Page 1 of 1

Amalthea textures need correction

Posted: 12.02.2004, 16:03
by danielj
See this link :
http://photojournal.jpl.nasa.gov/catalog/PIA02531
There is a bright streak near the south pole of Amalthea and this could be put in its texture.Additionally,there is a "sawtooth" terminator,that I don?t know if it is in the model/texture.
Last,a curiosity : Galileo was supposed to pass much nearer to the jovian moon,but we don?t have more detailed pictures.Was Galileo?s camera shut down in this passage?

Posted: 12.02.2004, 16:51
by granthutchison
The bright patch within the south pole crater Gaea is already present in the Celestia Amalthea texture - its brightness is very much exaggerated in the photo you link to.
Galileo took no pictures during its close encounter with Amalthea.

Grant

Posted: 12.02.2004, 21:01
by Bob Hegwood
Daniel,

The "sawtooth" effect you mentioned can be repaired by using Jestr's new model of Amalthea. Keep in mind, however, that you have to be using Celestia 1.3.2 Pre1 or later (I think!) in order to get the correct orientation now.

Have a look at this topic for more information on Jestr's website(s)?

Hope that helps...

Take care, Bob

Posted: 12.02.2004, 22:11
by granthutchison
Bob Hegwood wrote:The "sawtooth" effect you mentioned can be repaired by using Jestr's new model of Amalthea.
Bob, the sawtooth daniel is referring to is a real world effect noted on the photo of Amalthea he linked to - it's discussed in the associated caption.

Grant

Posted: 12.02.2004, 23:44
by Bob Hegwood
Sorry Mr. Hutchison...

The original model which came with Celestia *also* had a sawtooth effect on it - from the mesh of the model I presume.

I stand (or sit) corrected once again. I'll just keep my mouth shut from now on. :oops:

Thanks, Bob

Maybe, maybe not

Posted: 19.02.2004, 08:44
by sombrehombre
Two kinds of sawtooth effects are at work in the mesh - 1) from the
fixed threshold for sharp edge rendering under OpenGL/DirectX, and
2) because the seam between -180 degrees and +180 degrees in the
adaptations from spherical coordinates in most of the meshes was never
closed, graphics cards don't blend pixel values between the two surfaces.

Problem 1) is easily fixed by local editing to slightly adjust points
on the sharp edge (and perhaps adjacent) in your favorite 3d
modeling program, so that the edge is rendered as rounded. If you don't
give a damn about filesize or framerate, you can also subdivide faces
(a common solution hereabouts), which accomplishes the same thing
because many 3d programs will just try to minimise angles between
face normals when they subdivide. Problem 2) is addressed
by merging points in the mesh that are closer than some tiny threshold.

I made my own version, with a fixed mesh (actually, I've fixed the
meshes of every single 'real' minor planet or asteroid I've found - not
including Jestr's lovely but fanciful creations), and new texture based
on Phil Stooke's map that looks more like the Voyager image (http://www.solarviews.com/cap/jup/amalthea.htm) than the blood-red of the distribution texture. I could post it to my 1-day old website if
there's a demand.