GLSL: sampler3D in vertex shader, texture appears blank - opengl

EDIT: Think I've narrowed down the problem. Skip to the running section.
I'm trying to sample a 3d texture in my vertex shader, I'm going to use the texel values as corner value in Marching Cubes. The issue I'm having is that no matter what method I use to sample it, I always get (0,0,0,0). I've tried using texelFetch and texture3D and neither seem to work.
I'm also using transform feedback, but as far as I'm aware that shouldn't cause this issue.
Shader setup:
glEnable(GL_TEXTURE_3D);
Shader vertListTriangles(GL_VERTEX_SHADER_ARB);
vertListTriangles.setSource(lst_tri_vert); //Util to load from file.
vertListTriangles.compile();
vertListTriangles.errorCheck(); //Prints errors to console if they exist - shader compiles fine.
Shader geomListTriangles(GL_GEOMETRY_SHADER_ARB);
geomListTriangles.setSource(lst_tri_geom); //Util to load from file
geomListTriangles.compile();
geomListTriangles.errorCheck(); //Prints errors to console if they exist - shader compiles fine.
program.attach(vertListTriangles);
program.attach(geomListTriangles);
//Setup transform feedback varyings, also works as expected.
const GLchar* varyings1[1];
varyings1[0] = "gTriangle";
glTransformFeedbackVaryings(program.getID(), 1, varyings1, GL_INTERLEAVED_ATTRIBS);
program.link();
program.checkLink(); //Prints link errors to console - program links fine aparently.
Texture setup:
glBindTexture(GL_TEXTURE_3D, textureID);
errorCheck("texture bind"); //<- Detects GL errors, I actually get a GL_INVALID_OPERATION here, not sure if its the cause of the problem though as all subsuquent binds go smoothly.
if(!(glIsTexture(textureID)==GL_TRUE)) consolePrint("Texture Binding Failed."); //Oddly the texture never registers as failed despite the previous error message.
//Generate Texture
GLfloat volumeData[32768*3];
for(int z = 0; z < 32; z++)
{
for(int y = 0; y < 32; y++)
{
for(int x = 0; x < 32; x++)
{
//Set all 1s for testing purposes
volumeData[(x*3)+(y*96)+(z*3072)] = 1.0f;
volumeData[(x*3)+(y*96)+(z*3072)+1] = 1.0f;
volumeData[(x*3)+(y*96)+(z*3072)+2] = 1.0f;
}
}
}
glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MAX_LEVEL, 0);
glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_BASE_LEVEL, 0);
glTexImage3D(GL_TEXTURE_3D, 0, GL_RGB8, 32, 32, 32, 0, GL_RGB,
GL_FLOAT, volumeData);
glBindTexture(GL_TEXTURE_3D, 0);
Running Shader:
EDIT: Here it gets interesting. If I specify an incorrect uniform name or comment out the below lines it appears to work.
program.use();
//Disable Rastering
glEnable(GL_RASTERIZER_DISCARD);
//Input buffer: Initial vertices
glBindBuffer(GL_ARRAY_BUFFER, mInitialDataBuffer);
glEnableVertexAttribArray(0);
glVertexAttribIPointer(0, 1, GL_UNSIGNED_INT, 0, 0); //Initial input is array of uints
//Output buffer: Triangles
glBindBufferBase(GL_TRANSFORM_FEEDBACK_BUFFER, 0, mTriangleBuffer); //Triangle Markers, in the form of uints. NOT actual triangles.
//Texture setup
//If I comment out from here....
GLint sampler = glGetUniformLocation(program.getID(), "densityVol");
glUniform1i(sampler, GL_TEXTURE0);
glActiveTexture(GL_TEXTURE0);
//To here. It appears to work.
glBindTexture(GL_TEXTURE_3D, textureID);
//Just using this to debug texture.
//test is all 1s, so the texture is uploading correctly.
GLfloat test[32768*3];
memset(test, 0, sizeof(test));
glGetTexImage(GL_TEXTURE_3D, 0, GL_RGB, GL_FLOAT, test);
//Transform Feedback and Draw
glBeginTransformFeedback(GL_POINTS);
glDrawArrays(GL_POINTS, 0, 29790);
glEndTransformFeedback();
//Re-enable Rastering and cleanup
glDisable(GL_RASTERIZER_DISCARD);
glDisableVertexAttribArray(0);
glBindBuffer(GL_ARRAY_BUFFER, 0);
My code is a little more spread out in reality, but I hope I managed to edit it into something cohesive. Anyway if I map to the output buffer it does indeed output some information, however it processes as if all the texture data is 0s. I hacked the shader to just output some test results instead but I can't find any evidence the shader is using the texture correctly:
#version 410
#extension GL_EXT_gpu_shader4 : require
layout (location = 0) in int x_y_z;
uniform sampler3D densityVol;
out Voxel
{
/*
Each triangle is the edges it joins. There are 12 edges and so we need 12 bits. 4 For each edge.
There are up to 32 voxels, which means we need 6 bits for each coord, which is 18.
30 bits total.
int format 00xxxxxxyyyyyyzzzzzz111122223333
*/
uint triangles[5];
uint triangleCount;
} vVoxel;
//... Omitted some huge ref tables.
void main()
{
vec4 sample0 = texture3D(densityVol, vec3(0.1,0.1,0.1) );
vec4 sample1 = texture3D(densityVol, vec3(0.9,0.9,0.9) );
vec4 sample2 = texture3D(densityVol, vec3(0.1,0.1,0.9) );
vec4 sample3 = texture3D(densityVol, vec3(0.9,0.9,0.1) );
if(sample0.r > 0.0f)
{
vVoxel.triangles[1] = 1;
}
if(sample1.r > 0.0f)
{
vVoxel.triangles[2] = 2;
}
if(sample2.r > 0.0f)
{
vVoxel.triangles[3] = 3;
}
if(sample3.r > 0.0f)
{
vVoxel.triangles[4] = 4;
}
vVoxel.triangleCount = 5;
}
Not the best designed test, but I didn't want to write something from scratch. If I change the if clauses to if(true) the test outputs correctly. When the shader is compiled as above, the buffer is blank. I'm using a GS for pass through.
Can anyone see an obvious mistake in there? I've been stumped for about 2 hours now and I can't see what I'm doing different from many of the GLSL texturing tutorials.

