Shader two outputs feedback - opengl

I want to get two outputs from my shader, then feed them back into the shader to use as an iterator och accumulator. This works fine for the default "channel", but for two... So, here are the shaders:
#version 420
uniform sampler2DRect inData0;
uniform sampler2DRect inData1;
out float outData0;
out float outData1;
void main(void)
{
outData0 = 1 + texture(inData0, gl_FragCoord.xy).r;
outData1 = -1 + texture(inData1, gl_FragCoord.xy).r;
}
and
#version 420
in vec2 position;
void main()
{
gl_Position = vec4(position, 0, 1);
}
then the standard stuff
Int32 frameBufferId = GL.GenFramebuffer();
GL.BindFramebuffer(FramebufferTarget.Framebuffer, frameBufferId);
Int32 textureId0 = GL.GenTexture();
GL.ActiveTexture(TextureUnit.Texture0);
GL.BindTexture(TextureTarget.TextureRectangle, textureId0);
GL.TexImage2D(TextureTarget.TextureRectangle, 0, PixelInternalFormat.R32f, 1024, 1024, 0, PixelFormat.Red, PixelType.Float, new Single[1024 * 1024]);
GL.FramebufferTexture2D(FramebufferTarget.Framebuffer, FramebufferAttachment.ColorAttachment0, TextureTarget.TextureRectangle, textureId0, 0);
Int32 textureId1 = GL.GenTexture();
GL.ActiveTexture(TextureUnit.Texture1);
GL.BindTexture(TextureTarget.TextureRectangle, textureId1);
GL.TexImage2D(TextureTarget.TextureRectangle, 0, PixelInternalFormat.R32f, 1024, 1024, 0, PixelFormat.Red, PixelType.Float, new Single[1024 * 1024]);
GL.FramebufferTexture2D(FramebufferTarget.Framebuffer, FramebufferAttachment.ColorAttachment1, TextureTarget.TextureRectangle, textureId1, 0);
GL.TexParameter(TextureTarget.TextureRectangle, TextureParameterName.TextureMinFilter, (Int32)TextureMinFilter.Linear);
Single[] arrayBufferData = new Single[4 * 2] { -1, -1, 1, -1, 1, 1, -1, 1 };
Int32 arrayBufferId = GL.GenBuffer();
GL.BindBuffer(BufferTarget.ArrayBuffer, arrayBufferId);
GL.BufferData(BufferTarget.ArrayBuffer, new IntPtr(4 * 2 * sizeof(Single)), arrayBufferData, BufferUsageHint.StaticDraw);
Int32 positionId = GL.GetAttribLocation(programId, "position");
GL.EnableVertexAttribArray(positionId);
GL.VertexAttribPointer(positionId, 2, VertexAttribPointerType.Float, false, 0, 0);
GL.Viewport(0, 0, 1024, 1024);
So, here is where I managed to get somewhere
Single[] result = new Single[1024 * 1024];
GL.**DrawBuffers**(2, new DrawBuffersEnum[] { DrawBuffersEnum.ColorAttachment0, DrawBuffersEnum.ColorAttachment1 });
GL.DrawArrays(PrimitiveType.Quads, 0, 4);
GL.**ReadBuffer**(ReadBufferMode.ColorAttachment0);
GL.ReadPixels(0, 0, 1024, 1024, PixelFormat.Red, PixelType.Float, result);
GL.**ReadBuffer**(ReadBufferMode.ColorAttachment1);
GL.ReadPixels(0, 0, 1024, 1024, PixelFormat.Red, PixelType.Float, result);
Which gives me both outputs properly. For one "channel", I'd call DrawArrays again and it would feed the output0 into input0, but here I want output1 to also go into input1. Is this doable? This R32f / Red is for testing -- they will be replaced by vec4.
Edit:
changed:
layout(binding = 0) uniform sampler2DRect inData0
layout(binding = 1) uniform sampler2DRect inData1
layout(location = 0) out float outData0;
layout(location = 1) out float outData1;
and it works.

Bind input, and point the output to correct locatio.
layout(binding = 0) uniform sampler2DRect inData0
layout(binding = 1) uniform sampler2DRect inData1
layout(location = 0) out float outData0;
layout(location = 1) out float outData1;

Related

Multiple output textures from the same program

I'm trying to learn how to do multiple outputs from the same program in WebGL2 leveraging gl.drawBuffer() capabilities.
I looked at the book "OpenGL ES 3.0 Programming Guide", chapter 11 where it lists what is needed for multi-output to take place. However the shader source example is very trivial outputting only constant values.
I'd like to know if someone has a better example? or if one could explain what happened to the TextureCoordinates varying? In normal shader code I would use that to find data values from my inputs and write them out. Now in the face of multiple layouts, how would the TextureCoordinates varying correspond to each layout? What happens to the dimensions of my viewPort? which output Texture does that correspond with?
