animated gas giants?

The place to discuss creating, porting and modifying Celestia's source code.
Topic author
Sum0
Posts: 273
Joined: 10.03.2002
With us: 22 years 8 months
Location: Norwich, UK

animated gas giants?

Post #1by Sum0 » 03.04.2002, 13:22

Recently there's been a cool hack for Celestia that makes the clouds spin round Venus in 4 days while the actual planet goes much slower - as in real life.
Couldn't you make animated gas giants in this way? Ages ago there was a discussion on this topic. What I think you could do is make the texture of the gas giant the usual texture, and have it spin however fast you like. But, very close to the planets surface, you would have a cloud layer, textured with part of that planets texture, and the rest transparent. This would spin at a different speed to the "actual" planet.
On Jupiter, the bands might be the cloud layer, over the normal planet texture. As the planet spins the bands would spin at a different speed, creating an illusion of movement between the bands (cloud layer) and the rest of the planet (normal texture). Not that I know much about the weather systems of Jupiter... Does this make any sense? I hope so. I'll trying making an example gas giant.
"I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able to rightly apprehend the kind of confusion of ideas that could provoke such a question."

Guest

Cluods on Gas Giant

Post #2by Guest » 05.04.2002, 19:23

After modified the Venus data, I thoutght to do something similar with Jupiter (but so far I hadn't the time to work on it), as you suggested.

As you know, to study the motion of the cluods on Jupiter we use two different coordinate systems, spinning at two different speeds (called System I and System II, almost in Italian): the trick is to make a texture rotate at System I speed and the other at System II speed.
One would be setted as a ground map (modifying the speed of rotation of the entire planet, if necessary), and the other as a cloud texture that rotate with a definite speed and an altitude of few km.
After I woould add a thin Atmosphere, as for Earth, Venus and Mars (it's a gas giant, after all!).

It would be more realistic and more funny if we could add more than one cloud texture i.e. another cloud layer above or under the actual one, so we can make clouds spinning in a more continous way, at a differential velocity... Chris, it's possible?
(a second texture layer on planets could be useful also for another reason: we could place on the map of a planet transparent textures with the names of planet's more important formations...)

A last thing: it's possible to set textures on the planet so the aspect is as I expect from calculations or real observations, such as transits of the Great Red Spot on the meridian of Jupiter? So Celestia could became useful also to preview our telescopic observations...
There's some work to do...
Regards
Anarion

Matt McIrvin
Posts: 312
Joined: 04.03.2002
With us: 22 years 8 months

Cluods on Gas Giant

Post #3by Matt McIrvin » 06.04.2002, 03:38

Anonymous wrote:After I woould add a thin Atmosphere, as for Earth, Venus and Mars (it's a gas giant, after all!).


If I recall correctly from the last time I tried this, in 1.2.2, at least, there is not proper support for atmospheres with oblate planets. So you might have to make the planet spherical for now.

Topic author
Sum0
Posts: 273
Joined: 10.03.2002
With us: 22 years 8 months
Location: Norwich, UK

Post #4by Sum0 » 06.04.2002, 13:07

I have done a test on a gas giant around Alpha Centurai A, which should appear on Bruckner's add-on page soon. It doesn't look very good, and I just copied the planet data from some extrasolar planet, but it is cool to watch (especially if you add a moon and watch from there...)
"I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able to rightly apprehend the kind of confusion of ideas that could provoke such a question."

Matt McIrvin
Posts: 312
Joined: 04.03.2002
With us: 22 years 8 months

Cluods on Gas Giant

Post #5by Matt McIrvin » 06.04.2002, 17:43

Anonymous wrote:It would be more realistic and more funny if we could add more than one cloud texture i.e. another cloud layer above or under the actual one, so we can make clouds spinning in a more continous way, at a differential velocity... Chris, it's possible?
(a second texture layer on planets could be useful also for another reason: we could place on the map of a planet transparent textures with the names of planet's more important formations...)


Currently the sky color seen from the ground is modeled as a background color, which has the disadvantage that it appears behind astronomical objects, such as the Moon (whose night side therefore appears black when seen from Earth in the daytime, when it should be blue). An additional cloud layer could be used as a transparent "roof" to simulate the blue sky more accurately. (This might not be the most efficient way to do it.)

Also, of course, real planets have multiple cloud layers.


[/i]

Topic author
Sum0
Posts: 273
Joined: 10.03.2002
With us: 22 years 8 months
Location: Norwich, UK

Post #6by Sum0 » 07.04.2002, 09:18

You can get the moving cloud add-on here (thanks Bruckner)
"I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able to rightly apprehend the kind of confusion of ideas that could provoke such a question."

Mikeydude750
Posts: 169
Joined: 31.01.2002
With us: 22 years 9 months
Location: Wisconsin

Post #7by Mikeydude750 » 08.04.2002, 11:21

Or, why not just make the core a few thousand miles in diameter, and make the atmosphere like it really is, super-thick near the core and lighter as you get higher up. Of course, you should be able to turn that off, if it becomes annoying.