Okay figured it out.
glUniform1i(sampler, GL_TEXTURE0);
The GL_TEXTURE0 is incorrect here.
glUniform1i(sampler, 0);
Is how it should be.

Related

Cannot create/ sample 3D Texture (Qt/ OpenGL)

I am trying to display a layer in a 3D texture created from some 3D data, but all points sampled are always black (I guess my texture creation/ allocation is failing somehow). It is being rendered on a plane using window coords. I have checked my data, it is a vector with the correct values. glEnable(GL_TEXTURE_3D) has been called earlier. Any clues why this would fail?
Function that creates the texture:
bool VolumeRender::setVolumeData(QOpenGLShaderProgram *program, vector<unsigned short> v, int x, int y, int z){
voxels.resize(v.size(), 0);
cout << "Processing texture" << endl;
unsigned short sMax = 0;
unsigned short sMin = 32768;
for (unsigned int i =0; i< voxels.size(); i++){
sMax = max(sMax, v[i]);
sMin = min(sMin, v[i]);
} for (unsigned int i =0; i< voxels.size(); i++) voxels[i] = (v[i] - sMin)/(float)(sMax-sMin);
cout << "Loading 3D texture" << endl;
gl.glActiveTexture(GL_TEXTURE0);
gl.glGenTextures(1, &volumeTexture);
gl.glBindTexture(GL_TEXTURE_3D, volumeTexture);
gl.glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
gl.glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
gl.glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_BORDER);
gl.glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_BORDER);
gl.glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_WRAP_R, GL_CLAMP_TO_BORDER);
gl.glTexImage3D(GL_TEXTURE_3D, 0, GL_RED, x, y, z, 0, GL_RED, GL_FLOAT, &voxels[0]);
gl.glBindTexture(GL_TEXTURE_3D, 0);
program->bind();
program->setUniformValue("VOXELS", 0);
program->release();
voxelsLoaded = true;
return true;
}
Simple fragment shader:
#version 330 core
uniform sampler3D VOXELS;
uniform vec2 SIZE;
out vec4 color;
void main(){
vec2 coords = (gl_FragCoord.xy - 0.5) / SIZE;
vec3 texcoords = vec3(coords, 0.5);
color = texture(VOXELS, texcoords);
}
glEnable(GL_TEXTURE_…) has no effect when using shaders. It's a relic from the fixed function pipeline era. On the other hand the texture must be actually bound when drawing.
In your code you have
gl.glBindTexture(GL_TEXTURE_3D, 0);
program->bind();
program->setUniformValue("VOXELS", 0);
program->release();
Now since this is in initialization code, it's unclear if you actually understand the consequences of these lines. So let's break it down:
gl.glBindTexture(GL_TEXTURE_3D, 0);
This means that texture 0 (which with shaders is the nil texture, but in old and busted OpenGL-1.0 it actually could be sampled form) is bount to texture unit 0. From there on, when trying to sample from texture unit 0, it will not sample anything.
program->bind();
program->setUniformValue("VOXELS", 0);
program->release();
Set the sampler uniform with name "VOXELS" to sample from texture unit 0. Whatever texture is bound to that texture unit at the moment of calling the draw function, that texture will be sampled from.
Somewhere in your program you're making a draw call. You didn't show us where. But in order for your drawing to actually sample from your texture you have to bind your 3d texture to texture unit 0.
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_3D, volumeTexture);
draw_stuff();

