Celestia Forums

I got an email from the official Cassini site and noticed a very nice high res cross section of Saturn's rings. Here is the LINK
http://ciclops.org/view.php?id=3858
I looked at the ring image that comes with Celelstia and tried to create the image with the higher res image. No Luck! The Celestia image is a png file and has some transparency but no alpha channel. Who knows how this images works and is mapped on to the Saturn ring polygon?

Thought it would be nice to get a higher res image of the rings as an option.

Don

BTW They are noting that Cassini has been out in space for ten years and have posted several nice images of Saturn and related moons that have been taken over the years.
http://ciclops.org/view_event.php?id=71

Don,

The resolution of the rings in Celestia is limited by the size of your graphics card's texture input buffer: 1K, 2K or 4K. There is no workaround for this. The saturn-rings.png image shipped with Celestia is only 1K so that it can work on almost all platforms. It does include an alpha channel: that's what makes it partially transparent.

Grant Hutchison spent a lot of time in 2004 creating 4K Saturn ring texture from the best available data using density profies measured in 1989. It's available at http://www.lepp.cornell.edu/~seb/celest ... rings.html

Of course the saturn rings in Celestia do have an alpha channel. You just have to know how to extract it! .

The alpha channel of the rings was actually the more tedious part of the work, largely done by Grant Hutchison (and some help & discussions by me).

The alpha channel contains the earthbound hires measurements of the ring /transparency/ as a grayscale image. The transparency was obtained by measuring the amount of transmitted light during the passage of the rings in front of a star (of known brightness)! The RGB part of the saturn-ring texture then is to contain ONLY the ring albedo (reflective part)! A clean separation of these two intrinsically different components is not quite easy...

Hence you cannot simply combine the two textures without doing some proper "renormalization"...

Bye Fridger

Don,

You SHOULD be able to take a small vertical slice of that image and create
a PNG with alpha and transparency using the Gimp. If you're not familiar
with that program just hang on a while. I've downloaded the ring images
to see if I can create a texture for use on my machine.

Keep in mind that I'm virtually Brain-Dead though, so I can't promise
anything. We'll see...

Is it possible to have two sets of rings around a planet? One set like what you see around Saturn and then a second set that is perpendicular and encircles the planet from pole to pole?

Reiko,

I'm not sure if you're asking if it's physically possible or if Celestia can show it.

It's physically possible only for a very short time with planets as oblate as Saturn. The differential gravitational effects will quickly (on an astronomical time scale ) drag the particles into an equatorial orbit.

Celestia's Ring definition draws rings only in an equatorial plane. However, one can create a 3D model of a set of rings and position it at any angle desired.

p.s. One could also define a "fake planet" with a diameter smaller than the visible planet and tilted at any desired angle. Another set of Rings could be defined around that planet. In either case, (model or fake planet) i seem to recall that Celestia may generate some undesirable visual effects.

selden wrote:Reiko,

I'm not sure if you're asking if it's physically possible or if Celestia can show it.

It's physically possible only for a very short time with planets as oblate as Saturn. The differential gravitational effects will quickly (on an astronomical time scale ) drag the particles into an equatorial orbit.

Celestia's Ring definition draws rings only in an equatorial plane. However, one can create a 3D model of a set of rings and position it at any angle desired.

p.s. One could also define a "fake planet" with a diameter smaller than the visible planet and tilted at any desired angle. Another set of Rings could be defined around that planet. In either case, (model or fake planet) i seem to recall that Celestia may generate some undesirable visual effects.

I didn't think it would be possible in real life. I was just wondering if it was possible in Celestia. It would make an interesting sci-fi planet.
I'll try the fake planet approach and see how it looks.

Thank you.

Reiko wrote:It would make an interesting sci-fi planet.
I'll try the fake planet approach and see how it looks.

Thank you.

I think it looks kind of "elegant" more like the vertical stabilizer of a plane. Notably if a "rudder" could be added to that unusual ring, such a ringed planet would always know where to fly

Bye Fridger

Thanks for the replies.

I figured it out. I have CS3 and know Photoshop (older versions) pretty well. I am getting up to speed in CS3. For some reason CS3 does not show the alpha channel in the channels area. It does show the transparency. When I loaded the image in an earlier version of PS I then saw the alpha were I thought it should be (in the channels area). I am sure there is some stupid reason for hiding it in the newer version of PS but i have not figured that out.

Thanks Fridger, I did not know that about the alpha map. I think I can do what I wanted to do. I will just use the alpha channel from Celestias image and put the more detailed rgb image from the Cassini sight. I will then upload it to motherload for others to use if they want. It will be 2048 x 2 pixal image for the high res folder. It won't have the little ripples because it is only 2 pix wide.

DonAVP wrote:Thanks for the replies.

I figured it out. I have CS3 and know Photoshop (older versions) pretty well. I am getting up to speed in CS3. For some reason CS3 does not show the alpha channel in the channels area. It does show the transparency. When I loaded the image in an earlier version of PS I then saw the alpha were I thought it should be (in the channels area). I am sure there is some stupid reason for hiding it in the newer version of PS but i have not figured that out.

