Struct InstanceBuffer

pub struct InstanceBuffer<K: BufferLayout> {
    pub unprocessed: Vec<Vec<OrderedIndex>>,
    pub buffers: Vec<Option<InstanceDetails>>,
    pub clipped_buffers: Vec<Vec<ClippedInstanceDetails>>,
    pub buffer: Buffer<K>,
    pub layer_size: usize,
    /* private fields */
}

Instance buffer holds all the Details to render with instances with a Static VBO. This stores and handles the orders of all rendered objects to try and reduce the amount of GPU uploads we make.

Fields

unprocessed: Vec<Vec<OrderedIndex>>

Unprocessed Buffer Data.

buffers: Vec<Option<InstanceDetails>>

Buffers ready to Render

clipped_buffers: Vec<Vec<ClippedInstanceDetails>>

Clipped Buffers ready to Render.

buffer: Buffer<K>

The main Buffer within GPU memory.

layer_size: usize

Size each Buffer Layer gets allocated to for Future buffers.

Implementations

impl<K: BufferLayout> InstanceBuffer<K>

pub fn create_buffer( gpu_device: &GpuDevice, data: &[u8], layer_size: usize, ) -> Self

Used to create a InstanceBuffer. Only use this for creating a reusable buffer.

Arguments

• data: The contents to Create the Buffer with.
• layer_size: The capacity allocated for any future elements per new Buffer Layer.

pub fn add_buffer_store( &mut self, renderer: &GpuRenderer, index: OrderedIndex, buffer_layer: usize, )

Adds the Buffer to the unprocessed list so it can be processed in InstanceBuffer::finalize This must be called in order to Render the Object.

Arguments

• index: The Order Index of the Object we want to render.
• buffer_layer: The Buffer Layer we want to add this Object too.

pub fn finalize(&mut self, renderer: &mut GpuRenderer)

Processes all unprocessed listed buffers and uploads any changes to the gpu This must be called after InstanceBuffer::add_buffer_store in order to Render the Objects.

pub fn new(gpu_device: &GpuDevice, layer_size: usize) -> Self

Creates an InstanceBuffer with a default buffer size. Buffer size is based on the initial crate::BufferLayout::default_buffer length.

Arguments

• layer_size: The capacity allocated for any future elements per new Buffer Layer.

pub fn count(&self) -> u32

Returns the instances count.

pub fn len(&self) -> u64

Returns the instances byte count.

pub fn is_empty(&self) -> bool

Returns if the instance buffer is empty

pub fn max(&self) -> usize

Returns instance buffers max size in bytes.

pub fn is_clipped(&self) -> bool

Returns if the buffer is clipped or not to deturmine if you should use buffers or clipped_buffers.

pub fn set_as_clipped(&mut self)

Sets the Buffer into Clipping mode. This will Produce a clipped_buffers instead of the buffers which will still be layered but a Vector of individual objects will Exist rather than a grouped object per layer. Will make it less Efficient but allows Bounds Clipping.

pub fn stride(&self) -> usize

Returns buffer’s stride.

pub fn instances(&self, bounds: Option<Range<u64>>) -> BufferSlice<'_>

Returns wgpu::BufferSlice of vertices. bounds is used to set a specific Range if needed. If bounds is None then range is 0..vertex_count.

pub fn with_capacity( gpu_device: &GpuDevice, capacity: usize, layer_size: usize, ) -> Self

Creates an InstanceBuffer with a buffer capacity. Buffer size is based on the initial crate::BufferLayout::default_buffer length.

Arguments

• capacity: The capacity of the Buffers instances for future allocation.
• layer_size: The capacity allocated for any future elements per new Buffer Layer.