glGetPixels on Offscreen framebuffer opengl

I generate a PointCloud in my program, and now, I want to be able to click on a point in this point cloud rendered to my screen using OpenGL.
In order to do so, I used the trick of giving to each pixel in an offscreen render a colour based on its index in the VBO. I use the same camera for my offscreen render and my onscreen render so they move together, and when I click, I get values of my offscreen render to retrieve the position in the VBO to get the point I clicked on. This is the theory since when I click, I have only (0,0,0). I believe that means my FBO is not well renderer but I'm not sure whether it is that or if the problem comes from somewhere else...
So here are the steps. clicFBO is the FBO I'm using for offscreen render, and clicTextureColorBuf is the texture in which I write in the FBO
glGenFramebuffers(1, &clicFBO);
glBindFramebuffer(GL_FRAMEBUFFER, clicFBO);
glGenTextures(1, &clicTextureColorBuf);
glBindTexture(GL_TEXTURE_2D, clicTextureColorBuf);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB, SCR_WIDTH, SCR_HEIGHT, 0, GL_RGB, GL_UNSIGNED_BYTE, NULL);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, clicTextureColorBuf, 0);
GLenum DrawBuffers[1] = { GL_COLOR_ATTACHMENT0 };
glDrawBuffers(1, DrawBuffers);
After that, I wrote a shader that gives to each point the color of its index in the VBO...
std::vector<cv::Point3f> reconstruction3D; //Will contain the position of my points
std::vector<float> indicesPointsVBO; //Will contain the indexes of each point
for (int i = 0; i < pts3d.size(); ++i) {
reconstruction3D.push_back(pts3d[i].pt3d);
colors3D.push_back(pt_tmp);
indicesPointsVBO.push_back(((float)i / (float)pts3d.size() ));
}
GLuint clicVAO, clicVBO[2];
glGenVertexArrays(1, &clicVAO);
glGenBuffers(2, &clicVBO[0]);
glBindVertexArray(clicVAO);
glBindBuffer(GL_ARRAY_BUFFER, clicVBO[0]);
glBufferData(GL_ARRAY_BUFFER, reconstruction3D.size() * sizeof(cv::Point3f), &reconstruction3D[0], GL_STATIC_DRAW);
glEnableVertexAttribArray(0);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (GLvoid*)0);
glEnable(GL_PROGRAM_POINT_SIZE);
glBindBuffer(GL_ARRAY_BUFFER, clicVBO[1]);
glBufferData(GL_ARRAY_BUFFER, indicesPointsVBO.size() * sizeof(float), &indicesPointsVBO[0], GL_STATIC_DRAW);
glEnableVertexAttribArray(1);
glVertexAttribPointer(1, 1, GL_FLOAT, GL_FALSE, 0, (GLvoid*)0);
glBindBuffer(GL_ARRAY_BUFFER, 0);
glBindVertexArray(0);
glBindFramebuffer(GL_FRAMEBUFFER, 0);
and the vertex shader:
layout (location = 0) in vec3 pos;
layout (location = 1) in float col;
out float Col;
uniform mat4 model;
uniform mat4 view;
uniform mat4 projection;
uniform int pointSize;
void main()
{
gl_PointSize = pointSize;
gl_Position = projection * view * model * vec4(pos, 1.0);
Col = col;
}
And the Fragment:
#version 330 core
layout(location = 0) out vec4 FragColor;
in float Col;
void main()
{
FragColor = vec4(Col, Col, Col ,1.0);
}
And this is how I render this texture:
glm::mat4 view = camera.GetViewMatrix();
glm::mat4 projection = glm::perspective(glm::radians(camera.Zoom), (float)SCR_WIDTH / (float)SCR_HEIGHT, 1.0f, 100.0f);
glBindFramebuffer(GL_FRAMEBUFFER, clicFBO);
clicShader.use();
glDisable(GL_DEPTH_TEST);
glClearColor(1.0f, 1.0f, 1.0f, 1.0f);
glClear(GL_COLOR_BUFFER_BIT);
clicShader.setMat4("projection", projection);
