Dear all,

thank you for your sugegstions! Please let me comment on them in the following:

1)

The suggestion to use "intelligent" antialias-options, so each frame only takes
the rendering time really required is difficult for my animation. This clip
starts with a huge number of particles in front of a black background, which -
at the beginning of the animation - are such far away from the camera that their
visible size is much less than the size of a pixel in the animation. So all
particles together should appear as a kind of fog. In order to avoid flickering
in the animation which would stem from the fact, that most of the rays don't hit
the particles, but sometimes incidentally do, I need to use these values:

Sampling_Method=2
Antialias_Depth=3
Antialias_Threshold=0

The zero-threshold is required to force POV-Ray to apply many rays even if the
brightness difference between the test rays is zero, since most test rays hit
the black background, and only a very small fraction would hit the particle
inside this pixel. Or do you have a better idea? Such slow antialias-options
would strongly slow down the rendering in the ongoing animation, when big
isosurfaces dominate the scene, which could perfectly and much faster be
rendered with e. g.

Sampling_Method=1
Antialias_Depth=2
Antialias_Threshold=0.75

2) Writing different ini-files will work, but is not a very "smart" way

3) What about the suggestion, that I can put the commands of the ini-file
alternatively into the scene-file, such as:

ini_settings{Sampling_Method=2 Antialias_Depth=3
Antialias_Threshold=frame_number/1000}

Would this be an idea for future development?

Post a reply to this message