Struct Framebuffer 

pub struct Framebuffer { /* private fields */ }

SIMD-aligned framebuffer for efficient pixel operations.

The pixel buffer is aligned to 64 bytes for optimal SIMD performance on AVX-512 and other wide SIMD architectures.

SIMD Acceleration

Operations like clear(), fill_rect(), and blend() are SIMD-accelerated using trueno’s automatic backend selection (SSE2/AVX2/AVX512/NEON).

Implementations

impl Framebuffer

pub fn new(width: u32, height: u32) -> Result<Self>

Create a new framebuffer with the given dimensions.

The buffer is aligned to 64 bytes for optimal SIMD performance.

Errors

Returns an error if width or height is zero.

Example

use trueno_viz::framebuffer::Framebuffer;

let fb = Framebuffer::new(800, 600).unwrap();
assert_eq!(fb.width(), 800);
assert_eq!(fb.height(), 600);

pub const fn width(&self) -> u32

Get the width in pixels.

pub const fn height(&self) -> u32

Get the height in pixels.

pub const fn stride(&self) -> usize

Get the stride (row width in bytes, including any padding).

pub const fn pixel_count(&self) -> usize

Get the total number of pixels.

pub fn pixels(&self) -> &[u8]

Get the raw pixel data as a slice.

pub fn pixels_mut(&mut self) -> &mut [u8]

Get the raw pixel data as a mutable slice.

pub fn row(&self, y: u32) -> Option<&[u8]>

Get a row of pixels as a slice.

pub fn row_mut(&mut self, y: u32) -> Option<&mut [u8]>

Get a row of pixels as a mutable slice.

pub fn clear(&mut self, color: Rgba)

Clear the framebuffer to a solid color.

This operation is optimized for SIMD by processing 16 pixels at a time (64 bytes = 16 RGBA pixels on AVX-512).

pub fn fill_rect(&mut self, x: u32, y: u32, w: u32, h: u32, color: Rgba)

Fill a rectangular region with a solid color.

Coordinates are clamped to framebuffer bounds.

pub fn get_pixel(&self, x: u32, y: u32) -> Option<Rgba>

Get the color at a specific pixel coordinate.

Returns None if the coordinates are out of bounds.

pub fn set_pixel(&mut self, x: u32, y: u32, color: Rgba)

Set the color at a specific pixel coordinate.

Does nothing if the coordinates are out of bounds.

pub fn blend_pixel(&mut self, x: u32, y: u32, color: Rgba)

Blend a color at a specific pixel coordinate using alpha blending.

Uses the standard “over” compositing operation: out = src * src_alpha + dst * dst_alpha * (1 - src_alpha)

pub fn blend_over(&mut self, other: &Framebuffer, alpha: f32) -> Result<()>

Blend an entire framebuffer over this one using SIMD-accelerated operations.

Uses trueno’s Vector operations for alpha blending.

Errors

Returns an error if the framebuffers have different dimensions.

pub fn adjust_brightness(&mut self, factor: f32)

Apply a brightness adjustment using SIMD-accelerated operations.

factor of 1.0 is no change, < 1.0 darkens, > 1.0 brightens.

pub fn luminance_stats(&self) -> (f32, f32, f32)

Get statistics about the framebuffer using SIMD-accelerated reduction.

Returns (min_luminance, max_luminance, avg_luminance).

pub fn is_aligned(&self) -> bool

Check if the pixel buffer is properly aligned for SIMD.

pub fn to_compact_pixels(&self) -> Vec<u8>

Get pixel data as a compact buffer without stride padding.

This is useful for encoding to formats like PNG that expect tightly-packed pixel data.

pub fn backend() -> Backend

Get the selected SIMD backend.

Trait Implementations

impl Clone for Framebuffer

fn clone(&self) -> Framebuffer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Framebuffer

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

Formats the value using the given formatter.