Field Noise

Read with AI
All docs in one file (llms-full.txt)
FieldNoise #Field #Node
Field Noise

Generates a flow field (FlowField) from curl noise. It takes the curl of a potential function (2-octave fbm of valueNoise), so the result is a flow of vortices only, with no sources and no sinks. The time phase is Speed × deltaTime integrated every frame, and the Offset input is added on top of it.

  • Scale — the spatial scale. Clamped to 0.25-16, default 3. The larger it is, the finer the vortices. The strength of the field is also proportional to Scale (the output is curl × 0.02 × Scale)
  • Speed — how fast the phase advances. Clamped to -4 to 4, default 0.5. Negative values play in reverse
  • Offset — a value added directly to the phase (Offset input port / offsetInput). Not clamped
  • Seed — an integer. When a node is newly created it is a random number determined automatically from the runtime value (a saved value is restored as is)

The output resolution is 1/4 of the global resolution. The field is a low-frequency signal, so it is built light on the assumption that the consumer reads it with bilinear filtering. The format is RGHalf (signed; the values are “displacement in UV space per second”).

To increase or decrease the strength, place a Field Scale after it.

Video

Input ports

Port IDType
Offsetfloat

Output ports

Port IDType
FlowFlowFieldOutput

Scriptable members

PathType
offsetInputfloat
currentScalefloat
currentSpeedfloat
currentOffsetfloat
currentSeedint

Synapse Apps (scripting)

await synapse.modules.create({ type: "FieldNoise" })

Pages linking here