Here are some steps the way I understood them:
Create a Color attachment array GL_COLOR_ATTACHMENT0, ...
Create a framebuffer object for each output
Create output textures
For each FB:
BindFramebuffer
BindTexture
Associate texture with FBO: frameBufferTexture2D (..., color_attchment_from_step1)
call drawBuffers passing the color attachment array
Inside the shader access output values like this:
layout(location = 0) out vec4 fragData0;
layout(location = 1) out vec4 fragData1;
You only need one framebuffer object. You attach all the textures to it. So your steps would be
Create a framebuffer object and BindFramebuffer
Create output textures
For each texture:
Associate texture with FBO: frameBufferTexture2D(...)
Create a Color attachment array GL_COLOR_ATTACHMENT0, ...
call drawBuffers passing the color attachment array
function main() {
const gl = document.querySelector('canvas').getContext('webgl2');
if (!gl) {
return alert("need WebGL2");
}
const vs = `
#version 300 es
void main() {
gl_PointSize = 300.0;
gl_Position = vec4(0, 0, 0, 1);
}
`;
const fs = `
#version 300 es
precision mediump float;
layout(location = 0) out vec4 outColor0;
layout(location = 1) out vec4 outColor1;
layout(location = 2) out vec4 outColor2;
layout(location = 3) out vec4 outColor3;
void main() {
outColor0 = vec4(1, .5, .3, .7); // orange
outColor1 = vec4(.6, .5, .4, .3); // brown
outColor2 = vec4(.2, .8, .0, 1); // green
outColor3 = vec4(.3, .4, .9, .6); // blue
}
`
const program = twgl.createProgram(gl, [vs, fs]);
const textures = [];
const fb = gl.createFramebuffer();
gl.bindFramebuffer(gl.FRAMEBUFFER, fb);
for (let i = 0; i < 4; ++i) {
const tex = gl.createTexture();
textures.push(tex);
gl.bindTexture(gl.TEXTURE_2D, tex);
const width = 1;
const height = 1;
const level = 0;
gl.texImage2D(gl.TEXTURE_2D, level, gl.RGBA, width, height, 0,
gl.RGBA, gl.UNSIGNED_BYTE, null);
// attach texture to framebuffer
gl.framebufferTexture2D(gl.FRAMEBUFFER, gl.COLOR_ATTACHMENT0 + i,
gl.TEXTURE_2D, tex, level);
}
// our framebuffer textures are only 1x1 pixels
gl.viewport(0, 0, 1, 1);
// tell it we want to draw to all 4 attachments
gl.drawBuffers([
gl.COLOR_ATTACHMENT0,
gl.COLOR_ATTACHMENT1,
gl.COLOR_ATTACHMENT2,
gl.COLOR_ATTACHMENT3,
]);
// draw a single point
gl.useProgram(program);
{
const offset = 0;
const count = 1
gl.drawArrays(gl.POINT, 0, 1);
}
// --- below this is not relevant to the question but just so we
// --- we can see it's working
// render the 4 textures
const fs2 = `
#version 300 es
precision mediump float;
uniform sampler2D tex[4];
out vec4 outColor;
void main() {
vec4 colors[4];
// you can't index textures with non-constant integer expressions
// in WebGL2 (you can in WebGL1 lol)
colors[0] = texture(tex[0], vec2(0));
colors[1] = texture(tex[1], vec2(0));
colors[2] = texture(tex[2], vec2(0));
colors[3] = texture(tex[3], vec2(0));
vec4 color = vec4(0);
for (int i = 0; i < 4; ++i) {
float x = gl_PointCoord.x * 4.0;
float amount = step(float(i), x) * step(x, float(i + 1));
color = mix(color, colors[i], amount);
}
outColor = vec4(color.rgb, 1);
}
`;
const prgInfo2 = twgl.createProgramInfo(gl, [vs, fs2]);
gl.bindFramebuffer(gl.FRAMEBUFFER, null);
gl.viewport(0, 0, gl.canvas.width, gl.canvas.height);
gl.useProgram(prgInfo2.program);
// binds all the textures and set the uniforms
twgl.setUniforms(prgInfo2, {
tex: textures,
});
gl.drawArrays(gl.POINTS, 0, 1);
}
main();
<script src="https://twgljs.org/dist/4.x/twgl.min.js"></script>
<canvas></canvas>

Texture Coordinates OSG GLSL

I have an image 500x500 pixels which I converted as Texture2D to GLSL and return back the raw data to C++/OSG. I have faced problems with texture coordinates (the coordinates on GLSL goes from 0 to 1). Can someone help me with this point?
