Celestia Forums

Earlier this week, I was attempting to render Venus' atmosphere using the OpenGL 2.0 atmosphere code, and I made it using trial and error. I tried to replicate what I could remember about the venus surface pictures I saw a while back.

This is the result:


Venus with clouds.


Venus without clouds.


Venus surface with clouds.


Venus surface without clouds.

This is the code.


Of course, input is appreciated, and I'm sure there could be improvements made to this code. I'm open to any changes that might need to be made.

Don't suppose you or someone else might summarize for me what the different variables in atmosphere colors mean?

I was just about to ask the same thing.

A kinda related question: Why are the values so small for almost all atmospheres?

Have a look in our Celestial Matters site.
http://forum.celestialmatters.org/

There is e.g. a Mie theory workshop related to Celestia and beyond with ~22000 hits of the thread and a knowledgable host

Manifestations of Atmospheric Light Scattering (Mie Theory)
http://forum.celestialmatters.org/viewforum.php?f=9
http://forum.celestialmatters.org/viewt ... 6&start=75

Topics include playing with an improved Mie theory simulation tool written by Chris L., reproducing the blue Mars sunset or optimizing Earth's atmosphere display etc


To understand Mie theory, you have to know some physics, however....

You may also use the shatters.net search engine for locating many of my respective contributions here. E.g how to set up the subtle atmosphere of Titan...

Fridger

Thank you.

t00fri wrote:optimizing Earth's atmosphere display

Earth's atmosphere with the new Mie and Rayleigh settings looks like it's 5-7 PM when it's noon, there's so much redsky decay (that's what we call it over at Planetside, a Terragen community). Yet it looks (almost) photorealistic from space. It needs to be fixed so it looks as good from the ground as it does from orbit.

Terragen 2 is awesome! If only Terragen planet models could be used in Celestia somehow...

W0RLDBUILDER wrote:

t00fri wrote:optimizing Earth's atmosphere display

Earth's atmosphere with the new Mie and Rayleigh settings looks like it's 5-7 PM when it's noon, there's so much redsky decay (that's what we call it over at Planetside, a Terragen community). Yet it looks (almost) photorealistic from space. It needs to be fixed so it looks as good from the ground as it does from orbit.

The present implementation of Mie scattering in Celestia is substantially simplified (e.g. for reasons of speed), and is thus reliably applicable only for space-bound views. By design, ground-based views are not a main aspect of Celestia. For a significant improvement, the atmosphere and surface rendering code would have to be largely rewritten and huge amounts of harddisk space would be necessary for the required scientific elevation maps.

In this context, let me also refer you to an old thread of mine where a persistent problem wrto atmospheres and surface color rendering is discussed:

viewtopic.php?f=10&t=10611
viewtopic.php?f=5&t=12534&start=13

There I emphasized also the limitations of Celestia's Mie implementation wrto surface-based views. I hope these examples will make clear that these remaining issues are not just a matter of "fixing". Rather, it would all require an atmosphere recoding largely from scratch...

Just to recall for Celestia newcomers: The official Celestia distribution is entirely based on scientific astrophysical data (rather than on fiction), and there are only VERY few celestial objects, where sufficient data are available for a realistic surface-based simulation. All add-ons (e.g. fictional ones) are user projects and hence NOT in the responsability of the Celestia development team.

Note that the Mie stuff only works appropriately with the OpenGL 2.0 rendering path.

Fridger

t00fri wrote:The official Celestia distribution is entirely based on scientific astrophysical data (rather than on fiction), and there are only VERY few celestial objects, where sufficient data are available for a realistic surface-based simulation. All add-ons (e.g. fictional ones) are user projects and hence NOT in the responsibility of the Celestia development team.

So you're saying:
1. Scientific realism over aesthetic realism.
2. Anything based on more than pure data and photographs is fictional.
3. Addons can't be included in an official Celestia distribution, no matter how much it enhances realism.
Also, I fixed your spelling error.

--- edit ---

As to the ongoing effort of improving Celestia's atmospheres, please, have a look into this new thread:
viewtopic.php?f=10&t=15965&start=0

W0RLDBUILDER wrote:

t00fri wrote:The official Celestia distribution is entirely based on scientific astrophysical data (rather than on fiction), and there are only VERY few celestial objects, where sufficient data are available for a realistic surface-based simulation. All add-ons (e.g. fictional ones) are user projects and hence NOT in the responsibility of the Celestia development team.

So you're saying:
1. Scientific realism over aesthetic realism.
2. Anything based on more than pure data and photographs is fictional.
3. Addons can't be included in an official Celestia distribution, no matter how much it enhances realism.
Also, I fixed your spelling error.

Celestia has been designed by it's authors (including myself) to represent a scientific-level 3D visualization of the Universe.

Yes, the task is to achieve scientific level standards in Celestia's positional accuracy, 3D graphics, data base and documentation. Celestia is part of various research projects, notably the STA project by ESA and a number of partner Universities, where scientific realism is required.

http://sta.estec.esa.int/Space_Trajecto ... /Home.html

However due to Celestia's highly modular design, the software can be and actually is used for many other purposes: Educational, planetariums, geology, SciFi, worldbuilding, astrology ... The nature of add-ons by Celestia users is highly diverse and their number is strongly growing.

(1) "aesthetic realism" is poorly defined.

(2) Astrodynamics/astromechanics is a highly quantitative field of theoretical physics. Therefore besides using data catalogs and photos, we can simply calculate many aspects in Celestia to highest accuracy from underlying physical principles. That perhaps most important part you forgot.

(3) Good ideas are always welcome! They may be implemented into the official Celestia distribution via proper C++ coding or scripting (Lua). Often the quality of external add-ons by users is quite poor compared to the pro-level standards we apply for the official distribution. In the "extras-standard" directory we have included a very limited number of standard space missions: cassini, galileo, hubble, iss, mir. You might call these "official add-ons".

Fridger

t00fri wrote:Celestia has been designed by it's authors (including myself) to represent a scientific-level 3D visualization of the Universe.
Fridger

And this devotion to the "scientific-level 3D visualization of the Universe" is why Celestia
is so popular in my opinion. Keep at it there Good Doctor. Some of us really do appreciate your efforts here.

Thanks, Brain-Dead

BobHegwood wrote:

t00fri wrote:Celestia has been designed by it's authors (including myself) to represent a scientific-level 3D visualization of the Universe.
Fridger

And this devotion to the "scientific-level 3D visualization of the Universe" is why Celestia
is so popular in my opinion. Keep at it there Good Doctor. Some of us really do appreciate your efforts here.

Thanks, Brain-Dead

Hi Bob,

good to read you from time to time!

F.

BobHegwood wrote:Some of us really do appreciate your efforts here.

Absolutely!!