clicShader.setMat4("view", view);
model = glm::mat4();
clicShader.setMat4("model", model);
clicShader.setInt("pointSize", pointSize);
glBindVertexArray(clicVAO);
glDrawArrays(GL_POINTS, 0, (GLsizei)reconstruction3D.size());
glBindFramebuffer(GL_FRAMEBUFFER, 0);
And then, when I click, I Use this piece of Code:
glBindFramebuffer(GL_FRAMEBUFFER, clicFBO);
glReadBuffer(GL_COLOR_ATTACHMENT0);
int width = 11, height = 11;
std::array<GLfloat, 363> arry{ 1 };
glReadPixels(Xpos - 5, Ypos - 5, width, height, GL_RGB, GL_UNSIGNED_BYTE, &arry);
for (int i = 0; i < 363; i+=3) { // It's 3 time the same number anyways for each number
std::cout << arry[i] << " "; // It gives me only 0's
}
std::cout << std::endl << std::endl;
glBindFramebuffer(GL_FRAMEBUFFER, clicFBO);
I know the error might be really stupid but I still have some problems with how OpenGL works.
I put what I thought was necessary to understand the problem (without extending too much), but if you need more code, I can write it too.
I know this is not a question in which you can say Yes or No and it's more like debugging my program, but since I really don't find from where the problem comes from, I'm looking toward someone who can explain to me what I did wrong. I do not necessarily seek the solution itself, but clues that could help me understand where my error is ...
Using a framebuffer object FBO to store a "object identifier" is a cool method. But also want to see the objects, right? Then you must render also to the default frame buffer (let me call it "defFB", which is not a FBO).
Because you need to render to two different targets, you need one of these techniques:
Draw objects twice (e.g. with two glDrawArrays calls), one to the FBO and a second one to the defFB.
Draw to two FBO's images at once and later blit one of then (with colors) to the defFB.
For the first technique you may use a texture attached to a FBO (as you currently do). Or you can use a "Renderbuffer" and draw to it.
The second approach needs a second "out" in the fragment shader:
layout(location = 0) out vec3 color; //GL_COLOR_ATTACHMENT0
layout(location = 1) out vec3 objID; //GL_COLOR_ATTACHMENT1
and setting the two attachments with glDrawBuffers.
For the blit part, read this answer.
Note that both "out" have the same format, vec3 in this example.
A fail in your code is that you set a RGB texture format and also use this format at glReadPixels, but your "out" in the FS is vec4 instead of vec3.
More concerns are:
Check the completeness with glCheckFramebufferStatus
Using a "depth attachment" to the FBO may be needed, even it will not be used for reading.
Disabling the depth test will put all elements if the frame. Your point-picking will select the last drawn, not the nearest.
I found the problem.
There were 2 failures in my code :
The first one is that in OpenGL, there is an Y inversion between the image and the framebuffer. So in order to pick the good point, you have to flip Y using the size of the viewport : I did it like this :
GLint m_viewport[4];
glGetIntegerv(GL_VIEWPORT, m_viewport);
int YposTMP = m_viewport[3] - Ypos - 1;
The second one is the use of
glReadPixels(Xpos - 2, Ypos - 2, width, height, GL_RGB, GL_UNSIGNED_BYTE, &pixels[0]);, the 6th parameter must be GL_FLOAT since the datas i'm returning are float.
Thanks all!
Best regards,
R.S