C++ code:
cv::Mat test = cv::Mat::zeros(512, 512, CV_8UC3);
test(cv::Rect( 0, 0, 255, 255)).setTo(cv::Scalar(255,0,0));
test(cv::Rect(256, 0, 255, 255)).setTo(cv::Scalar(0,255,0));
test(cv::Rect( 0, 256, 255, 255)).setTo(cv::Scalar(0,0,255));
test(cv::Rect(256, 256, 255, 255)).setTo(cv::Scalar(255,255,255));
osg::ref_ptr<osg::Image> image = new osg::Image;
image->setImage(512, 512, 3, GL_RGB, GL_BGR, GL_UNSIGNED_BYTE, test.data, osg::Image::NO_DELETE, 1);
osg::ref_ptr<osg::Texture2D> texture = new osg::Texture2D;
texture->setTextureSize(512, 512);
texture->setImage(image);
// Pass the texture to GLSL as uniform
osg::StateSet* ss = scene->getOrCreateStateSet();
osg::Uniform* samUniform = new osg::Uniform(osg::Uniform::SAMPLER_2D, "vertexMap");
samUniform->set(0);
ss->addUniform(samUniform);
ss->setTextureAttributeAndModes(0, texture, osg::StateAttribute::ON);
Vertex code:
#version 130
void main() {
gl_Position = gl_ModelViewProjectionMatrix * gl_Vertex;
gl_TexCoord[0] = gl_TextureMatrix[0] * gl_MultiTexCoord0;
}
Fragment code:
#version 130
uniform sampler2D vertexMap;
out vec4 out_data;
void main() {
vec3 value = texture2D(vertexMap, gl_TexCoord[0].st).xyz;
out_data = vec4(value, 1);
}
This is my input data:
Output data from shader:
I solved by replacing texture2D by texelFetch on fragment shader. The difference between these two functions can be found here: https://gamedev.stackexchange.com/questions/66448/how-do-opengls-texelfetch-and-texture-differ
#version 130
uniform sampler2D vertexMap;
out vec4 out_data;
void main() {
vec3 value = texelFetch(vertexMap, ivec2(gl_FragCoord.xy), 0).xyz;
out_data = vec4(value, 1);
}

fragment shader for stipple pattern

I'm trying to draw a stipple pattern using a shader, like as in glStipple. However it does not seem to be working as expected.
The idea is to compare pixel coordinates against a stipple mask and discard fragments if the pixel coordinate mod mask size is 0.
However, two odd things are occurring which suggests the approach has some flaw I'm missing. The first is that the resulting stipple pattern does not match the mask. A single diagonal line in an 8x8 mask results in what appears to be diagonals spaced 1 pixel apart in the drawn shape. The second is that testing against 1, instead of 0, does not give a similar (but shifted) pattern.
The vertex shader looks like this:
static const char *vertexShaderSource =
"attribute highp vec4 posAttr;\n"
"attribute highp vec4 colAttr;\n"
"varying lowp vec4 col;\n"
"varying highp vec2 coord;\n"
"uniform highp mat4 matrix;\n"
"void main() {\n"
" col = colAttr;\n"
" gl_Position = matrix * posAttr;\n"
" coord = gl_Position.xy;\n"
"}\n";
and the fragment shader:
static const char *fragmentShaderSource =
"varying lowp vec4 col;\n"
"varying highp vec2 coord;\n"
"uniform int stipple[64];\n"
"uniform int width;\n"
"uniform int height;\n"
"void main() {\n"
" if (stipple[abs(mod(coord.x * width,8)) + abs(mod(coord.y * height,8)) * 8] == 1)\n"
" {\n"
" discard;\n"
" }\n"
" gl_FragColor = col;\n"
"}\n";
where width and height are the w/h of the viewport and the stipple is e.g.
GLint stipple[64] = {
0, 0, 0, 0, 0, 0, 0, 1,
0, 0, 0, 0, 0, 0, 1, 0,
0, 0, 0, 0, 0, 1, 0, 0,
0, 0, 0, 0, 1, 0, 0, 0,
0, 0, 0, 1, 0, 0, 0, 0,
0, 0, 1, 0, 0, 0, 0, 0,
0, 1, 0, 0, 0, 0, 0, 0,
1, 0, 0, 0, 0, 0, 0, 0
};
Any ideas appreciated, thanks.
You can use gl_FragCoord. It contains the pixel's center coordinate. So for a resolution of 800x600 then gl_FragCoord would be in the range of vec2(0.5, 0.5) and vec2(799.5, 599.5). Thereby there's no need to multiply by the resolution, since we already basically have window coordinates.
Since gl_FragCoord wouldn't result in a negative number you can remove abs().
