Models in OpenGL2.0 render path

All about models for Celestia. How-to, programs, work in progress...
Topic author
fungun
Posts: 315
Joined: 30.07.2007
Age: 63
With us: 17 years 3 months
Location: Iowa, USA

Models in OpenGL2.0 render path

Post #1by fungun » 17.12.2011, 15:09

I am lost for an answer to this problem.
Some of the models I use do not work right in the OpenGL 2.0 render path. See pics.
Does this have to do with transparent parts?

Thanks,
Tim

Avatar
selden
Developer
Posts: 10192
Joined: 04.09.2002
With us: 22 years 2 months
Location: NY, USA

Re: Models in OpenGL2.0 render path

Post #2by selden » 18.12.2011, 16:55

I don't know if this is your problem exactly, but Celestia has always had problems depth sorting transparent objects. If an object's transparency is generated in a material's surface texture image, it helps a little if you define the material itself to have an opacity slightly less than 1.0 (e.g. 0.99), but that doesn't fix the problem in many cases.

Unfortunately, showing what is "wrong" doesn't help much if the observer doesn't know what is "right". Your pictures need to have some explanations added.
Selden

Topic author
fungun
Posts: 315
Joined: 30.07.2007
Age: 63
With us: 17 years 3 months
Location: Iowa, USA

Re: Models in OpenGL2.0 render path

Post #3by fungun » 18.12.2011, 23:25

The vger cloud should be the blue one,(using the opengl vertext render path) not the grey tone colors as in the inset image.(using the opengl 2.0 render path)

The Enterprise bussard should look like the inset image,,(using the opengl vertext render path) not the bright yellow you see in the main image..(using the opengl 2.0 render path)

Does it make more sense now?

Thanks,
Tim

Avatar
Fenerit M
Posts: 1880
Joined: 26.03.2007
Age: 17
With us: 17 years 7 months
Location: Thyrrenian sea

Re: Models in OpenGL2.0 render path

Post #4by Fenerit » 19.12.2011, 08:33

As for Enterprise' bussard should seem matter either of material's emissivity, and in particular an emissivity toward the yellow (1 1 0), or yellow specular color or both. For Viger's cloud, no clues about the blue-b/w shift. You could try to assign the blue color to the transparent mesh; albeit a point sprite maybe should best to render cloudy objects.
Never at rest.
Massimo


Return to “Modelling”