Procedural Content Generation Framework Node Reference

Remaps the density of each point in the point data to another density value according to the provided curve.

Final Density = Curve Remap(Input Density)

Applies a linear transform to the point densities. Optionally, this can be set to not affect values outside the input range.

D' = (Out_Max - Out_Min) * (D - In_min) / (In_max - In_min) + Out_Min

Uses Geometry Script functions to test if points are inside a tagged actors' static meshes.

Requires the Procedural Generation Framework (PCG) Geometry Script Interop plugin.

Loads an Alembic file into PCG point data.

Requires the Procedural Content Generation (PCG) External Data Interop plugin.

Using the Match And Set Type option, finds a match for each point based on the selection criteria, then applies the value to an attribute.

A common use case is to select meshes to be used downstream in a Static Mesh Spawner node with the By Attribute selector.

Computes the value of the Frac mathematical operation. Takes in an input value and returns the fractional portion of that value.

For example, for an input value X, the result is X minus the Floor of X. The output value ranges from zero to one, inclusive on the low end, exclusive on the high end.

Computes the value of the Sign mathematical operation and writes the result to an attribute. Evaluates an input value and indicates whether it is positive, negative, or exactly zero.

• If the input is negative, this node outputs -1.
• If the input is exactly 0, this node outputs 0.
• If the input is positive, this node outputs 1.

Retrieves the contents of a property from the actor holding the PCG component (or higher in the object hierarchy).

By default, it looks at actor-level properties (useful for Blueprint variables), but it can look at component properties as well using the Select Component option.

Samples points on a specified static mesh. Note that this operation is costly.

Requires the PCG Geometry Script Interop plugin and the Geometry Script plugin.

Samples points using the spline as the source material. Sampling on the spline means directly on the spline curve, while the Horizontal, Vertical and Volume options sample on the spline volume (driven by the radius of the control points in the Y/Z axes).

Sampling inside the spline requires the spline to be closed.

Samples points on a Surface data, in a regular grid pattern. This node contains the following options:

• Point Extents: Defines the basic grid cell size on the surface.
• Looseness: Defines cell size to allow for variation. In practice, the cell size is point extents * (1 + Looseness).
• Points Per Square Meter: Computes the ratio of kept cells. This property limits overcrowding when your grid is large.

Creates a Spline from the input point data. Contains options to create the following:

• Create Component or Create Data Only
• Closed or Linear Splines
• Apply custom in and out tangents from Attribute names.

Outputs the result of the difference of each source against the union of the differences. This node has the following options:

Density Function: Controls which density function is used after the operation is complete. Contains the following options:

• Minimum: Final density is equal to the density of the source minus the max density of all the differences.
• Clamped Subtraction: Final density is equal to the density of the source minus the sum density of all the differences. This value is clamped between 0 and 1.
• Binary: Final density is equal to 0 if the density of the difference is greater than zero. Otherwise the final density is equal to the density of the source.

Mode: Controls behavior of difference in the presence of concrete spatial data vs continuous data (other types, more akin to distribution functions). Note that in some cases where we want to select a concrete element, this should be set to Discrete otherwise the resulting density function could not be flat. Contains the following options:

• Inferred: Discrete if either the source or the differences are points, and continuous otherwise.
• Continuous: Does not collapse the Density function.
• Discrete: Collapses the density function, meaning it converts the data to point data.

Takes all inputs and generates a single collection holding all the input data. Used mainly for organization.

Contains a Dependency Only pin to sequence execution in cases where it is important (such as World Ray Hit Query vs. content spawned in a given graph). Note that all data provided to this pin will not be passed to the output.

General version of the Get … Data nodes. Reads data from an Actor using the Actor Filter and the Mode.

It contains the following options:

Actor Filter: Determines which Actors to consider when fetching Actor data.

Include Children: Determines whether to consider any child Actors of the input.