All in all with the few modifications, the fragment shader ends up looking like this:
varying vec4 color;
uniform int stipple[64];
void main()
{
ivec2 coord = ivec2(gl_FragCoord.xy - 0.5);
if (stipple[int(mod(coord.x, 8) + mod(coord.y, 8) * 8)] == 0)
discard;
gl_FragColor = color;
}

Textures not working (Black Textures) - OpenGL + GLSL

I am trying to detect silhouette edges and render some textures (varies based on the diffuse term) on these edges using OpenGL and shaders. I am rendering a quad using the geometry shader and also assign the texture coordinates here. In the fragment shader, i am trying to use the diffuse term calculated in vertex shader to render different textures based on diffTerm's value. There are two issues with my code.
1) The diffuse term should vary from (-1,1) but it seems to be stuck at 0 when i rotate the model and it reaches negative values at certain positions.
2) The textures are always black and I cant seem to find out what is causing this issue.
"MeshViewer.cpp" - The main file
Mesh* mesh;
GLuint* texID = new GLuint[5];
float rotn_x = 0.0, rotn_y = 0.0, fov;
GLuint matrixLoc1, matrixLoc2, matrixLoc3,texLoc1, texLoc2, texLoc3, texLoc4, texLoc5;
float cam_near, cam_far; //Near and far planes of the camera
const float PI = 3.14159265f;
glm::mat4 view; //View and projection matrices
void loadTextures()
{
glGenTextures(5, texID); //Generate 1 texture ID
glActiveTexture(GL_TEXTURE0); //Texture unit 0
glBindTexture(GL_TEXTURE_2D, texID[0]);
loadTGA("Pencil0.tga");
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glActiveTexture(GL_TEXTURE1); //Texture unit 0
glBindTexture(GL_TEXTURE_2D, texID[1]);
loadTGA("Pencil1.tga");
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glActiveTexture(GL_TEXTURE2); //Texture unit 0
glBindTexture(GL_TEXTURE_2D, texID[2]);
loadTGA("Pencil2.tga");
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glActiveTexture(GL_TEXTURE3); //Texture unit 0
glBindTexture(GL_TEXTURE_2D, texID[3]);
loadTGA("Brick.tga");
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glActiveTexture(GL_TEXTURE4); //Texture unit 0
glBindTexture(GL_TEXTURE_2D, texID[4]);
loadTGA("Brick.tga");
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
}
void initialise()
{
// --- Mesh object ---
GLuint lgtLoc;
mesh = new Mesh();
if(!mesh->loadMeshOFF("Camel.off")) cout << "Error reading mesh data file." << endl;
// --- Camera parameters ---
float win_width = (mesh->_xmax - mesh->_xmin) * 1.5f;
float win_height = (mesh->_ymax - mesh->_ymin) * 1.5f;
if(win_width > win_height) win_height = win_width; //Maintain aspect ratio = 1
cam_near = 2*(mesh->_zmax) - mesh->_zmin;
cam_far = 2*(mesh->_zmin) - mesh->_zmax;
float cam_posx = (mesh->_xmax + mesh->_xmin) * 0.5f;
float cam_posy = (mesh->_ymax + mesh->_ymin) * 0.5f;
float cam_posz = cam_near + win_height;
fov = 27.0f; //Approx. atan(0.5)
// --- Uniform locations ---
GLuint program = createShaderProg("MeshViewer.vert", "MeshViewer.frag", "MeshViewer.geom");
matrixLoc1 = glGetUniformLocation(program, "mvMatrix");
matrixLoc2 = glGetUniformLocation(program, "mvpMatrix");
matrixLoc3 = glGetUniformLocation(program, "norMatrix");
lgtLoc = glGetUniformLocation(program, "lightPos");
GLint lineWidth = glGetUniformLocation(program, "HalfWidth");
if (lineWidth > -1)
glUniform1f(lineWidth, 0.005f);
GLint overhangLength = glGetUniformLocation(program, "OverhangLength");
if (overhangLength > -1)
glUniform1f(overhangLength, 0.15f);
texLoc1 = glGetUniformLocation (program, "tex1");
glUniform1i(texLoc1, 0);
texLoc2 = glGetUniformLocation (program, "tex2");
glUniform1i(texLoc2, 1);
texLoc3 = glGetUniformLocation (program, "tex3");
glUniform1i(texLoc3, 2);
texLoc4 = glGetUniformLocation (program, "tex4");
glUniform1i(texLoc4, 3);
texLoc5 = glGetUniformLocation (program, "tex5");
glUniform1i(texLoc5, 4);
view = glm::lookAt(glm::vec3(cam_posx, cam_posy, cam_posz), glm::vec3(cam_posx, cam_posy, 0.0), glm::vec3(0.0, 1.0, 0.0)); //view matrix
glm::vec4 light = glm::vec4(100.0, 50.0, 100.0, 1.0); //Light's position
glm::vec4 lightEye = view*light; //Light position in eye coordinates
glUniform4fv(lgtLoc, 1, &lightEye[0]);
// --- OpenGL ---
glClearColor(1.0f, 1.0f, 1.0f, 1.0f);
glEnable(GL_DEPTH_TEST);
glPolygonMode(GL_FRONT_AND_BACK, GL_FILL); //Wireframe
mesh->setColor(0, 0, 1); //Mesh color = blue.
