funny lightining on lo, help ?

[adyga]

hi, I am really enjoying Celestia, and I was using it since the old openuniverse, I guess its getting better by the times, good job from you programmers and artists and astronauts.
today as I opened up Celestia 1.3.0 on my Windowz machine, as usual it was set to "lo" and holy**** I saw a funny look at "lo", as it was covered by Jupiter's shadow:ambient light was set to medium, as anybody can see (maybe) the "lo" was split into 2 different lightened areas, although it was all of it in the shadow of Jupiter!! is that a bug ?? or a bad hardware ?
*another thing is: I downloaded the jupiter's 2k tex. from t00fri's page and I discovered that there is a missmatch between the mid tex. and this tex.help?
my system is:
win2k
GF2MX400
detonator4345
..
greetings

[Christophe]

I confirm, this is a bug. Night side and eclipsed zones do not react to ambient light in the same way which leads to this strange look.

[adyga]

thanks a lot for help, but what about Jupiter's high texture ?? is there a mismatch ?

[Christophe]

What do you mean exactly by mismatch?

[adyga]

sorry for my poor english, but I mean that if I look at Jupiter where the big spot is at mid tex. and then I switch to hi tex (t00fri's) I lose the spot and now its on another coordinates !! same goes for the whole tex not only spot

[Christophe]

There's been a change in orientation of textures between Celestia 1.2.5 (when Fridger released is textures on TexFoundry) and 1.3.0 which you're using.

Fridger was supposed to update his site with corrected textures soon...

[adyga]

ohh now I can sleep well,at least its not my poor nvidia card
thanks a lot for help !

[JackHiggins]

Heres how to solve the problem with the GRS (great red spot)

Open the original texture, and find out how many pixels there are between the center of the GRS, and the left edge of the image.

Put this number over the width of the image, and get a percentage. Then multiply this by 360 to find out how many degrees this is.

Now do the same for fridger's texture.

If the first was 180 degrees, and the second was, (say) 270 degrees, you know the difference is 90 degrees to the East.

Now, add the line Orientation [ -90 0 1 0 ] to the Jupiter section of solarsys.ssc

This means to rotate jupiter's texture 90 degrees West, to compensate for the texture being misaligned.

If i've got anything wrong here, correct me, but i think i'm pretty much right...?!!

[Guest]

Jack, you are absolutely right, but I discovered that the textures are somehow imaginery and when I wanted to fix that problem of Jupiter, there was a mismatch between the hi res tex and the rest! considering your talk about realism of t00fri's texs in another thread of this forum, which tex is the most native one ?? the tex packed with celestia or t00fri's 2k ??
well even if t00fri's was fake , that does not say its not a piece of art though, and I can close my eyes and put the tex in any orientation that I would like, but which has the GRS at the right orientation ??
by the way, great site Jack, I admire and feel jealous of this state of art ed models !

[JackHiggins]

The standard one is based on galileo images, and t00fri's is based on cassini images. (I think!!)

But the standrad celestia one is definitely oriented correctly.

Since jupiter's atmosphere changes so rapidly, any jupiter texture becomes innacurate almost as soon as it's made. t00fri's is a higher res though, & that's always good!

[granthutchison]

There really isn't a "correct" texture position for the Great Red Spot, since it drifts in longitude relative to the rotation of Jupiter's magnetic field, which is the basis for the definition of the prime meridian on Jupiter.
There's some discussion on how to correctly position individual textures here:
http://www.shatters.net/forum/viewtopic.php?t=1497

To correctly position Fridger's 2k texture, you need to revise the RotationOffset for Jupiter to

Code: Select all

   RotationOffset     -41 #jupiter2k.jpg


Grant

[JackHiggins]

To correctly position Fridger's 2k texture, you need to revise the RotationOffset for Jupiter to

Code: Select all

   RotationOffset     -41 #jupiter2k.jpg


But that's still the same as saying

Code: Select all

Orientation [ -41 0 1 0 ]

isn't it? Or is there a difference?

[granthutchison]

You can certainly achieve the same results using Orientation and RotationOffset. But there are a couple of differences in the details:

Trivially:
Orientation rotates things clockwise around the principal axes.
RotationOffset follows the standard anticlockwise direction of planetary rotation.

But also:
Orientation rotates the object relative to the coordinate system.
RotationOffset rotates the object and coordinate system together.

So Orientation is handy if you want to move a texture into alignment with the correct longitude of the planet it is attached to.

Grant

[JackHiggins]

Ahhh - now I get it!

[Guest]

I changed the configuration of rotationoffset to -41, well it works, but for the hi res tex only and now the other are screwed, however I will stick with hires, since I don't see a use for low res tex on my machine
if Jupiter's tex is imaginery and all it matters is the GRS, why nobody wants to make an 8k monster tex for the monster planet of our GREAT solar system ?? well I support that and pray for you !
thanks a lot for information.