Framebuffer depth-artifacts - c++

There's some Artifacts on my FrameBuffer Depth Texture I can't get rid off:
The Code used to init the FrameBuffer:
glGenFramebuffers(1, &fbo);
glBindFramebuffer(GL_FRAMEBUFFER, fbo);
glGenTextures(1, &color);
glBindTexture(GL_TEXTURE_2D, color);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, DISPLAY_WIDTH, DISPLAY_HEIGHT, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glGenTextures(1, &depth);
glBindTexture(GL_TEXTURE_2D, depth);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, DISPLAY_WIDTH, DISPLAY_HEIGHT, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glFramebufferTexture(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, color, 0);
glFramebufferTexture(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT1, depth, 0);
GLuint attachments[2] = { GL_COLOR_ATTACHMENT0, GL_COLOR_ATTACHMENT1};
glDrawBuffers(2, attachments);
if (glCheckFramebufferStatus(GL_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE) {
std::cout << "FBO FEHLER" << std::endl;
}
Code used to draw the FrameBuffer:
shader->bind();
// Bind Textures
glEnable(GL_TEXTURE_2D);
glEnable(GL_BLEND);
glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, color);
glActiveTexture(GL_TEXTURE1);
glBindTexture(GL_TEXTURE_2D, depth);
glUniform1i(shader->getUniform("tex"), 0);
glUniform1i(shader->getUniform("depth"), 1);
glUniformMatrix4fv(shader->getUniform("matrix"), 1, GL_FALSE, glm::value_ptr(ortho));
// Draw Call
glBindBuffer(GL_ARRAY_BUFFER,fbovbo);
glEnableVertexAttribArray(0);
glEnableVertexAttribArray(1);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 20, (void*)0);
glVertexAttribPointer(1, 2, GL_FLOAT, GL_FALSE, 20, (void*)12);
glDrawArrays(GL_TRIANGLE_STRIP,0,4);
glBindBuffer(GL_ARRAY_BUFFER, 0);
// Unbind
shader->unbind();//*/
FragmentShader of the actual rendering:
#version 330 core
layout(location = 0) out vec4 out_color;
layout(location = 1) out vec4 out_depth;
in float temp;
void main(void){
out_color = vec4(1,0.0,0.0,1.0);
out_depth = vec4(gl_FragCoord.z);
}
FragmentShader of the FrameBuffer rendering:
#version 330 core
in vec2 fuv;
out vec4 color;
uniform sampler2D tex;
uniform sampler2D depth;
void main(){
vec4 d = texture2D(depth, fuv);
gl_FragDepth = d.a;
vec4 c = texture2D(tex,fuv);
if(c.a<0.1){
discard;
}
color = c;
//color = vec4(texture2D(depth, fuv).zzz,1.0);
//color.a = 0.8;
//color = vec4(1,0,0,0.5);
}
The red "Mesh" is behind the brown surface, but its borders still appear. thats the problem

The problem might come from the limited precision of the depth buffer. Basically, the range between near and farplane has to be represented by the depth-buffer. In your case, the depth-buffer has 8 bit, thus there are only 256 different depth values that can be represented. When two objects are closer together than this, they will be mapped to the same value in the depth buffer.
To overcome this, you can try to increase the precision of the depth buffer, for example, by using
glGenTextures(1, &depth);
glBindTexture(GL_TEXTURE_2D, depth);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA16, DISPLAY_WIDTH, DISPLAY_HEIGHT, 0, GL_RGBA, GL_UNSIGNED_SHORT, NULL);
Side-note: I'm not sure why the depth texture has four channels. If you don't need them, I would change this to a one channel format.

Related

Framebuffer texture coming out white (fragment shader doesn't affect it)

So I have an fbo and trying to output a depth texture from the light source perspective. Unfortunately the depth texture is coming out pure white even if I hard code black in the frag shader.
