Enum bevy::render::render_resource::VertexStepMode

#[repr(C)]
pub enum VertexStepMode {
    Vertex = 0,
    Instance = 1,
}

Whether a vertex buffer is indexed by vertex or by instance.

Consider a call to RenderPass::draw like this:

render_pass.draw(vertices, instances)

where vertices is a Range<u32> of vertex indices, and instances is a Range<u32> of instance indices.

For this call, wgpu invokes the vertex shader entry point once for every possible (v, i) pair, where v is drawn from vertices and i is drawn from instances. These invocations may happen in any order, and will usually run in parallel.

Each vertex buffer has a step mode, established by the step_mode field of its VertexBufferLayout, given when the pipeline was created. Buffers whose step mode is Vertex use v as the index into their contents, whereas buffers whose step mode is Instance use i. The indicated buffer element then contributes zero or more attribute values for the (v, i) vertex shader invocation to use, based on the VertexBufferLayout’s attributes list.

You can visualize the results from all these vertex shader invocations as a matrix with a row for each i from instances, and with a column for each v from vertices. In one sense, v and i are symmetrical: both are used to index vertex buffers and provide attribute values. But the key difference between v and i is that line and triangle primitives are built from the values of each row, along which i is constant and v varies, not the columns.

An indexed draw call works similarly:

render_pass.draw_indexed(indices, base_vertex, instances)

The only difference is that v values are drawn from the contents of the index buffer—specifically, the subrange of the index buffer given by indices—instead of simply being sequential integers, as they are in a draw call.

A non-instanced call, where instances is 0..1, is simply a matrix with only one row.

Corresponds to WebGPU GPUVertexStepMode.

Variants

Vertex = 0

Vertex data is advanced every vertex.

Instance = 1

Vertex data is advanced every instance.

Trait Implementations

impl Clone for VertexStepMode

fn clone(&self) -> VertexStepMode

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for VertexStepMode

fn fmt(&self, f: &mut Formatter<'_>) -> Result<(), Error>

Formats the value using the given formatter.

impl Default for VertexStepMode

fn default() -> VertexStepMode

Returns the “default value” for a type.

impl Hash for VertexStepMode

fn hash<__H>(&self, state: &mut __H)

where __H: Hasher,

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for VertexStepMode

fn eq(&self, other: &VertexStepMode) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for VertexStepMode

impl Eq for VertexStepMode

impl StructuralEq for VertexStepMode

impl StructuralPartialEq for VertexStepMode