mesh->createVAO(); //Create buffer objects for the mesh
}
void display()
{
glm::mat4 proj;
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glm::mat4 matrix = glm::mat4(1.0);
matrix = glm::rotate(matrix, rotn_x, glm::vec3(1.0, 0.0, 0.0)); //rotation about x
matrix = glm::rotate(matrix, rotn_y, glm::vec3(0.0, 1.0, 0.0)); //rotation about y
glm::mat4 prodMatrix1 = view*matrix; //Model-view matrix
proj = glm::perspective(fov, 1.0f, cam_near, cam_far); //perspective projection matrix
glm::mat4 prodMatrix2 = proj*prodMatrix1; //The model-view-projection transformation
glm::mat4 invMatrix = glm::inverse(prodMatrix1); //Inverse of model-view matrix for normal transformation
glUniformMatrix4fv(matrixLoc1, 1, GL_FALSE, &prodMatrix1[0][0]);
glUniformMatrix4fv(matrixLoc2, 1, GL_FALSE, &prodMatrix2[0][0]);
glUniformMatrix4fv(matrixLoc3, 1, GL_TRUE, &invMatrix[0][0]); //Use transpose matrix here
mesh->render();
glFlush();
}
void specialKeys(int key, int x, int y)
{
if(key == GLUT_KEY_LEFT) rotn_y -= 5.0;
else if(key == GLUT_KEY_RIGHT) rotn_y += 5.0;
else if(key == GLUT_KEY_UP) rotn_x -= 5.0;
else if(key == GLUT_KEY_DOWN) rotn_x += 5.0;
else if(key == GLUT_KEY_PAGE_UP) fov --;
else if(key == GLUT_KEY_PAGE_DOWN) fov ++;
if(fov < 1.0) fov = 1.0;
else if(fov > 80.0) fov = 80.0;
glutPostRedisplay();
}
int main(int argc, char** argv)
{
glutInit(&argc, argv);
glutInitDisplayMode (GLUT_SINGLE | GLUT_RGB | GLUT_DEPTH);
glutInitWindowSize (600, 600);
glutInitWindowPosition (20, 10);
glutCreateWindow ("Mesh Viewer");
glutInitContextVersion (4, 2);
glutInitContextProfile ( GLUT_CORE_PROFILE );
if(glewInit() == GLEW_OK)
{
cout << "GLEW initialization successful! " << endl;
cout << " Using GLEW version " << glewGetString(GLEW_VERSION) << endl;
}
else
{
cerr << "Unable to initialize GLEW ...exiting." << endl;
exit(EXIT_FAILURE);
}
initialise ();
glutDisplayFunc(display);
glutSpecialFunc(specialKeys);
glutMainLoop();
return 0;
}
Vertex Shader:
#version 330
layout (location = 0) in vec4 position;
layout (location = 1) in vec3 normal;
layout (location = 2) in vec3 cols;
layout (location = 3) in vec2 texC;
uniform mat4 mvMatrix;
uniform mat4 mvpMatrix;
uniform mat4 norMatrix;
uniform vec4 lightPos;
out float diffTerm;
out vec4 vColour;
out float viewTerm;
out float silhoutte;
out vec2 TexC;
void main()
{
vec4 grey = vec4(0.2, 0.2, 0.2, 1.0);
vec4 posnEye = mvMatrix * position;
vec4 normalEye = norMatrix * vec4(normal, 0);
vec4 lgtVec = normalize(lightPos - posnEye);
vec4 viewVec = normalize(vec4(-posnEye.xyz, 0));
float viewTerm = max(dot(viewVec, normalEye),0);
vec4 material = vec4(cols, 1.0);
vec4 lgtAmb = grey * material;
diffTerm = max(dot(lgtVec, normalEye), 0);
vec4 lgtDiff = material * diffTerm;
silhoutte = dot(viewVec, normalEye);
gl_Position = mvpMatrix * position;
vColour = vec4(cols, 1);
TexC = texC;
}
Geometry Shader:
#version 430 core
layout(triangles_adjacency) in;
layout(triangle_strip, max_vertices = 6) out;
in vec2 texC[];
out vec2 TexCoord;
in vec4 vColour[];
out vec4 colorv;
in float viewTerm[];
out float viewTermg;
in float diffTerm[];
out float diffTermg;
in vec2 TexC[];
out vec2 TexCg;
uniform float HalfWidth;
uniform float OverhangLength;
out float gDist;
out vec3 gSpine;
bool IsFront(vec3 A, vec3 B, vec3 C)
{
float area = (A.x * B.y - B.x * A.y) + (B.x * C.y - C.x * B.y) + (C.x * A.y - A.x * C.y);