This is my frame buffer initialization
//Render frame to a texture
m_FrameBuffer = 0;
glGenFramebuffers(1, &m_FrameBuffer);
glBindFramebuffer(GL_FRAMEBUFFER, m_FrameBuffer); // once frame buffer is bound, must draw to it or black screen
glEnable(GL_TEXTURE_2D);
glGenTextures(1, &m_depthTexture);
glBindTexture(GL_TEXTURE_2D, m_depthTexture);
glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT16, 1024, 1024, 0, GL_DEPTH_COMPONENT, GL_FLOAT, 0);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_COMPARE_FUNC, GL_LEQUAL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_COMPARE_MODE, GL_COMPARE_R_TO_TEXTURE);
glFramebufferTexture(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, m_depthTexture, 0);
glDrawBuffer(GL_NONE);
//Always check that our framebuffer is ok
if (glCheckFramebufferStatus(GL_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE)
printf("frame buffer binding error");
glBindTexture(GL_TEXTURE_2D, 0);
this is my rendering of frame buffer
glBindFramebuffer(GL_FRAMEBUFFER, m_FrameBuffer);
glViewport(0, 0, 1024, 1024);
glEnable(GL_CULL_FACE);
glCullFace(GL_BACK);
glBindVertexArray(m_VAO);
glBindBuffer(GL_ARRAY_BUFFER, m_VBO);
glEnableVertexAttribArray(0);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, 0);
glDrawArrays(GL_TRIANGLES, 0, vertices.size());
glDisableVertexAttribArray(0);
glBindVertexArray(0);
glBindFramebuffer(GL_FRAMEBUFFER, 0);
This is my fragment shader
#version 330 core
// Ouput data
layout(location = 0) out float fragmentdepth;
void main()
{
//fragmentdepth = gl_FragCoord.z;
fragmentdepth = 0;
}
This is my main loop
while (!window.closed())
{
glEnable(GL_DEPTH_TEST);
glEnable(GL_CULL_FACE);
glDepthFunc(GL_LESS);
window.clear();
CameraControls();
depth.enable();
//depthMVP = depthProjectionMatrix * depthViewMatrix * depthModelMatrix;
depth.setUniformMat4("projection", projection);
depth.setUniformMat4("view", camera);
depth.setUniformMat4("model", shape1->modelMatrix);
shape1->RenderToTexture();
depth.disable();
window.clear();
basic.enable();
basic.setUniformMat4("proj", projection);
//basic.setUniform3f("light_Pos", lightPos);
basic.setUniformMat4("view", camera);
basic.setUniformMat4("model", shape1->modelMatrix);
basic.setUniformMat4("DepthBiasMVP", biasMatrix);
//NEED TO CHANGE THIS TO BIAS MATRIX X DEPTHMVP
//MVP = projection * camera * shape1->modelMatrix;
//basic.setUniformMat4("MVP", MVP);
shape1->Render(basic);
basic.disable();
window.update();
}
I am following this tutorial http://www.opengl-tutorial.org/intermediate-tutorials/tutorial-16-shadow-mapping/
Just answering my own question after looking at my old questions that weren't resolved. If anyone is wondering, my problem was that I had fused two tutorials into one and one of them had different near and far values for the frustum. The framebuffer was black because the object was too far from the camera.

Load and display texture with OpenGL and OpenCV

I am a bloody newbie to OpenGL and it still looks sometimes like black magic to me. I just want to load and display an texture for testing purposes and I do not want to use the tutorial libraries to really understand what happens. But I just get a black screen and I cannot find the reason for it.
My c++ source code
int main(){
new cpp_logger();
GLint program_rgb_;
Mat img = imread("plane.jpg");
std::string file_path = "plane.jpg";
// Set viewport window size and clear color bit.
