Celestia Forums

While the version without emissivity crashed my Mac (total freeze of OS X !), the version with emissivity appears to be working well (? I'm not even sure of that yet !). The rendering is almost right (??), except that there isn't any lumpy shadows inside the dusty ring.

Preview (click for a large picture) :


Here's the model and its SSC, to play with it :


The model has 10 000 points in it.

I can generate very easily any kind of models like this, with Mathematica. Too bad it crashes the Mac when emissivity is OFF

I changed the picture used for the sprites. Adding some noise may help a bit, but we can feel a periodic pattern in the ring, because of the noise repeating itself on each small blob. Maybe we should have an option to randomize the picture orientation (Chris ?) :



Here's the texture used for this version :

Here's another experiment :



There is differential rotation, in this case.

First time I load point sprites in Celestia via my brand new config! It works well but I don't have the same result:

ElChristou wrote:First time I load point sprites in Celestia via my brand new config! It works well but I don't have the same result:

Image 1.jpg

Maybe this is because I just changed the opacity in the model. Try this in the CMOD file :

Nope, I still don't have the difference in your shot between the ring in front and behind the planet. My "ring" is quite uniform, the opacity change but all over the ring...

Okay, try this version. Looks nice in Celestia (but ...) :

and add the following model into the models folder :


This version is associated with the two last pictures given above (my previous post with pictures).

Same; I have the two rings, but they are uniform... I guess it's an hardware symptom...

Try this URL :

cel://Follow/Sol:Jupiter/2008-10-05T23: ... 2048&ver=2

From this place, it's ugly :
cel://Follow/Sol:Jupiter/2008-10-06T01: ... 2048&ver=2

I feel that the sprites code isn't very good for this kind of stuff (accretion disks). The disk is becoming too opaque at a distance. Same thing as for the dust rings without sprites (CMOD dots only). Chris ?

Nope, I will have to try on a fresh build... (a bit later)

It's pretty!

It makes my computer cry though. Frame rate plummets to 1 or 2 fps when I'm close to it.

Yes, the distance fading problem is annoying... pity, because it would be possible to make a really neat Beta Pictoris add-on with this kind of technique.

Do someone has tried one point sprites' nebulae? It's fine. Nevertheless, there are pros and cons.
Pros:
Nebulae are seen in the same portrait from everywhere (right toggle doesn't change the view) so one is sure that the shape is that and no more "line" billboard viewed from different angle.

Cons; no one real nebulae rest in the same position when turned around...

I made a new CMOD nebula model with sprites. It's working well in Celestia, but there's a strong problem, as shown on the second picture below.

The planetary nebula prototype at close range. Looks pretty, in full 3D :


The problem is shown here at large scale. There are many small red dots visible in the Milky Way, which is a bad thing. So the sprites code isn't well implemented in Celestia :


Here's the model for you to test, with Selden's old planetary nebulae (modified) addon :

Modifying a simple parameter in the CMOD file gives some more convincing model :


However, there is still the problem with luminosity at large range (and sprite size with distance) :


Please Chris...

EDIT : Here's a variation with two colors :

Nice nebula experiment (the only problem is the sprites model high luminosity at large distance ) :

Another nice experiment with sprites (click the thumbnail for a larger picture) :

Yet aother sprites experiment. Dark matter halo :


This spherical shell model behaves very well in Celestia, even at large distance.

If you want feedback, better post the models, else...

ElChristou wrote:If you want feedback, better post the models, else...

There's a test addon, on the previous page!