On this page |
This node generates a noise pattern using a fractal approach. The same underlying noise is sampled at increasing frequencies and mixed together. Use the Fractal Noise 3D COP if you want the source location to be the pixel’s world coordinates instead of its image coordinates.
For example, use this node to create cloudy, smoky, or TV static patterns.
Parameters ¶
Signature
The layer type that the source accepts.
See Signatures for more information.
Range ¶
Amplitude
The scale of the noise, which the node applies before the Center adjustment.
Note
Layers can store negative numbers and values above one, so this may result in out-of-bounds values. You can use a Clamp COP to enforce the range afterwards.
Center
The center of the output noise range.
Contrast
The amount of contrast to apply to the noise before the Amplitude and Center parameters. You can use this parameter to make the noise more extreme without exceeding the 0
to 1
range.
Pattern ¶
Noise Type
The underlying noise function that’s iteratively sampled to generate the fractal noise.
Torus
Create 4D simplex noise evaluated on a flat torus. This generates noise that’s periodic by construction, which means it doesn’t clamp or round settings to enforce periodic behavior (unlike other noise types).
Perlin
Create Perlin noise on a regular grid. This type of noise generates more natural results.
Worley Cellular F1
Create Worley noise using its primary distance value, which is the distance to the closest worley point. This generates a cell-like noise.
Worley Cellular F2-F1
Create Worley noise using the difference between its second closest and closest points. This generates a cell border-like noise.
White (Random)
Create white noise where each noise element is given a constant but random value.
Per Component
Computes a separate noise for each channel of the output. If off, the output is grayscale for mono and non-mono layers.
Metric
When Noise Type is Worley Cellular F1
or Worley Cellular F2-F1
, this is how to define different metrics for computing the distance to a point.
Euclidean
The usual L2 distance metric, which results in circular shapes.
Manhattan
The maximum of the two axial distances, which results in diamond shapes.
Chebyshev
The sum of the two axial distances, which results in square shapes.
Element Size
The size (in image coordinates) of the basic element of the noise. You can turn on the Per-Component Controls button to adjust this further using the Element Scale parameter.
Element Scale
When the Per-Component Controls button is on, this is the per-axis scaling of the element size for anistropic noise.
Offset
The offset of the noise function in image coordinates.
Tile Size
The size of a single tile of noise. The noise periodically repeats in this size. The size is in image coordinates, so the default is for the entire default canoncial image. If you have a non-square image, this should match the aspect ratio.
Note
If on, values for parameters like Element Size and Lacunarity are rounded or clamped to make them valid. This is because these types of parameters must meet certain conditions for the noise to be tileable.
3D Noise ¶
3D Noise
Evaluates the noise in a 3D space. The first two dimensions are the location in the image and the third is controlled separately.
Animate
Implicitly adds time to the third coordinate.
Pulse Length
The feature size of the noise in the third dimension. This is the rate at which the noise switches appearances.
Time Offset
The fixed offset for evaluating the noise.
Time Scale
The amount by which to scale the computed time.
Loop Length (sec)
This is the time in seconds at which to repeat. This clamps other options to enforce periodic behavior.
Fractal ¶
Max Octaves
The amount of times to scale add together the noise.
Lacunarity
The amount to scale the noise for each iteration, which is rounded to an integer value for periodic noise.
Roughness
The amount to scale the amplitude of each successive noise.
Inputs ¶
size_ref
A representative layer that determines the size of the output image and controls the metadata.
pos
An optional UV layer with a value that’s used instead of the pixel’s image coordinates for the noise.
time
Each pixel uses this layer’s value for the time.
Outputs ¶
noise
The computed noise.
See also |