return area > 0;
}
void EmitEdge(vec3 P0, vec3 P1)
{
vec3 E = OverhangLength * vec3(P1.xy - P0.xy, 0);
vec2 V = normalize(E.xy);
vec3 N = vec3(-V.y, V.x, 0) * 0.005;
vec3 S = -N;
float D = HalfWidth;
gSpine = P0;
gl_Position = vec4(P0 + S - E, 1); gDist = +D; TexCoord=vec2(0.0,0.0); colorv = vColour[0]; EmitVertex();
gl_Position = vec4(P0 + N - E, 1); gDist = -D; TexCoord=vec2(1.0,0.0); colorv = vColour[1]; EmitVertex();
gSpine = P1;
gl_Position = vec4(P1 + S + E, 1); gDist = +D; TexCoord=vec2(1.0,1.0); colorv = vColour[0]; EmitVertex();
gl_Position = vec4(P1 + N + E, 1); gDist = -D; ; TexCoord=vec2(0.0,1.0); EmitVertex();
EndPrimitive();
}
void main()
{
vec3 v0 = gl_in[0].gl_Position.xyz / gl_in[0].gl_Position.w;
vec3 v1 = gl_in[1].gl_Position.xyz / gl_in[1].gl_Position.w;
vec3 v2 = gl_in[2].gl_Position.xyz / gl_in[2].gl_Position.w;
vec3 v3 = gl_in[3].gl_Position.xyz / gl_in[3].gl_Position.w;
vec3 v4 = gl_in[4].gl_Position.xyz / gl_in[4].gl_Position.w;
vec3 v5 = gl_in[5].gl_Position.xyz / gl_in[5].gl_Position.w;
if (IsFront(v0, v2, v4)) {
if (!IsFront(v0, v1, v2)) EmitEdge(v0, v2);
//if (!IsFront(v2, v3, v4)) EmitEdge(v2, v4);
//if (!IsFront(v0, v4, v5)) EmitEdge(v4, v0);
}
}
Fragment Shader:
#version 330
in vec4 vColourg;
in float diffTermg;
in float silhoutte;
in vec2 TexCg;
in vec2 TexCoord;
uniform sampler2D tex1;
uniform sampler2D tex2;
uniform sampler2D tex3;
uniform sampler2D tex4;
uniform sampler2D tex5;
void main()
{
vec4 texColor1 = texture(tex1, TexCoord);
vec4 texColor2 = texture(tex2, TexCoord);
vec4 texColor3 = texture(tex3, TexCoord);
vec4 texColor4 = texture(tex4, TexCoord);
vec4 texColor5 = texture(tex5, TexCoord);
vec4 blue = vec4(0.0,0.0,1.0,0.0);
vec4 red = vec4(1.0,0.0,0.0,0.0);
vec4 yellow = vec4(1.0,1.0,0.0,0.0);
if (diffTermg<0)
{
gl_FragColor = blue;
}
else if (diffTermg ==0)
{
gl_FragColor = texColor5;
}
else if (diffTermg > 0 && diffTermg < 0.2)
gl_FragColor = yellow;
else if (diffTermg > 100)
gl_FragColor = blue;
}
EDIT:
Shader.h
GLuint loadShader(GLenum shaderType, string filename)
{
ifstream shaderFile(filename.c_str());
if(!shaderFile.good()) cout << "Error opening shader file." << endl;
stringstream shaderData;
shaderData << shaderFile.rdbuf();
shaderFile.close();
string shaderStr = shaderData.str();
const char* shaderTxt = shaderStr.c_str();
GLuint shader = glCreateShader(shaderType);
glShaderSource(shader, 1, &shaderTxt, NULL);
glCompileShader(shader);
GLint status;
glGetShaderiv(shader, GL_COMPILE_STATUS, &status);
if (status == GL_FALSE)
{
GLint infoLogLength;
glGetShaderiv(shader, GL_INFO_LOG_LENGTH, &infoLogLength);
GLchar *strInfoLog = new GLchar[infoLogLength + 1];
glGetShaderInfoLog(shader, infoLogLength, NULL, strInfoLog);
const char *strShaderType = NULL;
cerr << "Compile failure in shader: " << strInfoLog << endl;
delete[] strInfoLog;
}
return shader;
}
GLuint createShaderProg(string vertShader, string fragShader, string geomShader)
{
GLuint shaderv = loadShader(GL_VERTEX_SHADER, vertShader);
GLuint shaderf = loadShader(GL_FRAGMENT_SHADER, fragShader);
GLuint shaderg = loadShader(GL_GEOMETRY_SHADER, geomShader);
GLuint program = glCreateProgram();
glAttachShader(program, shaderv);
glAttachShader(program, shaderf);
glAttachShader(program, shaderg);
glLinkProgram(program);
GLint status;