Mode: Contains the following options:

• Parse Actor Components: Builds PCG data from the components on the Actor.
• Get Single Point: Return one point data containing one point per actor with its transform and local bounds. Note that in most cases a Merge or Union node is used after this node.
• Get Data from PCG Component: Returns the result of another actor's PCG execution. Note that only data passed to the Output node is stored on the PCG component and only some types are supported (Point Data & Attribute Set), and will be queried accordingly. PCG will make sure that the execution of the remote actor will be done before continuing execution.
• Get Data from PCG component or Parse Actor components: Returns PCG output first, if present. Then parses Actors, if needed.
• Expected Pins: Returns data on custom output node pins.

Computes the inner intersection between all data provided to the node, regardless of their pins.

Example: Input contains [A, B, C]

Result: A ∩ B ∩ C

Computes an outer intersection for each data provided in the Primary Source pin, where the result is one intersection per data on the primary source pin against the union of data on each other Source pins.

Example: Primary Source contains [A, B] Source 1 contains [C, D] Source 2 contains [E].

Result: A ∩ (C ∪ D) ∩ E, B ∩ (C ∪ D) ∩ E

Changes the points transforms (either in place or to an attribute with Apply to Attribute) using basic random rules. Each component of the transform (translation, rotation, scale) can be set to Absolute instead of relative to allow for more control.

It contains the following options:

Uniform Scale: Scales point data to the same X, Y, Z ratio.

Recompute Seed: Forces the points seed to be updated according to their new world position.

Example: use a Transform Points node with Absolute Rotation and rotation Z being 0 will make certain that the points are pointing in Z up.This makes sure the points are upwards after sampling from the landscape.

Creates a logical union between data, from a distribution function perspective. Result depends on the density function option chosen.

Density Function: Controls which density function is used after the operation is complete. Contains the following options:

• Maximum: Final density is equal to the max density of the source.
• Clamped Addition: Final density is equal to the sum of the densities in all the differences. This value is clamped between 0 and 1.
• Binary: Final density is equal to 1 if any density of the source is greater than zero. Note that this is rarely useful except in the context of a binary difference.

Creates a surface-like data that performs ray casts in the physics world. It can pass data to any node that expects a surface data.

By default, the size and orientation of the rays is driven by the source component's actor properties (most likely a volume), but the ray properties can be overridden. It contains the following options:

Apply Metadata From Landscape: Gets the landscape layer values if the raycast hits the landscape. Note that there is a small performance cost involved with this.

Ignore PCG Hits: Ignores all PCG generated assets. Is useful when there is no ordering enforced vs other nodes creating data in the world (or other graphs).

Optionally, can return the physical material and the actor hit.

The Filtering elements are used for finer-grained control to ignore or keep only some of the hits.

Creates a volume-like data that gathers points from the physics world. It can pass data to any node that expects a surface data.

The Search for overlap check controls whether overlaps are returned as points or only queries overlapping nothing (subject to filtering) are returned.

Can also optionally return the actor "found" in that volume.

Spawns either the contents of an actor or an actor per point in the provided input data. The actor is driven by the template actor class or the instanced templated actor or by attribute depending on the settings.

It contains the following options:

Template Actor Class: List of available Actors in your project.

Option:

• Collapse Actors: Gathers some of the actor components (Static Mesh Components and PCG Components) and acts collapsed inside of the target actor.
• Merge PCG only: Spawns one actor per point If the spawned actor has a PCG component, its inputs are bundled into a single graph execution.
• No Merging: Spawns one actor per point.

Spawn one static mesh per point in the provided point data.

Static Mesh options are added to the Mesh Entries array and selected using each entry's Weight. This is done by taking the sum of all the weight values and converting them to a percentage for each entry. For example, if there are four entries in the array that each have a value of 1, the value of the sum is 4. Each entry's weight is then divided by the sum and converted to a percentage. This means each entry has a 25% chance of spawning.

Selection of the static mesh is done using variations based on the selected Mesh Selector Type option.

It contains the following options:

• PCG Mesh Selector Weighted: Selects an entry based on the total weight of entries.
• PCG Mesh Selector By Attribute: Selects an entry based on an attribute present on the mesh.
• PCG Mesh Selector Weighted By Category: Selects a category by looking up an Attribute on the point, then selects an entry based on weight.