Ortolan
Posts: 120
Joined: 31.03.2002
With us: 22 years 7 months
Location: Melbourne, Australia

Post #8by Ortolan » 09.04.2002, 09:24

I made a new Neptune texture and cloud layer for it to see what an 'animated' gas giant looks like. Looks pretty good at timex1000, I might try Jupiter next. You can download them from my website if you want:
http://www.skullcave.com/neptune.jpg
http://www.skullcave.com/neptune-clouds.png
Just edit your solarsys.ssc file to include atmosphere details for Neptune (eg. copy them from the Earth section and change the name of the cloud texture), I made the cloud height and speed faster than Earth's.
I still reckon fully-animated textures for the giants would kick ass! : )

Matt McIrvin
Posts: 312
Joined: 04.03.2002
With us: 22 years 8 months

Animated Jupiter

Post #9by Matt McIrvin » 10.04.2002, 04:06

I'm working on an "animated" two-level Jupiter texture of my own; it looks pretty good already, and I'll probably write some instructions, zip it up, and send it to Bruckner soon.

The basic idea is to use a cloud texture for Jupiter that is identical to the "ground" texture, but with an alpha channel that is a contrast-enhanced copy of the cloud brightness. Since the bright clouds on Jupiter tend to be the higher ones, this should, at least in theory, give OK results. Give the cloud map a nonzero CloudSpeed, and the lighter clouds race around the planet ahead of the dark ones.

The big problem for Jupiter is how to deal with the Red Spot. The technique above will cause high clouds to obscure part of the Red Spot most of the time, and a transparent, ghostly Pink Spot will move around the planet-- this is not what we want. So some cheat is needed.

Of course it's impossible to animate the Red Spot accurately; so I essentially carved out a transparent trench in the alpha channel so that the high clouds will at least leave it undisturbed.

Ortolan
Posts: 120
Joined: 31.03.2002
With us: 22 years 7 months
Location: Melbourne, Australia

Post #10by Ortolan » 10.04.2002, 05:10

The big problem for Jupiter is how to deal with the Red Spot. The technique above will cause high clouds to obscure part of the Red Spot most of the time, and a transparent, ghostly Pink Spot will move around the planet-- this is not what we want. So some cheat is needed.

Easiest way of fixing that is to have the red spot on the cloud texture at the correct contrast and everything, and remove the spot from the ground texture.
I've been toying with the idea of a hack that would enable the red spot to actually 'swirl'. Essentially, you code a synchronous orbit for a smaller planet that resides INSIDE Jupiter but closer to the sun so that a small circle of its surface breaches Jupiter's surface and is visible. You then tell the small planet to spin on its Z axis ala Uranus so that the visible spot actually rotates! I have no idea if the celestia engine will allow this but I intend to find out with a few tedious hours of trial and error.. ; )

Ortolan
Posts: 120
Joined: 31.03.2002
With us: 22 years 7 months
Location: Melbourne, Australia

Post #11by Ortolan » 10.04.2002, 07:05

D'oh! The celestia engine doesn't react well to planets & moons in close proximity. Ah well, no swirling red spot for now... : )

chris
Site Admin
Posts: 4211
Joined: 28.01.2002
With us: 22 years 9 months
Location: Seattle, Washington, USA

Post #12by chris » 10.04.2002, 16:43

Ortolan wrote:D'oh! The celestia engine doesn't react well to planets & moons in close proximity. Ah well, no swirling red spot for now... : )

What happened when you tried it? It should work fine, though there might be some occlusion errors. And eclipse shadows might be a little weird . . .

--Chris

Ortolan
Posts: 120
Joined: 31.03.2002
With us: 22 years 7 months
Location: Melbourne, Australia

Post #13by Ortolan » 11.04.2002, 10:36

Well I added an Earth-sized moon in orbit around Jupiter so close that 90% of the moon was inside Jupiter. It rendered correctly side-on until the moon passed in front of Jupiter at which point the entire moon was drawn instead of only the small circle of surface that should be visible. I next tried making the moon a smaller flat disc with the same aspect as the red spot, plus extended its orbit past Jupiter's surface. That rendered even worse - the disc would disappear as it began orbiting in front then reappear, then disappear again. (wishing there was a 'capture avi' option)

Rassilon
Posts: 1887
Joined: 29.01.2002
With us: 22 years 9 months
Location: Altair

Post #14by Rassilon » 11.04.2002, 19:52

If your referring to Celestia and making movie clips...there is a capture AVI function within the Capture Movie dialog box accessed under the File menu...MPG, AVI uncompressed and AVI DivX...
I'm trying to teach the cavemen how to play scrabble, its uphill work. The only word they know is Uhh and they dont know how to spell it!

Ortolan
Posts: 120
Joined: 31.03.2002
With us: 22 years 7 months
Location: Melbourne, Australia

Post #15by Ortolan » 12.04.2002, 01:01

If your referring to Celestia and making movie clips...there is a capture AVI function within the Capture Movie dialog box accessed under the File menu...MPG, AVI uncompressed and AVI DivX...

Is that only in the win32 version or a more recent build or something? I'm using 1.2.2 on i686 Linux.

Rassilon
Posts: 1887
Joined: 29.01.2002
With us: 22 years 9 months
Location: Altair

Post #16by Rassilon » 12.04.2002, 01:53

Ortolan wrote:
If your referring to Celestia and making movie clips...there is a capture AVI function within the Capture Movie dialog box accessed under the File menu...MPG, AVI uncompressed and AVI DivX...
Is that only in the win32 version or a more recent build or something? I'm using 1.2.2 on i686 Linux.


Im not quite sure for linux...I suspect chris has put that functionality in all versions...And for win32 it was present in the 1.2.2 version...but if its not located in that specified location...then no you cannot do it in linux...
I'm trying to teach the cavemen how to play scrabble, its uphill work. The only word they know is Uhh and they dont know how to spell it!


Return to “Development”