use of Celestia education for that appeal.

Word is also spreading around the world about Celestia's value in education. Based upon feedback from the motherlode site and from emails I have received, it now appears to be in use in hundreds of high schools worldwide, with a particular concentration in Australia and New Zealand.

FYI

Frank

Frank,

I think the case is by far not that easy! You were enumerating a lot of DETAILS about which I of course agree with you.

But the coding pace of Mitaka is considerably faster that that in Celestia for a number of obvious reasons. So DETAILS can easily and quickly be corrected for and polished away...

It's the underlying PLOT that one has to consider. It's the target where they are planning to go.

But so what, I can't help it if none of you is seing what I mean. We'll talk about it later ...When it's too late.

Bye Fridger

I once made an Oort cloud model, ready to be used in Celestia, and very similar to Mitaka's Oort version. It's very convincing in Celestia (fully volumetric model) :



Its only problem is the magnitude which appears to be increasing while the observer is moving farther from it. I already asked Chris to add an option at the SSC (and DSC) level so it can have a magnitude dependant on distance. Until now, he's not very interested.

Compare my model with the one shown in Mitaka (of course, I can easily made a model with a larger thickness and bigger dispersion of dots) :

That still looks cool. Do you have it available for download?

Here's a small addon which places a generic Oort cloud around the solar system. Notice that it disappears abruptly at some distance. This is Celestia's fault and there's nothing I can do about it.

http://nho.ohn.free.fr/celestia/Cham/Oort.zip (408 KB zip file)

Feedbacks would be appreciated.

Cham wrote:Here's a small addon which places a generic Oort cloud around the solar system.... Feedbacks would be appreciated.

Strangely I get a bug with the sun halo using this addon (on old render path); the halo cover all the screen between 1.7280 and 7.0337 ua of distance. Before 1.7280 and after 7.0337, all is normal...

Someone confirm?

ElChristou wrote:

Cham wrote:Here's a small addon which places a generic Oort cloud around the solar system.... Feedbacks would be appreciated.

Strangely I get a bug with the sun halo using this addon (on old render path); the halo cover all the screen between 1.7280 and 7.0337 ua of distance. Before 1.7280 and after 7.0337, all is normal...

Someone confirm?

I don't have your bug. Is that with Dirkpitt's last build or my build (with color corrections) ?

And use that SSC definition, instead of the one in my previous link. The size is better (about 10000 AU from Sol) :

Cham wrote:

ElChristou wrote:

Cham wrote:Here's a small addon which places a generic Oort cloud around the solar system.... Feedbacks would be appreciated.

Strangely I get a bug with the sun halo using this addon (on old render path); the halo cover all the screen between 1.7280 and 7.0337 ua of distance. Before 1.7280 and after 7.0337, all is normal...

Someone confirm?

I don't have your bug. Is that with Dirkpitt's last build or my build (with color corrections) ?

And use that SSC definition, instead of the one in my previous link. The size is better :

Code: Select all

"Oort" "Sol"
{
   Mesh "Oort.cmod"
   Emissive true
   Radius 2000000000000

   EllipticalOrbit {
      Period   9E12
      SemiMajorAxis   0
   }
}



The bug is gone using this new ssc whatever app I use.

Could you redo an asteroid belt but not not a simple belt, this time more a sphere like your Oort cloud, but with a higer density on equator (forming the belt)...

Cham:

I was able to fix the brightening of the Cloud as you back away and approach 1 LY. I simply increased its radius to 5000000000000. Now, it brightens only slightly, then disappears at 1 LY as usual (Celestia bug).

Here is the revised ssc file:



Since many astronomical references estimate the outer edges of the cloud at 1 LY or beyond, the boost in radius is still scientifically accurate.

Thanks for the add-on. It fits perfectly into one of my educational activities.

regards

Frank

fsgregs,

the reason why I used the radius given above (2, instead of 5) is twofold :

I can get a nice global view without having the model to disappear suddenly, and I can also show some comets coming from the Oort cloud and moving inside :

I can only see a red cross instead of the image.
In a public computer,the image showed NOTHING,just empty space.No option to show its true size,too...

Cham wrote:I once made an Oort cloud model, ready to be used in Celestia, and very similar to Mitaka's Oort version. It's very convincing in Celestia (fully volumetric model) :



Its only problem is the magnitude which appears to be increasing while the observer is moving farther from it. I already asked Chris to add an option at the SSC (and DSC) level so it can have a magnitude dependant on distance. Until now, he's not very interested.

Compare my model with the one shown in Mitaka (of course, I can easily made a model with a larger thickness and bigger dispersion of dots) :

Strange.Now I saw...but where??s the Oort Cloud in Celestia?It??s very dark and hard to see in my monitor,at least...

Cham wrote:Here's a small addon which places a generic Oort cloud around the solar system. Notice that it disappears abruptly at some distance. This is Celestia's fault and there's nothing I can do about it.

http://nho.ohn.free.fr/celestia/Cham/Oort.zip (408 KB zip file)

Feedbacks would be appreciated.

I installed it and it looks great so far. Sorry if my feedback isn't helpful but I haven't noticed any problems or bugs with it.

Cham wrote:Notice that it disappears abruptly at some distance. This is Celestia's fault and there's nothing I can do about it.

Cham, this sounds like the "1LY for SSC's" limitation. See here: http://celestiaproject.net/forum/viewtopic.php ... highlight=

Cham wrote:

2- A set of custom classes "layers", that we could turn ON/OFF on the fly using the keyboard (ctrl-1, ctrl-2, ctrl-3, or command-1, command-2, .. on the Mac). We could associate many Selden like coordinates grids to these "layers", for example, magnetic field lines and other information.

I totally agree with this. If these custom classes are not intended as screen layer only, these could permit also some recursive switches among the meshes, useful for the addon developers who needs to "disjoint" specific particularities without hiding a quite lot of models for then re-showing it; because the most external model, or invisible "container"'s mesh, ever affect the verbose description.

Fridger wrote:

There is no screendump facility (yet) . Otherwise I had of course done it right away.


Fridger, try this:

http://www.gadwin.com/printscreen/

The not Pro version is freeware. It work by press the Printscreen key, so you do not have need to do with the taskswitch.

Cham, is it possible to use the same methods to model star streams around Milky Way? I think it could be really interesting to make the model of Milky Way in Celestia as accurate as possible. What we have now, is the main structure of stars in the Galaxy. But Milky Way is a product of many collisions between galaxies and its real structure is much more complicated. Please check this link:

http://www.astro.caltech.edu/~drlaw/MWstreams.html

Maybe it's possible to recreate these structures in Celestia in a similar way you did with Oort cloud?

s3nn0c wrote:Cham, is it possible to use the same methods to model star streams around Milky Way? I think it could be really interesting to make the model of Milky Way in Celestia as accurate as possible. What we have now, is the main structure of stars in the Galaxy. But Milky Way is a product of many collisions between galaxies and its real structure is much more complicated. Please check this link:

http://www.astro.caltech.edu/~drlaw/MWstreams.html

Maybe it's possible to recreate these structures in Celestia in a similar way you did with Oort cloud?

There's almost no limits on what I can do with Mathematica and lines and dots CMOD models. I was also planning all sorts of nebulas, dust clouds, accretion disks, magnetic fields, etc, using this technique. Unfortunately, there are some problems related to the luminosity of the model, as seen at a large distance, and Chris isn't really interested in supporting them

Chris wants to concentrate on sprites (which aren't working on the Mac, and which aren't working on all computers without OpenGL 2). I think the dots models I'm able to do (without sprites) have LOTS of potential, but I currently don't want to invest much more time on them because of the luminosity problem.