GLSL Instancing - Max number of inputs for vertex data? - opengl

I am trying to implement instancing in my OpenGL program. I got it to work, and then decided to make my GLSL code more efficient by sending the Model-View-Projection multiplication matrix as input to the GLSL program, so that the CPU computes it for each instance, opposed to the GPU. Here is my vertex shader code (most of it is irrelevant to my question):
#version 330 core
// Input vertex data, different for all executions of this shader.
layout(location = 0) in vec3 vertexPosition_modelspace;
layout(location = 2) in vec3 vertexColor;
layout(location = 3) in vec3 vertexNormal_modelspace;
layout(location = 6) in mat4 models;
layout(location = 10) in mat4 modelsV;
layout(location = 14) in mat4 modelsVP;
// Output data ; will be interpolated for each fragment.
out vec3 newColor;
out vec3 Position_worldspace;
out vec3 Normal_cameraspace;
out vec3 EyeDirection_cameraspace;
// Values that stay constant for the whole mesh.
uniform mat4 MVP;
uniform mat4 MV;
uniform mat4 P;
uniform mat4 V;
uniform mat4 M;
uniform int num_lights;
uniform vec3 Lights[256];
void main(){
// Output position of the vertex, in clip space : MVP * position
gl_Position = P * modelsV * vec4(vertexPosition_modelspace,1);
// Position of the vertex, in worldspace : M * position
Position_worldspace = (models * vec4(vertexPosition_modelspace,1)).xyz;
// Vector that goes from the vertex to the camera, in camera space.
// In camera space, the camera is at the origin (0,0,0).
vec3 vertexPosition_cameraspace = ( modelsV * vec4(vertexPosition_modelspace,1)).xyz;
EyeDirection_cameraspace = vec3(0,0,0) - vertexPosition_cameraspace;
// Normal of the the vertex, in camera space
Normal_cameraspace = ( modelsV * vec4(vertexNormal_modelspace,0)).xyz;
// UV of the vertex. No special space for this one.
newColor = vertexColor;
}
The above code works, but only because I'm not using the last input modelsVP to calculate gl_position. If I do use it (instead of computing P*modelsV), the instances won't be drawn, and I get this error:
Linking program
Compiling shader : GLSL/meshColor.vertexshader
Compiling shader : GLSL/meshColor.fragmentshader
Linking program
Vertex info
0(10) : error C5102: input semantic attribute "ATTR" has too big of a numeric index (16)
0(10) : error C5102: input semantic attribute "ATTR" has too big of a numeric index (16)
0(10) : error C5041: cannot locate suitable resource to bind variable "modelsVP". Possibly large array.
I'm sure I'm linking it correctly in my OpenGL code, because if I swap the input location modelsVP with modelsV so that it is 10 instead of 14, I am able to use it, but not modelsV. Is there a maximum number of inputs you can have for your vertex shader? I really can't think of any other idea of why else I would get this error...
I'll include more of my OpenGL code that is relevant here, but I'm pretty sure that it's correct (it's not all in the same class or method):
// Buffer data for VBO. The numbers must match the layout in the GLSL code.
#define position 0
#define uv 1
#define color 2
#define normal 3
#define tangent 4
#define bitangent 5
#define model 6 // 4x4 matrices take 4 positions
#define modelV 10
#define modelVP 14
#define num_buffers 18
GLuint VBO[num_buffers];
glGenBuffers(num_buffers, VBO);
for( int i=0; i<ModelMatrices.size(); i++ )
{
mvp.push_back( projection * view * ModelMatrices.at(i) );
mv.push_back( view * ModelMatrices.at(i) );
}
glBindBuffer(GL_ARRAY_BUFFER, VBO[model]);
glBufferData(GL_ARRAY_BUFFER, sizeof(glm::mat4) * ModelMatrices.size(), &ModelMatrices[0], GL_DYNAMIC_DRAW);
for (unsigned int i = 0; i < 4 ; i++) {
glEnableVertexAttribArray(model + i);
glVertexAttribPointer(model + i, 4, GL_FLOAT, GL_FALSE, sizeof(glm::mat4),
(const GLvoid*)(sizeof(GLfloat) * i * 4));
glVertexAttribDivisor(model + i, 1);
}
glBindBuffer(GL_ARRAY_BUFFER, VBO[modelV]);
glBufferData(GL_ARRAY_BUFFER, sizeof(glm::mat4) * mv.size(), &mv[0], GL_DYNAMIC_DRAW);
for (unsigned int i = 0; i < 4 ; i++) {
glEnableVertexAttribArray(modelV + i);
glVertexAttribPointer(modelV + i, 4, GL_FLOAT, GL_FALSE, sizeof(glm::mat4),
(const GLvoid*)(sizeof(GLfloat) * i * 4));
glVertexAttribDivisor(modelV + i, 1);
}
glBindBuffer(GL_ARRAY_BUFFER, VBO[modelVP]);
glBufferData(GL_ARRAY_BUFFER, sizeof(glm::mat4) * mvp.size(), &mvp[0], GL_DYNAMIC_DRAW);
for (unsigned int i = 0; i < 4 ; i++) {
glEnableVertexAttribArray(modelVP + i);
glVertexAttribPointer(modelVP + i, 4, GL_FLOAT, GL_FALSE, sizeof(glm::mat4), (const GLvoid*)(sizeof(GLfloat) * i * 4));
glVertexAttribDivisor(modelVP + i, 1);
}

OpenGL mandates implementations offer a minimum of 16 4-component vertex attributes. Therefore an index of 16 is not guaranteed to be supported by all implementations; see GL_MAX_VERTEX_ATTRIBS for more details.
Your mat4 vertex attributes count as 4 4-component attributes, so an index of 14 is out of range on implementations that only support 16 4-component vertex attributes.