glGetProgramiv (program, GL_LINK_STATUS, &status);
if (status == GL_FALSE)
{
GLint infoLogLength;
glGetProgramiv(program, GL_INFO_LOG_LENGTH, &infoLogLength);
GLchar *strInfoLog = new GLchar[infoLogLength + 1];
glGetProgramInfoLog(program, infoLogLength, NULL, strInfoLog);
fprintf(stderr, "Linker failure: %s\n", strInfoLog);
delete[] strInfoLog;
program = 0;
}
glUseProgram(program);
return program;
}
loadTGA.h
void loadTGA(string filename)
{
char id, cmap, imgtype, bpp, c_garb;
char* imageData, temp;
short int s_garb, wid, hgt;
int nbytes, size, indx;
ifstream file( filename.c_str(), ios::in | ios::binary);
if(!file)
{
cout << "*** Error opening image file: " << filename.c_str() << endl;
exit(1);
}
file.read (&id, 1);
file.read (&cmap, 1);
file.read (&imgtype, 1);
if(imgtype != 2 && imgtype != 3 ) //2= colour (uncompressed), 3 = greyscale (uncompressed)
{
cout << "*** Incompatible image type: " << (int)imgtype << endl;
exit(1);
}
//Color map specification
file.read ((char*)&s_garb, 2);
file.read ((char*)&s_garb, 2);
file.read (&c_garb, 1);
//Image specification
file.read ((char*)&s_garb, 2); //x origin
file.read ((char*)&s_garb, 2); //y origin
file.read ((char*)&wid, 2); //image width
file.read ((char*)&hgt, 2); //image height
file.read (&bpp, 1); //bits per pixel
file.read (&c_garb, 1); //img descriptor
nbytes = bpp / 8; //No. of bytes per pixels
size = wid * hgt * nbytes; //Total number of bytes to be read
imageData = new char[size];
file.read(imageData, size);
//cout << ">>>" << nbytes << " " << wid << " " << hgt << endl;
if(nbytes > 2) //swap R and B
{
for(int i = 0; i < wid*hgt; i++)
{
indx = i*nbytes;
temp = imageData[indx];
imageData[indx] = imageData[indx+2];
imageData[indx+2] = temp;
}
}
switch (nbytes)
{
case 1:
glTexImage2D(GL_TEXTURE_2D, 0, 1, wid, hgt, 0, GL_LUMINANCE, GL_UNSIGNED_BYTE, imageData);
break;
case 3:
glTexImage2D(GL_TEXTURE_2D, 0, 3, wid, hgt, 0, GL_RGB, GL_UNSIGNED_BYTE, imageData);
break;
case 4:
glTexImage2D(GL_TEXTURE_2D, 0, 4, wid, hgt, 0, GL_RGBA, GL_UNSIGNED_BYTE, imageData);
break;
}
delete imageData;
}
You are requesting a core profile context:
glutInitContextProfile(GLUT_CORE_PROFILE);
But your glTexImage2D() calls are not compatible with the core profile:
glTexImage2D(GL_TEXTURE_2D, 0, 1, wid, hgt, 0, GL_LUMINANCE, GL_UNSIGNED_BYTE, imageData);
glTexImage2D(GL_TEXTURE_2D, 0, 3, wid, hgt, 0, GL_RGB, GL_UNSIGNED_BYTE, imageData);
glTexImage2D(GL_TEXTURE_2D, 0, 4, wid, hgt, 0, GL_RGBA, GL_UNSIGNED_BYTE, imageData);
Using the number of components as the internal format (argument 3) is legacy from OpenGL 1.0, and was finally eliminated when the core profile was introduced. GL_LUMINANCE is also gone. The modern (core profile) equivalent of those calls is:
glTexImage2D(GL_TEXTURE_2D, 0, GL_R8, wid, hgt, 0, GL_RED, GL_UNSIGNED_BYTE, imageData);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB8, wid, hgt, 0, GL_RGB, GL_UNSIGNED_BYTE, imageData);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA8, wid, hgt, 0, GL_RGBA, GL_UNSIGNED_BYTE, imageData);
The post contains a lot of code, and I did not study it in detail to see if there are other problems. I strongly recommend the use of glGetError(), which would have reported these invalid arguments, and should also be helpful to check if there are any additional problems.