glClearColor(0, 0, 0, 0);
glClear(GL_COLOR_BUFFER_BIT);
glEnable(GL_DEPTH_TEST);
glDepthFunc(GL_LESS);
//set viewport
glViewport(0, 0, img.size().width, img.size().height);
// Create a Vertex Buffer Object and copy the vertex data to it
GLuint vbo;
glGenBuffers(1, &vbo);
GLfloat vertices[] = {-1, 1, -1, -1, 1, 1, 1, -1,
0, 0, 0, 1, 1, 0, 1, 1};
glBindBuffer(GL_ARRAY_BUFFER, vbo);
glBufferData(GL_ARRAY_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
//create program and shader
program_rgb_ = glCreateProgram();
const char* vertexShader = read_shader("vertex.glsl"); //simple method to read the shader files
const char* fragmentShader = read_shader("fragment.glsl");
CreateShader(program_rgb_, GL_VERTEX_SHADER, vertexShader); //simple method to link and compile shader programs
CreateShader(program_rgb_, GL_FRAGMENT_SHADER, fragmentShader);
glLinkProgram(program_rgb_);
glUseProgram(program_rgb_);
//use fast 4-byte alignment (default anyway) if possible because using OpenCV
glPixelStorei(GL_UNPACK_ALIGNMENT, (img.step & 3) ? 1 : 4);
//set length of one complete row in data (doesn't need to equal image.cols)
glPixelStorei(GL_UNPACK_ROW_LENGTH_EXT, img.step/img.elemSize());
flip(img, img, 0);
// Load texture
GLuint tex;
glGenTextures(1, &tex);
glBindTexture(GL_TEXTURE_2D, tex);
//set parameters
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB, img.cols, img.rows, 0, GL_RGBA, GL_UNSIGNED_BYTE, img.data);
glUniform1i(glGetUniformLocation(program_rgb_,"tex"), 0);
uint8_t* result = new uint8_t[img.cols * img.rows * 4];
while(true){
GLint pos_location = glGetAttribLocation(program_rgb_, "a_position");
GLint tc_location = glGetAttribLocation(program_rgb_, "a_texCoord");
// Specify the layout of the vertex data
glEnableVertexAttribArray(pos_location);
glVertexAttribPointer(pos_location, 2, GL_FLOAT, GL_FALSE, 0, 0);
//specify color data
glEnableVertexAttribArray(tc_location);
glVertexAttribPointer(tc_location, 2, GL_FLOAT, GL_FALSE, 0, static_cast<float*>(0) + 8);
// Draw a rectangle from the 2 triangles using 6 indices
glDrawArrays(GL_TRIANGLES, 0, 8);
glReadPixels(0, 0, img.rows, img.cols, GL_BGRA_EXT, GL_UNSIGNED_BYTE, (void*)result);
auto result_ = cv::Mat(img.rows, img.cols, CV_8UC4, result);
imshow("img", result_);
if(waitKey(1) >= 0) break;
}
destroyAllWindows();
return EXIT_SUCCESS;
}
My fragment shader
#version 150 core
precision mediump float;
in vec2 texCoord;
uniform sampler2D tex;
void main()
{
gl_FragColor = texture2D(tex, texCoord);
}
my vertex shader
#version 150 core
out vec2 texCoord;
attribute vec2 a_texCoord;
attribute vec4 a_position;
void main()
{
texCoord = a_texCoord;
gl_Position = a_position;
}
Thank you a lot for your help!
EDIT
Sorry for the missunderstanding. I meant: I do not want to use the libraries like SOIL and SFML and I want to go an alternative way to fully understand what actually happens.

Attribute Variable not found in OpenGL, possible optimization

I'm currently working in a deferred shading and I created a class which manages the FBOs and draw the buffers on the screen.
