#[macro_use]
extern crate glium;
#[allow(unused_imports)]
use glium::{glutin, Surface};
use glium::index::PrimitiveType;
mod support;
fn main() {
println!("This example demonstrates how to draw a lot of sprites in an efficient manner. \n\n\
Instead of drawing sprites one by one, it writes the list of sprites in a buffer \
and draws everything at once. Textures are accessed though a Texture2dArray to \
avoid the problem of binding textures one by one.\n\n\
Performances are limited by the synchronization required to write on the color \
buffer. Enabling depth test would likely increase the framerate.\n");
const SPRITES_COUNT: usize = 1024;
println!("Number of sprites: {}", SPRITES_COUNT);
let event_loop = glutin::event_loop::EventLoop::new();
let wb = glutin::window::WindowBuilder::new();
let cb = glutin::ContextBuilder::new();
let display = glium::Display::new(wb, cb, &event_loop).unwrap();
let texture = {
let images = (0 .. 64).map(|_| {
let color1: (f32, f32, f32) = (rand::random(), rand::random(), rand::random());
let color2: (f32, f32, f32) = (rand::random(), rand::random(), rand::random());
vec![vec![color1], vec![color2]]
}).collect::<Vec<_>>();
glium::texture::Texture2dArray::new(&display, images).unwrap()
};
let (mut vertex_buffer, index_buffer) = {
#[derive(Copy, Clone)]
struct Vertex {
i_position: [f32; 2],
i_tex_id: u32,
}
implement_vertex!(Vertex, i_position, i_tex_id);
let mut vb: glium::VertexBuffer<Vertex> = glium::VertexBuffer::empty_dynamic(&display,
SPRITES_COUNT * 4).unwrap();
let mut ib_data = Vec::with_capacity(SPRITES_COUNT * 6);
for (num, sprite) in vb.map().chunks_mut(4).enumerate() {
let tex_id: u32 = rand::random();
let tex_id = tex_id % texture.get_array_size().unwrap();
let position: (f32, f32) = (rand::random(), rand::random());
let position: (f32, f32) = (position.0 * 2.0 - 1.0, position.1 * 2.0 - 1.0);
sprite[0].i_position[0] = position.0 - 0.1;
sprite[0].i_position[1] = position.1 + 0.1;
sprite[0].i_tex_id = tex_id;
sprite[1].i_position[0] = position.0 + 0.1;
sprite[1].i_position[1] = position.1 + 0.1;
sprite[1].i_tex_id = tex_id;
sprite[2].i_position[0] = position.0 - 0.1;
sprite[2].i_position[1] = position.1 - 0.1;
sprite[2].i_tex_id = tex_id;
sprite[3].i_position[0] = position.0 + 0.1;
sprite[3].i_position[1] = position.1 - 0.1;
sprite[3].i_tex_id = tex_id;
let num = num as u16;
ib_data.push(num * 4);
ib_data.push(num * 4 + 1);
ib_data.push(num * 4 + 2);
ib_data.push(num * 4 + 1);
ib_data.push(num * 4 + 3);
ib_data.push(num * 4 + 2);
}
(vb, glium::index::IndexBuffer::new(&display, PrimitiveType::TrianglesList, &ib_data).unwrap())
};
let program = program!(&display,
140 => {
vertex: "
#version 140
in vec2 i_position;
in uint i_tex_id;
out vec2 v_tex_coords;
flat out uint v_tex_id;
void main() {
gl_Position = vec4(i_position, 0.0, 1.0);
if (gl_VertexID % 4 == 0) {
v_tex_coords = vec2(0.0, 1.0);
} else if (gl_VertexID % 4 == 1) {
v_tex_coords = vec2(1.0, 1.0);
} else if (gl_VertexID % 4 == 2) {
v_tex_coords = vec2(0.0, 0.0);
} else {
v_tex_coords = vec2(1.0, 0.0);
}
v_tex_id = i_tex_id;
}
",
fragment: "
#version 140
uniform sampler2DArray tex;
in vec2 v_tex_coords;
flat in uint v_tex_id;
out vec4 f_color;
void main() {
f_color = texture(tex, vec3(v_tex_coords, float(v_tex_id)));
}
"
},
110 => {
vertex: "
#version 110
in vec2 i_position;
in uint i_tex_id;
varying vec2 v_tex_coords;
flat varying uint v_tex_id;
void main() {
gl_Position = vec4(i_position, 0.0, 1.0);
if (gl_VertexID % 4 == 0) {
v_tex_coords = vec2(0.0, 1.0);
} else if (gl_VertexID % 4 == 1) {
v_tex_coords = vec2(1.0, 1.0);
} else if (gl_VertexID % 4 == 2) {
v_tex_coords = vec2(0.0, 0.0);
} else {
v_tex_coords = vec2(1.0, 0.0);
}
v_tex_id = i_tex_id;
}
",
fragment: "
#version 110
uniform sampler2DArray tex;
varying vec2 v_tex_coords;
flat varying uint v_tex_id;
void main() {
gl_FragColor = texture2DArray(tex, vec3(v_tex_coords, float(v_tex_id)));
}
"
},
100 => {
vertex: "
#version 100
attribute lowp vec2 i_position;
attribute uint i_tex_id;
varying lowp vec2 v_tex_coords;
flat varying uint v_tex_id;
void main() {
gl_Position = vec4(i_position, 0.0, 1.0);
if (gl_VertexID % 4 == 0) {
v_tex_coords = vec2(0.0, 1.0);
} else if (gl_VertexID % 4 == 1) {
v_tex_coords = vec2(1.0, 1.0);
} else if (gl_VertexID % 4 == 2) {
v_tex_coords = vec2(0.0, 0.0);
} else {
v_tex_coords = vec2(1.0, 0.0);
}
v_tex_id = i_tex_id;
}
",
fragment: "
#version 100
uniform sampler2DArray tex;
varying lowp vec2 v_tex_coords;
flat varying uint v_tex_id;
void main() {
gl_FragColor = texture2DArray(tex, vec3(v_tex_coords, float(v_tex_id)));
}
"
},
).unwrap();
support::start_loop(event_loop, move |events| {
{
let mut mapping = vertex_buffer.map();
for sprite in mapping.chunks_mut(4) {
let mv: (f32, f32) = (rand::random(), rand::random());
let mv = (mv.0 * 0.01 - 0.005, mv.1 * 0.01 - 0.005);
sprite[0].i_position[0] += mv.0;
sprite[0].i_position[1] += mv.1;
sprite[1].i_position[0] += mv.0;
sprite[1].i_position[1] += mv.1;
sprite[2].i_position[0] += mv.0;
sprite[2].i_position[1] += mv.1;
sprite[3].i_position[0] += mv.0;
sprite[3].i_position[1] += mv.1;
}
}
let ib_slice = index_buffer.slice(0 .. SPRITES_COUNT * 6).unwrap();
let mut target = display.draw();
target.clear_color(0.0, 0.0, 0.0, 0.0);
target.draw(&vertex_buffer, &ib_slice,
&program, &uniform! { tex: &texture }, &Default::default()).unwrap();
target.finish().unwrap();
let mut action = support::Action::Continue;
for event in events {
match event {
glutin::event::Event::WindowEvent { event, .. } => match event {
glutin::event::WindowEvent::CloseRequested => action = support::Action::Stop,
_ => ()
},
_ => (),
}
};
action
});
}