|
|
Jonathan Rafael Ghiglia <j.l### [at] liberoit> wrote:
> What's the difference between the float value which follows the
> pattern type (inside a normal statement) and the bump_size value?
> Which one should I play with in order to control the amount of
> perturbation? Can I set an accuracy value for normal patterns
> (to get better results with patterns which are not likely to be used
> as normal patterns)?
Specifying a float after the pattern name is simply a shortcut for
specifying a bump_size value - there is no difference in the rendering, and
either can be used to control the amount of perturbation.
As for accuracy, this is generally determined only by the rendering
resolution - all POV-Ray's internal patterns are algorithmic, and can be
scaled to any reasonable size without degradation (barring precision limits,
and a small bug with the current implementation of "noisy" patterns).
Level of detail becomes important with image based patterns (i.e. bump_map),
for which POV-Ray offers the interpolation feature, and turbulence that you
add to patterns. With turbulence the detail is determined by the settings
you use - generally, increasing the octaves value should add extra, smaller
perturbations which you might want if viewing a pattern very closely.
Post a reply to this message
|
|
|
|
In article <3a467ec5@news.povray.org>, "Chris Colefax"
<chr### [at] tagpovrayorg> wrote:
> As for accuracy, this is generally determined only by the rendering
> resolution - all POV-Ray's internal patterns are algorithmic, and can
> be scaled to any reasonable size without degradation (barring
> precision limits, and a small bug with the current implementation of
> "noisy" patterns).
Actually, the "accuracy" is a constant in the official version, and is
always set to 0.02, so it probably could give inaccurate results in
normals with finely scaled patterns. This only applies to normals, and
only normals using slope_maps or which aren't built-in.
MegaPOV allows you to set this "delta" value with the "accuracy"
keyword, which defaults to 0.02 in the normal statement.
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|