This is how it looks so far:
FBORender::FBORender(float screenWidth, float screenHeight) :
_screenWidth(screenWidth),
_screenHeight(screenHeight),
ProgramManager("defVertexShader.txt", "defFragShader.txt")
{
CreateProgram();
_vbo[0] = 0;
_vbo[1] = 0;
_vao = 0;
BuildQuad();
BuildVAO();
glGenFramebuffers(1, &_fbo);
glGenRenderbuffers(1, &_depthBuffer);
// Bind the depth buffer
glBindRenderbuffer(GL_RENDERBUFFER, _depthBuffer);
glRenderbufferStorage(GL_RENDERBUFFER, GL_DEPTH_COMPONENT24, (int)_screenWidth, (int)_screenHeight);
// Generate and bind the texture for diffuse
glGenTextures(1, &_diffuseBuffer);
glBindTexture(GL_TEXTURE_2D, _diffuseBuffer);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, (int)_screenWidth, (int)_screenWidth, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
// Generate and bind the texture for positions
glGenTextures(1, &_positionBuffer);
glBindTexture(GL_TEXTURE_2D, _positionBuffer);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA32F, (int)_screenWidth, (int)_screenWidth, 0, GL_RGBA, GL_FLOAT, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
// Generate and bind the texture for normals
glGenTextures(1, &_normalBuffer);
glBindTexture(GL_TEXTURE_2D, _normalBuffer);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA16F, (int)_screenWidth, (int)_screenWidth, 0, GL_RGBA, GL_FLOAT, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
// Bind the FBO so that the next operations will be bound to it.
glBindFramebuffer(GL_FRAMEBUFFER, _fbo);
glFramebufferRenderbuffer(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, GL_RENDERBUFFER, _depthBuffer);
// Attach the textures to the FBO
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, _diffuseBuffer, 0);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT1, GL_TEXTURE_2D, _positionBuffer, 0);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT2, GL_TEXTURE_2D, _normalBuffer, 0);
GLenum fboStatus = glCheckFramebufferStatus(GL_FRAMEBUFFER);
if (fboStatus != GL_FRAMEBUFFER_COMPLETE)
{
printf("DeferredLighting::Init: FrameBuffer incomplete: 0x%x\n", fboStatus);
exit(1);
}
glBindFramebuffer(GL_FRAMEBUFFER, 0);
}
FBORender::~FBORender()
{
glDeleteTextures(1, &_normalBuffer);
glDeleteTextures(1, &_positionBuffer);
glDeleteTextures(1, &_diffuseBuffer);
glDeleteFramebuffers(1, &_fbo);
glDeleteVertexArrays(1, &_vao);
glDeleteBuffers(2, _vbo);
}
void FBORender::Start()
{
// Bind the FBO and set the viewport to the proper size
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, _fbo);
glViewport(0, 0, (int)_screenWidth, (int)_screenWidth);
// Clear the render targets
GLenum windowBuffClear[] = { GL_COLOR_ATTACHMENT0, GL_COLOR_ATTACHMENT1, GL_COLOR_ATTACHMENT2 };
glDrawBuffers(3, windowBuffClear);
glClearColor(0.0f, 0.0f, 0.0f, 0.0f);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
GLenum windowBuffOpaque[] = { GL_COLOR_ATTACHMENT0, GL_COLOR_ATTACHMENT1, GL_COLOR_ATTACHMENT2 };
glDrawBuffers(3, windowBuffOpaque);
}
void FBORender::Draw(const glm::mat4 &Projection, const glm::mat4 &ModelView)
{
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
_ModelViewProjection = Projection * ModelView;
_ModelView = ModelView;
glDisable(GL_DEPTH_TEST);
Bind();
glActiveTexture(GL_TEXTURE2);
glBindTexture(GL_TEXTURE_2D, _normalBuffer);
glActiveTexture(GL_TEXTURE1);
glBindTexture(GL_TEXTURE_2D, _positionBuffer);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, _diffuseBuffer);
