New accretion disk method (WORK IN PROGRESS)

[Cham]

I think all this could give a lot more realism.

I don't think that it would be more realist with just sprites, especially since there isn't any drop shadow on sprites.

In the case of Saturn, it would be way too complicated to do. So it is clear to me that I will not do something for Saturn.

However, I could do it for something simpler : an exoplanet with a fictious ring system.

[Goofy]

I think all this could give a lot more realism.

I don't think that it would be more realist with just sprites, especially since there isn't any drop shadow on sprites. In the case of Saturn, it would be way too complicated to do. So it is clear to me that I will not do something for Saturn. However, I could do it for something simpler : an exoplanet with a fictious ring system.

OK Cham, never mind, and thank you anyway.
It's a matter of opinions, mine is different from yours because I think that Saturn deserves any attempt to improve it, so I'll give a try, even if I don't use MathLab, and this will take me a lot of time.
Obviously I'll inform here about results, if any.
Bye

Goofy

[Cham]

I just made a quick Mathematica run for 250 000 sprites !

I'm very surprised that there isn't any perceptible impact on the frame rate !

Sprites are now smaller (diameter of approximately 6 km). From a distance, they are not visible at all, so no impact on the usual rendering, even in the shadow (the sprites are now all emissive, to stay white everywhere). See the pictures below :

sat1.jpg

sat2.jpg

sat3.jpg

As you can see, even with 250 000 sprites, the density is very low (unrealistic at all !)

The CMOD file has a size of 14.7MB. So a million of sprites (4x more) would take almost 60MB, and the density would still be very low.

[Cham]

Three more views of the 250 000 sprites in Saturn's rings :

sat4.jpg

I turned the stars OFF, so we can better see the sprites :

sat5.jpg

sat6.jpg

[Goofy]

Wow Cham, this is very impressive.
Just to help me save time, can I please try your last Saturn 250,000 sprites cmod file, or the 1,000,000 sprites, if you'll produce it?
Thanks a lot once again.

Goofy

[Cham]

Prometheus and Pan are invaded of little guys, hehehee !

prom.jpg

pan1.jpg

pan2.jpg

So you get an idea of the size of the sprites (about 6km)...

[Cham]

I'm pushing this too far. Here's what I get with ... 1 million sprites !

I'm beginning to have a frame rate impact...

satu1.jpg

satu2.jpg

satu3.jpg

This rendering isn't realist at all ! So no more for Saturn and I move to a simpler, fictious ring system...

[Cham]

Now something more alien...

Let me introduce Volcanius, a protoplanet in Beta Pictoris.
200 000 sprites in its ring system...

vul1.jpg

vul2.jpg

vul3.jpg

[Cham]

The effect becomes interesting only at very high densities :

volc1.jpg

volc2.jpg

[Cham]

I polished the rings model for this planet. 800 000 sprites in there, around a planet of about the size of Earth.

part1.jpg

part2.jpg

I will not publish that part, because the addon is too big (large virtual texture on the planet, which I didn't made anyway).

Ring models like this one aren't really interesting, because each model is specific to a particular ring dimension. The same model cannot be used on another ring system, if the dimensions aren't exactly the same. So I will not pursue this, it's simply not worth it (especially for Saturn). Sorry Goofy.

[Cham]

For the ones who care about the particles in Saturn's rings, here's the crude model I made. To me, this isn't interesting at all.

The CMOD file contains 500 000 particles, duplicated in the SSC, so there's 1 million of particles in motion in Saturn's rings. You can add more if you want, by using the small trick I exposed in the SSC file.

Enjoy, or trash !

http://fsgregs.startlogic.com/Public_Fi ... _Rings.zip (6.9 MB zip file)

[Goofy]

For the ones who care about the particles in Saturn's rings, here's the crude model I made. To me, this isn't interesting at all.
The CMOD file contains 500 000 particles, duplicated in the SSC, so there's 1 million of particles in motion in Saturn's rings. You can add more if you want, by using the small trick I exposed in the SSC file.
Enjoy, or trash !

I'm enjoying, thanks a lot.
Bye

Goofy

[Hungry4info]

Thank-you for this download. I look forward to tinkering with it.

[Goofy]

Hi.
Same problem, but I modified the cmod file (following Cham's exhaustive instructions) with
Emissive false (instead of "true")
Color 0.6 0.6 0.6 (instead of 1 1 1)
It works. in my opinion.
Bye

Goofy

[Cham]

Oops !

Sorry guys, yes Goofy ; the Emissive directive should be set to false in the SSC file :

Code: Select all

Emissive false

or just comment it out like this :

Code: Select all

#Emissive true


I missed that. Strangely, Celestia 1.7.0 (the beta version I'm working with these days) doesn't do anything if I use Emissive true. Hmmm, maybe another bug with it

So just leave the CMOD file as it is (don't edit it), and just remove the Emissive line in the SSC file instead. You'll then have the same rendering as me.

[PlutonianEmpire]

One thing I was wondering, would it be possible to use or create sprites models for use as clouds? As in, weather clouds? Not necessarily for the whole planet of course, as framerates would probably take a huge hit, but rather, for surface scenes at specific locations on a world?

[Cham]

One thing I was wondering, would it be possible to use or create sprites models for use as clouds? As in, weather clouds? Not necessarily for the whole planet of course, as framerates would probably take a huge hit, but rather, for surface scenes at specific locations on a world?

Sprites could certainly be used for this. However, I think it would be difficult to achieve a realistic effect, to simulate some real clouds.

Before doing weather clouds, I'm much more interested in simulating the aurora on Earth (red and green blobs).
I'll have to use a special CELX script to define the aurora motion (I know how to do it). Currently, I don't have much time left for this project. Maybe this summer.

For the moment, I'm more interested in doing vast "primitive" Orion-like nebulae to be placed in other galaxies (I'm still working on NGC 404 right now). The absence of any symmetry (in primitive giant clouds) makes the project difficult. It was much easier for planetary nebulae and accretion disks, because of the spherical or axial symmetry. I'm searching for another way of doing it, especially to sculpt huge pillars, randomly oriented columns of cold clouds, etc...

[danielj]

But in the ssc, don?t have the term Color.Is it possible to add it? Where?

[selden]

This makes the Earth bright orange:

color.ssc

Code: Select all

Modify "Earth" "Sol" {
  BlendTexture true
  Color [ 1 0.5 0 ]
}