OpenGL Texture3D FragmentShader Write Into Layers

i want to filter a texture (no mipmapping) and write the filtered levels into a 3D-Texture. I already created a 3D-Texture with following code:
glGenTextures(1, &outTexture);
glBindTexture(GL_TEXTURE_3D, outTexture);
glTexParameterf(GL_TEXTURE_3D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexImage3D(GL_TEXTURE_3D, 0, GL_RGBA8, width, height, layer, 0, GL_RGBA, GL_INT, 0);
glBindTexture(GL_TEXTURE_3D, 0);
and i render my models like so:
glFramebufferTexture3D(GL_FRAMEBUFFER, attachment, GL_TEXTURE_3D, outTexture, 0, layer);
entity.render();
is there a way for a fragment-shader to write into multiple layers at the same time and knowing in which layer you are? Like so?
out_color[layer] = vec4(r,g,b,a);
PS: i need to know in which layer in the fragment shader (some filtering things & please no gl_Layer ... can be only used at OpenGL 4 and i can only use OpenGL 3) ... I could write it into a uniform and render multiple times but if there is a way to go with a one pass ... it would be nice :)
An instance of a fragment shader cannot choose what layer it is writing to. This is chosen for it by the geometry shader that constructed the primitive that generates the fragment in question. So if you want to render to multiple layers, you will need to use layered rendering and GS's.
If your Fragment Shader needs to know what layer it is writing to, then have the Geometry Shader pass that as a per-vertex parameter:
//Geometry shader
flat out int layerIndex;
void main()
{
for(int currLayer = 0; currLayer < numLayers; ++currLayer)
{
for(int currVertex = 0; currVertex < gl_in.length(); ++currVertex)
{
layerIndex = currLayer;
gl_Layer = currLayer;
//Compute other per-vertex outputs
EmitVertex();
}
EndPrimitive();
}
EmitVertex();
}
//Fragment shader
flat in int layerIndex;
void main()
{
//Do stuff with layerIndex;
}