glBindVertexArray(_vao);
LoadUniformVariables();
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_INT, 0);
glBindVertexArray(0);
UnBind();
}
void FBORender::Stop()
{
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, 0);
}
void FBORender::BuildQuad()
{
_coordinates[0] = 1.0f;
_coordinates[1] = 1.0f;
_coordinates[2] = 0.0f;
_coordinates[3] = -1.0f;
_coordinates[4] = 1.0f;
_coordinates[5] = 0.0f;
_coordinates[6] = 1.0;
_coordinates[7] = -1.0f;
_coordinates[8] = 0.0f;
_coordinates[9] = -1.0f;
_coordinates[10] = -1.0f;
_coordinates[11] = 0.0f;
_uv[0] = 1.0f;
_uv[1] = 1.0f;
_uv[2] = 0.0f;
_uv[3] = 1.0f;
_uv[4] = 1.0f;
_uv[5] = 0.0f;
_uv[6] = 0.0f;
_uv[7] = 0.0f;
}
void FBORender::BuildVAO()
{
// Generate and bind the vertex array object
glGenVertexArrays(1, &_vao);
glBindVertexArray(_vao);
// Generate and bind the vertex buffer object
glGenBuffers(2, _vbo);
glBindBuffer(GL_ARRAY_BUFFER, _vbo[0]);
glBufferData(GL_ARRAY_BUFFER, 12 * sizeof(float), _coordinates, GL_STATIC_DRAW);
glBindBuffer(GL_ARRAY_BUFFER, _vbo[1]);
glBufferData(GL_ARRAY_BUFFER, 8 * sizeof(float), _uv, GL_STATIC_DRAW);
LoadAttributeVariables();
glBindVertexArray(0);
}
void FBORender::LoadUniformVariables()
{
// OpenGL Matrices
GLuint ModelViewProjection_location = glGetUniformLocation(GetProgramID(), "mvMatrix");
glUniformMatrix4fv(ModelViewProjection_location, 1, GL_FALSE, glm::value_ptr(_ModelView));
// Texture buffers
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, _diffuseBuffer);
GLint Diffuse_location = glGetUniformLocation(GetProgramID(), "tDiffuse");
glUniform1i(Diffuse_location, 0);
glActiveTexture(GL_TEXTURE1);
glBindTexture(GL_TEXTURE_2D, _positionBuffer);
GLint Position_location = glGetUniformLocation(GetProgramID(), "tPosition");
glUniform1i(Position_location, 0);
glActiveTexture(GL_TEXTURE2);
glBindTexture(GL_TEXTURE_2D, _normalBuffer);
GLint Normal_location = glGetUniformLocation(GetProgramID(), "tNormals");
glUniform1i(Normal_location, 0);
}
void FBORender::LoadAttributeVariables()
{
// Vertex Attributes
GLuint VertexPosition_location = glGetAttribLocation(GetProgramID(), "vPosition");
glEnableVertexAttribArray(VertexPosition_location);
glBindBuffer(GL_ARRAY_BUFFER, _vbo[0]);
glVertexAttribPointer(VertexPosition_location, 3, GL_FLOAT, GL_FALSE, 0, 0);
GLuint TextureCoord_Location = glGetAttribLocation(GetProgramID(), "uvCoord");
printf("%d \n", TextureCoord_Location);
glEnableVertexAttribArray(TextureCoord_Location);
glBindBuffer(GL_ARRAY_BUFFER, _vbo[1]);
glVertexAttribPointer(TextureCoord_Location, 2, GL_FLOAT, GL_FALSE, 0, 0);
}
And those are my shaders:
#version 410 core
uniform mat4 mvMatrix;
in vec4 vPosition;
in vec2 uvCoord;
smooth out vec2 texCoord;
void main(void)
{
texCoord = uvCoord;
gl_Position = vPosition * mvMatrix;
}
#version 410 core
uniform sampler2D tDiffuse;
uniform sampler2D tPosition;
uniform sampler2D tNormals;
in vec2 texCoord;
out vec4 fragColor;
void main( void )
{
vec4 image = texture( tDiffuse, texCoord.st );
vec4 position = texture( tPosition, texCoord.st );
vec4 normal = texture( tNormals, texCoord.st );
fragColor.xyz = vec3(0.5f, 0.5f, 0.5f);
}
The problems is that when I try to set the attribute variables the one called uvCoord is not found. I think this may be due to an optimization, however, if that's so, what am I'm doing wrong or how should I do it? The uvCoord is used in the fragment shader.
The uvCoord is used in the fragment shader.
