Skip to contents

A smoothed Brownian bridge, as a path distortion

Source: R/noise_bridge.R
noise_bridge.Rd

noise_bridge bundles the settings needed to generate a smoothed Brownian bridge (via the internal smooth_bridge() helper): how many local-averaging smooth-ing passes to apply, and which seed to draw from.

Usage

noise_bridge(smooth = 3L, seed = 1L)

Arguments

smooth

Number of smoothing passes. Must be non-negative. Default 3L.

seed

Integer seed. Default 1L.

Details

Unlike noise_field (sampled at arbitrary (x, y) positions in the plane), a Brownian bridge has no spatial position to sample at – noise_sample()'s method for noise_bridge instead takes a point count n and a scale, returning a length-n displacement vector. This is the distortion behind shape_twist()'s wandering path, split out into its own class for the same reason noise_field was: so the distortion is a first-class, swappable object rather than bare constructor arguments, and so a future curve_twist() (an open, unfilled wandering path) can reuse it without duplicating shape_twist()'s path logic.

See also

Other noise helpers: noise_field(), noise_sample()

Examples

noise_bridge(smooth = 5L, seed = 4821)
#> <sketchpad::noise_bridge>
#>  @ smooth: int 5
#>  @ seed  : int 4821

# more smoothing passes give a gentler bridge; embedding it in
# shape_twist()'s path_distortion makes the effect easy to see
draw(shape_twist(
  x = 0, y = 0, xend = 1, yend = 0, width = 0.2,
  path_distortion = noise_bridge(smooth = 0L, seed = 7734)
))

draw(shape_twist(
  x = 0, y = 0, xend = 1, yend = 0, width = 0.2,
  path_distortion = noise_bridge(smooth = 20L, seed = 7734)
))