GLSL Render to Texture not working

I'm trying to do a compute pass where I render to a texture that will be used in a draw pass later on. My initial implementation was based on shader storage buffer objects and was working nicely. But I want to apply a computation method that is going to take advantage of the blend hardware of the GPU so I started porting the SSBO implementation to RTT one. Unfortunately the code has stopped working. Now when I read back the texture it is getting wrong values.
Here is my texture and frame buffer setup code:
glGenFramebuffers(1, &m_fbo);
glBindFramebuffer(GL_FRAMEBUFFER, m_fbo);
// Create render textures
glGenTextures(NUM_TEX_OUTPUTS, m_renderTexs);
m_texSize = square_approximation(m_numVertices);
cout << "Textures size: " << glm::to_string(m_texSize) << endl;
GLenum drawBuffers[NUM_TEX_OUTPUTS];
for (int i = 0 ; i < NUM_TEX_OUTPUTS; ++i)
{
glBindTexture(GL_TEXTURE_2D, m_renderTexs[i]);
// 1st 0: level, 2nd 0: no border, 3rd 0: no initial data
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, m_texSize.x, m_texSize.y, 0, GL_RGBA, GL_FLOAT, 0);
// XXX: do we need this?
// Poor filtering. Needed !
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);
// 0: level
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0 + i, GL_TEXTURE_2D, m_renderTexs[i], 0);
drawBuffers[i] = GL_COLOR_ATTACHMENT0 + i;
}
glDrawBuffers(NUM_TEX_OUTPUTS, drawBuffers);
if (glCheckFramebufferStatus(GL_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE)
{
cout << "Error when setting frame buffer" << endl;
// throw exception?
}
glBindFramebuffer(GL_FRAMEBUFFER, 0);
And this is the code to start the compute pass:
m_shaderProgram.use();
// setup openGL
glPolygonMode(GL_FRONT_AND_BACK, GL_LINE);
glDisable(GL_CULL_FACE);
glDisable(GL_DEPTH_TEST);
glViewport(0, 0, m_texSize.x, m_texSize.y); // setup viewport (equal to textures size)
// make a single patch have the vertex, the bases and the neighbours
glPatchParameteri(GL_PATCH_VERTICES, m_maxNeighbours + 5);
// Wait all writes to shader storage to finish
glMemoryBarrier(GL_SHADER_STORAGE_BARRIER_BIT);
glUniform1i(m_shaderProgram.getUniformLocation("curvTex"), m_renderTexs[2]);
glUniform2i(m_shaderProgram.getUniformLocation("size"), m_texSize.x, m_texSize.y);
glUniform2f(m_shaderProgram.getUniformLocation("vertexStep"), (umax - umin)/divisoes,
(vmax-vmin)/divisoes);
// Bind buffers
glBindFramebuffer(GL_FRAMEBUFFER, m_fbo);
glBindBuffer(GL_ARRAY_BUFFER, m_vbo);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, m_ibo);
glBindBufferBase(GL_UNIFORM_BUFFER, m_mvp_location, m_mvp_ubo);
// Make textures active
for (int i = 0; i < NUM_TEX_OUTPUTS; ++i)
{
glActiveTexture(GL_TEXTURE0 + i);
glBindTexture(GL_TEXTURE_2D, m_renderTexs[i]);
}
// no need to pass index array 'cause ibo is bound already
glDrawElements(GL_PATCHES, m_numElements, GL_UNSIGNED_INT, 0);
I then read back the textures using the following:
bool readTex(GLuint tex, void *dest)
{
glBindTexture(GL_TEXTURE_2D, tex);
glGetTexImage(GL_TEXTURE_2D, 0, GL_RGBA, GL_FLOAT, dest);
glBindTexture(GL_TEXTURE_2D, 0);
// TODO: check glGetTexImage return values for error
return true;
}
for (int i = 0; i < NUM_TEX_OUTPUTS; ++i)
{
if (m_tensors[i] == NULL) {
m_tensors[i] = new glm::vec4[m_texSize.x*m_texSize.y];
}
memset(m_tensors[i], 0, m_texSize.x*m_texSize.y*sizeof(glm::vec4));
readTex(m_renderTexs[i], m_tensors[i]);
}
Finally, the fragment shader code is:
#version 430
#extension GL_ARB_shader_storage_buffer_object: require
layout(pixel_center_integer) in vec4 gl_FragCoord;
layout(std140, binding=6) buffer EvalBuffer {
vec4 evalDebug[];
};
uniform ivec2 size;
in TEData {
vec4 _a;
vec4 _b;
vec4 _c;
vec4 _d;
vec4 _e;
};
layout(location = 0) out vec4 a;
layout(location = 1) out vec4 b;
layout(location = 2) out vec4 c;
layout(location = 3) out vec4 d;
layout(location = 4) out vec4 e;
void main()
{
a= _a;
b= _b;
c= _c;
d= _d;
e= _e;
evalDebug[gl_PrimitiveID] = gl_FragCoord;
}
The fragment coordinates are correct (each fragment is pointing to a x,y coordinate in the texture), so are all the input values (_a to _e), but I do not see them outputted correctly to the textures when reading back. I also tried accessing the texture in the shader to see if it was only a read-back error, but my debug SSBO returned all zeroes.
Am I missing some setup step?
I've tested both on linux and windows (titan and 540M geforces) and I'm using openGL 4.3.
As derhass pointed out in the comments above, the problem was with the texture format. I assumed that by passing GL_FLOAT as the data type it would use 32bit floats for each of the RGBA channels. It was not so.
As derhass said, the data type parameter here does not change the texture format. I had to change the internalFormat parameter to what I wanted (GL_RGBA32F) so that it would work as expected.
So, after changing glTexImage2D call to:
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA32F, m_texSize.x, m_texSize.y, 0, GL_RGBA, GL_FLOAT, 0);
I was able to correctly render the results to the texture and read it back. :)