No it is not. Your output is a constant color. As a result, the texture fetches before are all eliminated by a decent compiler, as is the texCoord varying. This ultimately results in the elimination of the input attribute, which simply does not affect the output of the program in any way. This is allowed by the GL spec. Your attribute is just not considered active, but only active attributes have locations.

OpenGL Render To Texture (To screen) blank

I'm drawing a lot of points to the screen, and I am trying to get them to render to a texture for post processing before then rendering the texture back to the screen. Currently I'm trying to just pass through and get the texture rendering but it appears to be rendering nothing.
My fragment shader works (without render to texture - used bellow in mEllipseTextureProgram) and the only change I have made is to include
out vec3 color;
to render to the texture itself (and obviously change to color= rather than gl_FragColor=). The second program referenced below (mScreenProgram) is as follows:
Vertex Shader:
#version 330 compatibility
in vec2 vUV;
out vec2 UV;
void main()
{
gl_Position = gl_Vertex;
UV = vUV;
}
Fragment Shader:
#version 330 core
in vec2 UV;
out vec3 color;
uniform sampler2D renderedTexture;
void main(){
color = texture( renderedTexture, UV ).xyz;
}
I set up my render to texture stuff like this:
glGenFramebuffers(1, &mGaussianFrameBuffer);
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, mGaussianFrameBuffer);
glGenTextures(1, &mGaussianRenderTexture);
glBindTexture(GL_TEXTURE_2D, mGaussianRenderTexture);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA,
1024,
768,
0, GL_RGBA, GL_UNSIGNED_BYTE,
NULL);
glBindTexture(GL_TEXTURE_2D, 0);
glFramebufferTexture(GL_DRAW_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, mGaussianRenderTexture, 0);
glGenTextures(1, &mGaussianDepthBuffer);
glBindTexture(GL_TEXTURE_2D, mGaussianDepthBuffer);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT);
glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT,
1024,
768,
0, GL_DEPTH_COMPONENT, GL_FLOAT,
NULL);
glBindTexture(GL_TEXTURE_2D, 0);
glFramebufferTexture(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, mGaussianDepthBuffer, 0);
which returns a successful GL_FRAMEBUFFER_COMPLETE.
Here are the three concerning rendering functions:
void Draw()
{
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
//gluPerspective(45, 1024.0/768.0, 3.0, 20000);
perspectiveGL(45, 1024.0/768.0, 1.0, 20000);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glPushMatrix();
gluLookAt(cameraData[0],cameraData[1], cameraData[2],
cameraData[3],cameraData[4], cameraData[5],
0,1,0);
glEnable(GL_DEPTH_TEST);
glClearColor( 0.2f, 0.2f, 0.9f, 0.0f );
glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
// OTHER STUFF POTENTIALLY DRAW HERE IN DIFFERENT WAYS
glUseProgram(mEllipseTextureProgram);
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, mGaussianFrameBuffer);
GLuint attachments[1] = {GL_COLOR_ATTACHMENT0};
glDrawBuffers(1, attachments);
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
DrawEllipseToTexture();
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, 0);
glUseProgram(mScreenProgram);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
DrawTextureToScreen();
}
void DrawEllipseToTexture()
{
glEnableClientState(GL_VERTEX_ARRAY);
glBindBuffer(GL_ARRAY_BUFFER, mVBO);
glVertexPointer(3, GL_FLOAT, 0, 0);
glEnable(GL_PROGRAM_POINT_SIZE);
glEnable(GL_POINT_SPRITE);
glEnable( GL_TEXTURE_2D );
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, mBMP);
glProgramUniform1i(mEllipseTextureProgram, mTextureLocation, 0);
glBindBuffer(GL_ARRAY_BUFFER, mUV);
glEnableVertexAttribArray(mTexCoordLocation);
