How do OpenGL fragment shaders know what pixel to sample in a texture?

Question

So I've got a triangle:

And I've got a vertex shader:

uniform mat4 uViewProjection;
attribute vec3 aVertexPosition;
attribute vec2 aTextureCoords;
varying vec2 vTextureCoords;

void main(void) {
  vTextureCoords = aTextureCoords;
  gl_Position = uViewProjection * vec4(aVertexPosition, 1.0);
}

And I've got a fragment shader:

precision mediump float;
uniform sampler2D uMyTexture;
varying vec2 vTextureCoords;

void main(void) {
  gl_FragColor = texture2D(uMyTexture, vTextureCoords);
}

And I feed in three sets of vertices and UVs, interleaved:

# x,  y,    z,   s,   t
0.0,  1.0,  0.0, 0.5, 1.0
-1.0, -1.0, 0.0, 0.0, 0.0 
1.0, -1.0,  0.0, 1.0, 0.0

How does the fragment shader know to draw pixel A differently from pixel B? What changes?

You fed in a texture right? Why would you expect them to be the same or are you asking how shaders work on the GPU? — Jesus Ramos
@JesusRamos I'm asking how shaders work. I don't understand what's changing and who's changing it between shader passes. — a paid nerd
Check out @genpfault's answer. I was going to say the same thing but he beat me to it :) — Jesus Ramos
Check this great answer stackoverflow.com/q/14246604/187752, when OpenGL might be sampling values you are not expecting. — Kimi

genpfault genpfault · Accepted Answer · 2013-01-31T18:57:05

As I understand it the rasterization stage of the GL pipeline interpolates vTextureCoords across the triangle face, running the fragment shader on each pixel with the interpolated value.

How do OpenGL fragment shaders know what pixel to sample in a texture?

1 Answers