GLSL - Incorrect results when retrieving values from shadow cubemap

When using cubemaps I'm getting inconsistent results in my shaders as opposed to my program.
For testing purposes I wrote a test-program that simply creates a depth cubemap texture and writes '1' to all sides of it:
unsigned int frameBuffer;
glGenFramebuffers(1,&frameBuffer);
unsigned int texture;
glGenTextures(1,&texture);
glBindFramebuffer(GL_FRAMEBUFFER,frameBuffer);
glBindTexture(GL_TEXTURE_CUBE_MAP,texture);
glDrawBuffer(GL_NONE);
glReadBuffer(GL_NONE);
glTexParameteri(GL_TEXTURE_CUBE_MAP,GL_DEPTH_TEXTURE_MODE,GL_LUMINANCE);
glTexParameteri(GL_TEXTURE_CUBE_MAP,GL_TEXTURE_COMPARE_FUNC,GL_LEQUAL);
glTexParameteri(GL_TEXTURE_CUBE_MAP,GL_TEXTURE_COMPARE_MODE,GL_COMPARE_R_TO_TEXTURE);
unsigned int width = 512;
unsigned int height = 512;
for(unsigned int i=0;i<6;i++)
{
glTexImage2D(
GL_TEXTURE_CUBE_MAP_POSITIVE_X +i,
0,
GL_DEPTH_COMPONENT16,
width,height,
0,GL_DEPTH_COMPONENT,
GL_FLOAT,
0
);
glFramebufferTexture2D(GL_FRAMEBUFFER,GL_DEPTH_ATTACHMENT,GL_TEXTURE_CUBE_MAP_POSITIVE_X +i,texture,0);
}
unsigned int status = glCheckFramebufferStatus(GL_FRAMEBUFFER);
if(status != GL_FRAMEBUFFER_COMPLETE)
return;
float *data = new float[width *height];
for(unsigned long long i=0;i<(width *height);i++)
data[i] = 1.f;
for(unsigned int i=0;i<6;i++)
{
glTexSubImage2D(
GL_TEXTURE_CUBE_MAP_POSITIVE_X +i,
0,
0,
0,
width,height,
GL_DEPTH_COMPONENT,
GL_FLOAT,
&data[0]
);
}
delete[] data;
// Check to see if data has been written correctly
data = new float[width *height];
for(unsigned int i=0;i<6;i++)
{
glFramebufferTexture2D(GL_READ_FRAMEBUFFER,GL_DEPTH_ATTACHMENT,GL_TEXTURE_CUBE_MAP_POSITIVE_X +i,texture,0);
glReadPixels(0,0,width,height,GL_DEPTH_COMPONENT,GL_FLOAT,&data[0]);
for(unsigned long long j=0;j<(width *height);j++)
{
if(data[j] != 1.f)
return;
}
}
delete[] data;
// Check end
Rendering:
float screenVerts[18] = {
-1.f,-1.f,0.f,
1.f,-1.f,0.f,
-1.f,1.f,0.f,
-1.f,1.f,0.f,
1.f,-1.f,0.f,
1.f,1.f,0.f
};
unsigned int vertexBuffer;
glGenBuffers(1,&vertexBuffer);
glBindBuffer(GL_ARRAY_BUFFER,vertexBuffer);
glBufferData(GL_ARRAY_BUFFER,sizeof(float) *18,&screenVerts[0],GL_STATIC_DRAW);
glUseProgram(shader);
glBindTexture(GL_TEXTURE_CUBE_MAP,texture);
glEnableVertexAttribArray(0);
glBindBuffer(GL_ARRAY_BUFFER,vertexBuffer);
glVertexAttribPointer(0,3,GL_FLOAT,GL_FALSE,0,(void*)0);
glDrawArrays(GL_TRIANGLES,0,6);
glDisableVertexAttribArray(0);
Vertex Shader:
#version 330 core
layout(location = 0) in vec3 vertPos;
out vec2 UV;
void main()
{
gl_Position = vec4(vertPos,1);
UV = (vertPos.xy +vec2(1,1)) /2.0;
}
Fragment Shader:
#version 330 core
in vec2 UV;
out vec3 color;
uniform samplerCubeShadow testShadow;
void main()
{
color.r = texture(testShadow,vec4(0,0,1,1));
// Just grab the value from a random direction and put it out as red color
}
(I've ported the C++ code from another language, so if you find some syntax errors in there, don't mind those, they're not in the actual code)
glGetError() does not return any errors.
ReadPixels proves that the writing process worked, however the result is a black screen. That means that the texture-call inside the shader returns 0, which should be impossible regardless of what I use as the direction vector.
What am I missing?
You consistently have the arguments for the glBind*() calls reversed. They all take a target as the first argument, and the object id (aka name) as the second argument. Instead of this:
glBindFramebuffer(frameBuffer,GL_FRAMEBUFFER);
glBindTexture(texture,GL_TEXTURE_CUBE_MAP);
glBindBuffer(vertexBuffer,GL_ARRAY_BUFFER);
It should be this:
glBindFramebuffer(GL_FRAMEBUFFER, frameBuffer);
glBindTexture(GL_TEXTURE_CUBE_MAP, texture);
glBindBuffer(GL_ARRAY_BUFFER, vertexBuffer);
There are multiple instances of some of these in the code, so make sure that you catch them all.
Other than that, if this is the complete code, you're not rendering to the screen. You have an FBO without color attachment bound, so the output goes nowhere. If you want to render to the screen, you'll need to unbind the FBO:
glBindFramebuffer(GL_FRAMEBUFFER, 0);
You're also leaving the g and b components of the fragment output undefined, since you only write to r.