glVertexAttribPointer(mTexCoordLocation, 2, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, mSpacial);
glEnableVertexAttribArray(mSpacialLocation);
glVertexAttribPointer(mSpacialLocation, 1, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, mNormals);
glEnableVertexAttribArray(mNormalLocation);
glVertexAttribPointer(mNormalLocation, 3, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, mVerticalSpat);
glEnableVertexAttribArray(mMajorLocation);
glVertexAttribPointer(mMajorLocation, 3, GL_FLOAT, GL_FALSE, 0, 0);
glBindBuffer(GL_ARRAY_BUFFER, mHorizontalSpat);
glEnableVertexAttribArray(mMinorLocation);
glVertexAttribPointer(mMinorLocation, 3, GL_FLOAT, GL_FALSE, 0, 0);
glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA);
glEnable( GL_BLEND );
glDrawArrays(GL_POINTS, 0, mNumberPoints);
glBindBuffer(GL_ARRAY_BUFFER, 0);
glDisable( GL_TEXTURE_2D );
glDisableVertexAttribArray(mSpacialLocation);
glDisable(GL_POINT_SPRITE);
glDisable( GL_POINT_SMOOTH );
glDisable(GL_PROGRAM_POINT_SIZE);
glDisableClientState(GL_VERTEX_ARRAY);
}
void DrawTextureToScreen()
{
glEnableClientState(GL_VERTEX_ARRAY);
glEnable( GL_TEXTURE_2D );
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, mGaussianRenderTexture);
glProgramUniform1i(mScreenProgram, mGaussianTextureLocation, 0);
GLfloat vertices[] = { -1, -1, 2,
1, -1, 2,
1, 1, 2,
-1, 1, 2 };
GLfloat uv[] = { 0, 0,
1, 0,
1, 1,
0, 1,};
glVertexPointer(3, GL_FLOAT, 0, vertices);
glEnableVertexAttribArray(mGaussianUV);
glVertexAttribPointer(mGaussianUV, 2, GL_FLOAT, GL_FALSE, 0, uv);
glDrawArrays(GL_TRIANGLES, 0, 4);
glDisable ( GL_TEXTURE_2D );
glDisableClientState(GL_VERTEX_ARRAY);
}
I still don't know what was wrong with my solution above (to which I tried a lot of variations) - but following this tutorial got it working: Tutorial
I usually use `out vec4 color', I am not sure whether it will work after you setting your texture type is GL_RGBA and then return vec3 for each fragment.
And I didnt see you set glViewport() in your code which means there is no specify affine transformation to mapping the screen position (x,y) from NDC to Screen space. If you don't have a viewport during fbo render, that definitely gives you nothing.
I am same as you struggling in the similar situation. Read the functions you don't know from GL documentation instead of blog is a much better way to understand things correctly...
: )

opengl shader problems with mrt

I have two textures with same size attached to one fbo, and I want to render another texture into one of the attached textures, and I want to render a blue shape of the figure of the input texture to the other fbo attached texture.
the (updated) vertex shader:
#version 330
in vec2 coord;
out vec2 f_coord;
void main() {
gl_Position = vec4(coord, 0.0, 1.0);
f_coord = (1.0+coord)/2.0;
}
the (updated) fragment shader:
#version 330
uniform sampler2D my_texture;
in vec2 f_coord;
out vec4 FragData[2];
void main() {
if(texture(my_texture,f_coord).a == 0.0)
discard;
FragData[0]=texture(my_texture,f_coord);
FragData[1]=vec4(0.5, 0.5, 1.0, 1.0);
}
With this fragment shader everything works fine. The texture and the blue stamp of the texture contures are rendered properly.
Here is a screenshot:
http://s14.directupload.net/file/d/3487/kvkqdbl7_png.htm
But if I want to render the texture into gl_FragData[1] and the blue conture stamp into gl_FragData[0]
#version 330
uniform sampler2D my_texture;
in vec2 f_coord;
out vec4 FragData[2];
void main() {
if(texture(my_texture,f_coord).a == 0.0)
discard;
FragData[0]=vec4(0.5, 0.5, 1.0, 1.0);
FragData[1]=texture(my_texture,f_coord);
}
than I get two totally blue textures. The strange thing is: Both target textures are totally blue, but I have set the glViewport just to a part of the texture.