Thanks Fridger, I did not know that about the alpha map. I think I can do what I wanted to do. I will just use the alpha channel from Celestias image and put the more detailed rgb image from the Cassini sight. I will then upload it to motherload for others to use if they want. It will be 2048 x 2 pixal image for the high res folder. It won't have the little ripples because it is only 2 pix wide.

Are you sure that the RGB map you were citing
http://ciclops.org/view.php?id=3858
is a pure reflection map??? Where is it precisely specified?

F.

Fridger

No I don't think it is. I did try and create something but the older/smaller alpha map did not line up with the newer image. It looked interesting but was not accurate and not as correct as the original. I have stopped working on it. If someone else wants to give it a shot be my guest. Near as I can do is guesstimate the alpha. BTW it had more vivid colors but as far as detail it did not and that is because of the 2 pix smear. I guess the best would be to image the rings on a disk but I have not seen any complete images of the rings without the planet's shadow.

Don

Well I HATE to mention it, but I DID manage to create a fabulous texture
for Saturn's rings from the image links posted originally. I used the Gimp
and I added a transparency layer to the whole thing so that the casual
passerby can even SEE THROUGH the rings now. I need to smooth it out
just a tad more, but this is encouraging.

If a Brain-Dead Bozo like me can create the ring texture in about 20
minutes, then what's the big deal here?

Don? If you want to have a look just let me know. I can e-mail a copy to
you as an attachment if you like. BTW, the texture is 4096x18 pixels, but it's
only about 64k in total size.

See ya, Bob

Bob,

Did you use Grant's transparency channel?
It should be possible to extract it from his ring texture.

selden wrote:Bob,

Did you use Grant's transparency channel?
It should be possible to extract it from his ring texture.

Nope...

Just created a brand-new texture from the image link posted above. I
KNOW that it's not to scale, but I just wanted to see how hard it was to
create the texture. What I ended up with was a really beautiful set of
Saturn rings which can be seen through. Again, I'm just playing here, but
my point is that it CAN'T be THAT hard to create a texture.

The real difficulty lies in the scaling of the texture, and - in THAT area - I
don't know BEANS about texturing.

Question: is this the right way to make a "Saturn rings"? Do not mind the large canvas, is just for the visual purpose.

selden wrote:
Grant Hutchison spent a lot of time in 2004 creating 4K Saturn ring texture from the best available data using density profies measured in 1989. It's available at http://www.lepp.cornell.edu/~seb/celest ... rings.html

I just tried using those textures on 1.5.0pre3 and got nowhere. The alpha channel doesn't even display.

DonAVP wrote:Fridger

I did try and create something but the older/smaller alpha map did not line up with the newer image. It looked interesting but was not accurate
Don

I disagree. I rebuilt and rescaled the rings using the alpha channel of Grant's 4k texture. The alpha channel pattern matched pretty well after proper rescaling. If I find some time I'll display the result for download in Celestial Matters (where else? )

Bye Fridger

srosenow_98 wrote:

selden wrote:
Grant Hutchison spent a lot of time in 2004 creating 4K Saturn ring texture from the best available data using density profies measured in 1989. It's available at http://www.lepp.cornell.edu/~seb/celest ... rings.html

I just tried using those textures on 1.5.0pre3 and got nowhere. The alpha channel doesn't even display.

Of course it does.

Bye F.

srosenow_98,

Exactly what kind of graphics hardware do you have?
This information is shown in Celestia's menu Help/ OpenGL info.

if you can't view Grant's ring pattern, then your graphics card most likely cannot display 4K textures. I just now downloaded the texture and verified that it works on an Nvidia card.

Most older ATI cards can only display 2K textures, for example, and many embedded graphics chipsets are limited to 1K textures.

The differences between Grant's 1K texture shipped with Celestia and his higher resolution 4K texture are relatively subtle. Here are images of the rings within the Cassini Division as shown by the lores and hires textures. Note the blurring of the rings as seen when using lores texture.



(these thumbnails link to much larger images)

So here comes my PRELIMINARY new Saturn ring composition as promised. The final version will be announced for download in due time at Celestial Matters.
At this point there follows a careful tuning of reflectivities vs transparency (alpha channel brightness!) to match the visual appearances perfectly. This will take some time.

What I did so far is: using the 4k alpha channel of Grant's Saturn ring and mapping it in size and LOCATION to the new true-color ring texture from Ciclops. This requires some non-trivial rescaling manipulations. Note in particular that the beginning and the ends of the 2 ring textures are NOT the same! It's standard image manipulation stuff and I have no time to explain it here. More in CM. Also the beginning and end of the rings have to be adjusted properly in solarsys.ssc.

In the Celestia images below you may see how accurate I did all this, since the moon Pan is always visible in the correct place (travelling in the empty Encke division!)

To start out: recent natural color image (Ciclops)


Detail view of the rings from that shot (Ciclops)


Next the new rings in Celestia:
(Click for a large image!)

(Click for a large image!)



Bye Fridger