Opengl 4.3 gl_VertexID not incrementing with glDrawArrays

I'm having difficulty in understanding why gl_VertexID is not properly incrementing for each new vertex in the code below for rendering "debug text". Hints/tips?
(Original code is referenced at the bottom of this post)
Hereafter is the vertex shader:
#version 430 core
layout( location = 0 ) in int Character;
out int vCharacter;
out int vPosition;
void main()
{
vPosition = gl_VertexID;
vCharacter = Character;
gl_Position = vec4(0, 0, 0, 1);
}
The geometry shader:
#version 430 core
layout(points) in;
layout(triangle_strip, max_vertices = 4) out;
in int vCharacter[1];
in int vPosition[1];
out vec2 gTexCoord;
uniform sampler2D Sampler;
uniform vec2 CellSize;
uniform vec2 CellOffset;
uniform vec2 RenderSize;
uniform vec2 RenderOrigin;
void main()
{
// Determine the final quad's position and size:
float x = RenderOrigin.x + float(vPosition[0]) * RenderSize.x * 2.0f;
float y = RenderOrigin.y;
vec4 P = vec4(x, y, 0, 1);
vec4 U = vec4(1, 0, 0, 0) * RenderSize.x;
vec4 V = vec4(0, 1, 0, 0) * RenderSize.y;
// Determine the texture coordinates:
int letter = vCharacter[0];
letter = clamp(letter - 32, 0, 96);
int row = letter / 16 + 1;
int col = letter % 16;
float S0 = CellOffset.x + CellSize.x * col;
float T0 = CellOffset.y + 1 - CellSize.y * row;
float S1 = S0 + CellSize.x - CellOffset.x;
float T1 = T0 + CellSize.y;
// Output the quad's vertices:
gTexCoord = vec2(S0, T1); gl_Position = P - U - V; EmitVertex();
gTexCoord = vec2(S1, T1); gl_Position = P + U - V; EmitVertex();
gTexCoord = vec2(S0, T0); gl_Position = P - U + V; EmitVertex();
gTexCoord = vec2(S1, T0); gl_Position = P + U + V; EmitVertex();
EndPrimitive();
}
The draw call and other relevant code:
glBindVertexArray(0);
glBindBuffer(GL_ARRAY_BUFFER, 0);
GLuint attribLocation = glGetAttribLocation(m_ProgramTextPrinter, "Character");
glVertexAttribIPointer(attribLocation, 1, GL_UNSIGNED_BYTE, 1, text.data()->c_str());
glEnableVertexAttribArray(attribLocation);
glDrawArrays(GL_POINTS, 0, text.data()->size());
Basically this code will be used for some text rendering. When I use this code, I see that my letters are put on top of each other. When I modify
glVertexAttribIPointer(attribLocation, 1, GL_UNSIGNED_BYTE, 1, text.data()->c_str());
into
glVertexAttribIPointer(attribLocation, 1, GL_UNSIGNED_BYTE, 2, text.data()->c_str());
I notice there is a shift in the x-direction as expected from the Geometry shader, nevertheless the letters are still on top of each other.
I'm using an NVIDIA Geforce GT 630M, driver version: 320.18 and an OpenGL 4.3 context.
Reference to the original author's code
I got the code working by using VBOs as Bartek hinted at: I basically replaced
glBindVertexArray(0);
glBindBuffer(GL_ARRAY_BUFFER, 0);
GLuint attribLocation = glGetAttribLocation(m_ProgramTextPrinter, "Character");
glVertexAttribIPointer(attribLocation, 1, GL_UNSIGNED_BYTE, 1, text.data()->c_str());
glEnableVertexAttribArray(attribLocation);
glDrawArrays(GL_POINTS, 0, text.data()->size());
with
GLuint vaoID, bufferID;
glGenVertexArrays(1, &vaoID);
glBindVertexArray(vaoID);
glGenBuffers(1, &bufferID);
glBindBuffer(GL_ARRAY_BUFFER, bufferID);
glBufferData(GL_ARRAY_BUFFER, text.data()->size() * sizeof(GL_UNSIGNED_BYTE), text.data()->data(), GL_DYNAMIC_DRAW);
GLuint attribLocation = glGetAttribLocation(m_ProgramTextPrinter, "Character");
glVertexAttribIPointer(attribLocation, 1, GL_UNSIGNED_BYTE, 0, 0);
glEnableVertexAttribArray(attribLocation);
glDrawArrays(GL_POINTS, 0, text.data()->size());
glDeleteVertexArrays(1, &vaoID);
glDeleteBuffers(1, &bufferID);