Here a screenshot:
http://s1.directupload.net/file/d/3487/9loqibdc_png.htm
The fbo attachments have the half screen size
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, glutGet(GLUT_WINDOW_WIDTH)/2, glutGet(GLUT_WINDOW_HEIGHT), 0, GL_RGBA, GL_BYTE, 0);
So I can render both textures to the screen for checking.
This is how I created the framebufferobject
void init_fbo()
{
glGenTextures(1, &fbo.texture0);
glBindTexture(GL_TEXTURE_2D, fbo.texture0);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, glutGet(GLUT_WINDOW_WIDTH)/2, glutGet(GLUT_WINDOW_HEIGHT), 0, GL_RGBA, GL_BYTE, 0);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glGenTextures(1, &fbo.texture1);
glBindTexture(GL_TEXTURE_2D, fbo.texture1);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, glutGet(GLUT_WINDOW_WIDTH)/2, glutGet(GLUT_WINDOW_HEIGHT), 0, GL_RGBA, GL_BYTE, 0);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glBindTexture(GL_TEXTURE_2D, 0);
glGenFramebuffers(1, &fbo.fbo);
glBindFramebuffer(GL_FRAMEBUFFER, fbo.fbo);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D,
fbo.texture0, 0);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT1, GL_TEXTURE_2D,
fbo.texture1, 0);
glBindFramebuffer(GL_FRAMEBUFFER, 0);
}
I use those global struct variables
struct
{
GLuint fbo;
GLuint texture0;
GLuint texture1;
} fbo;
struct
{
GLuint program;
GLint uni_texture;
GLint att_coord;
} shader_fbo;
struct
{
GLuint program;
GLint uni_texture;
GLint att_coord;
} shader_screen;
And this is my render function
void render()
{
GLenum buffers[2];
GLfloat vertices[] = { -1.0, -1.0,
-1.0, 1.0,
1.0, -1.0,
1.0, 1.0 };
GLubyte indices[] = { 0, 1, 2,
1, 2, 3 };
/*render to mrt fbo*/
glUseProgram(shader_fbo.program);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, texture.texID);
glUniform1i(shader_fbo.uni_texture, 0);
glEnableVertexAttribArray(shader_fbo.att_coord);
glVertexAttribPointer(shader_fbo.att_coord,
2,
GL_FLOAT,
GL_FALSE,
0,
vertices);
glBindFramebuffer(GL_FRAMEBUFFER, fbo.fbo);
buffers[0] = GL_COLOR_ATTACHMENT0;
buffers[1] = GL_COLOR_ATTACHMENT1;
glDrawBuffers(2, buffers);
glViewport(0, 0,
texture.width, texture.height);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, texture.texID);
glUniform1i(shader_fbo.uni_texture, 0);
glClear(GL_COLOR_BUFFER_BIT);
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_BYTE, indices);
glDisableVertexAttribArray(shader_fbo.att_coord);
glBindFramebuffer(GL_FRAMEBUFFER, 0);
glBindTexture(GL_TEXTURE_2D, 0);
/*render both textures now to screen*/
glUseProgram(shader_screen.program);
glEnableVertexAttribArray(shader_screen.att_coord);
glVertexAttribPointer(shader_screen.att_coord,
2,
GL_FLOAT,
GL_FALSE,
0,
vertices);
/*render the first texture to the left side of the screen*/
glViewport(0, 0,
glutGet(GLUT_WINDOW_WIDTH)/2, glutGet(GLUT_WINDOW_HEIGHT));
glClear(GL_COLOR_BUFFER_BIT);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, fbo.texture0);
glUniform1i(shader_screen.uni_texture, 0);
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_BYTE, indices);
/*render the second texture to the right side of the screen*/
glViewport(glutGet(GLUT_WINDOW_WIDTH)/2, 0,
glutGet(GLUT_WINDOW_WIDTH)/2, glutGet(GLUT_WINDOW_HEIGHT));
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, fbo.texture1);
glUniform1i(shader_screen.uni_texture, 0);
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_BYTE, indices);
glutSwapBuffers();
}