 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I'm rendering an earth fly-in using Pov-Ray 3.6.1/Athlon XP/homebrew Linux.
I'm using scattering media for the atmosphere.
Each frame calculates units so that the camera is 1000 POV units from the
origin (earth center).
The media color is divided that frame's meter size (i.e. POV units per
meter), to make it scale properly. This makes for some pretty
large-magnitude colors.
Problem:
As I zoom in, the atmosphere's brightness flickers, apparently in a sawtooth
wave (it takes a few frames to get bright and dark again).
Is there some obvious reason why this would happen? When I get rid of the
atmosphere, I have no flicker, so it's not my light source.
-Eric
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Looks like it was another scale issue. My density function:
1013.25 * pow(e,-.029*9.8*(pow(x*x+y*y+z*z,0.5)-earth_radius)/8.134/298)
with earth_radius a little less than 1000, may have pushed some
floating-point limits. Shrinking the entire scene helped, for now.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Eric Buddington" <ebu### [at] wesleyan edu> wrote:
> Problem:
> As I zoom in, the atmosphere's brightness flickers, apparently in a sawtooth
> wave (it takes a few frames to get bright and dark again).
>
> Is there some obvious reason why this would happen? When I get rid of the
> atmosphere, I have no flicker, so it's not my light source.
>
> -Eric
Sounds like it might be related to the number of samples (and/or intervals)
in the media. Too low, perhaps? If you have the time for an animation
experiment, try REDUCING the number of samples, to see if the problem gets
worse. (Each image should render a bit more quickly as well.)
Ken W.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Oh, forgot to mention:
Some of the default values given in the POV documentation for media are
incorrect. I posted something about this awhile ago...
http://news.povray.org/povray.documentation.inbuilt/thread/%3Cweb.434c8b2fe42c859ad82bb7c00%40news.povray.org%3E/
Also, since media density doesn't "scale" with an object, as your media
sphere(?) gets larger or smaller, the media samples and/or intervals need
to change along with it, to preserve the same "look" --that is, so
discontinuities don't start appearing in the media as the container is
scaled larger. At least, this has been my experience.
Ken
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