You are using too many vertex attributes. Here's how to reduce the number of attributes without changing anything much about your code (and any functional changes are improvements). The following assumes that models is the "model-to-world" matrix, modelsV is the "model-to-camera" matrix, and that modelsVP is the "model-to-projection" matrix:
#version 330 core
// Input vertex data, different for all executions of this shader.
layout(location = 0) in vec3 vertexPosition_modelspace;
layout(location = 2) in vec3 vertexColor;
layout(location = 3) in vec3 vertexNormal_modelspace;
layout(location = 6) in mat4 modelsV;
// Output data ; will be interpolated for each fragment.
out vec3 newColor;
//The fragment shader should work in *camera* space, not world space.
out vec4 Position_cameraspace;
out vec3 Normal_cameraspace;
//out vec3 EyeDirection_cameraspace; Can be computed from Position_cameraspace in the FS.
// Values that stay constant for the whole mesh.
uniform mat4 P;
void main()
{
Position_cameraspace = modelsV * vec4(vertexPosition_modelspace, 1.0);
gl_Position = P * Position_cameraspace;
Normal_cameraspace = ( modelsV * vec4(vertexNormal_modelspace,0)).xyz;
newColor = vertexColor;
}
See? Isn't that much simpler? Fewer uniforms in the vertex shader, fewer outputs to the fragment shader, fewer math computations, and fewer vertex attributes.
All you need to do is change your fragment shader to use the camera-space position, rather than the world-space position. Which should be a reasonably easy change.

Related

How to provide 14 float values per vertex to the shader?

I have downloaded a model, and have exported it as .fbx. The model contains several animations (6 to be precise), and I want to have one of them displayed. Following Anton Gerdelan's book on openGL I wrote an importer, which uses assimp to parse the model, buffers the relevant vertex data and retrieves the offset matrices needed for the animation.
Not having much experience with skeletal animation, I think I've been able to make the necessary changes to anton's importer, so that it can work on the more complex model that I need. However, the importer assumes that each vertex is only influenced by 1 bone, which unfortunately is not the case.
After some tinkering, I figured out that each vertex of the model can be influenced by at most 14 bones at a time. Since I am not sure how I could pass 14 values to the shader containing boneId and the relevant weight I tried changing the code to accommodate up to 4 bones at a time. This is the code that parses the bone id & weights and buffers them:
*bone_count = (int)mesh->mNumBones;
char bone_names[256][64];
struct vertexdata {
int IDs[4];
float Weights[4];
int ptr;
};
vector<vertexdata> vdata;
vdata.resize(*point_count);
for (int i = 0; i < *point_count; i++) {
vdata[i].ptr = 0;
}
for (int b_i = 0; b_i < *bone_count; b_i++) {
const aiBone* bone = mesh->mBones[b_i];
strcpy(bone_names[b_i], bone->mName.data);
printf("bone_names[%i]=%s\n", b_i, bone_names[b_i]);
bone_offset_mats[b_i] = convert_assimp_matrix(bone->mOffsetMatrix);
//getting weights for each bone
int num_weights = (int)bone->mNumWeights;
for (int w_i = 0; w_i < num_weights; w_i++) {
aiVertexWeight weight = bone->mWeights[w_i];
int vid = weight.mVertexId;
float vweight = weight.mWeight;
if (vdata[vid].ptr < 4) {
vdata[vid].IDs[vdata[vid].ptr] = b_i;
vdata[vid].Weights[vdata[vid].ptr] = vweight;
vdata[vid].ptr++;
}
int vertex_id = (int)weight.mVertexId;
}
}
//buffering bone id data
GLuint vbo1;
glGenBuffers(1, &vbo1);
glBindBuffer(GL_ARRAY_BUFFER, vbo1);
glBufferData(GL_ARRAY_BUFFER, sizeof(vdata[0]) * vdata.size(), &vdata[0], GL_STATIC_DRAW);
glEnableVertexAttribArray(3);
glVertexAttribIPointer(3, 4, GL_INT, sizeof(vertexdata), (const GLvoid*)0);
glEnableVertexAttribArray(4);
glVertexAttribPointer(4, 4, GL_FLOAT, GL_FALSE, sizeof(vertexdata), (const GLvoid*)16);
and in the shaders:
vertex shader
#version 330 core
layout (location = 0) in vec3 pos;
layout (location = 1) in vec3 norm;
layout (location = 2) in vec2 UV;
layout (location = 3) in ivec4 boneIDs;
layout (location = 4) in vec4 Weights;
uniform mat4 view, projection, model;
uniform mat4 bone_matrices[40];
out vec2 tCoords;
void main()
{
mat4 boneTransform = bone_matrices[boneIDs[0]] * Weights[0];
boneTransform += bone_matrices[boneIDs[1]] * Weights[1];
boneTransform += bone_matrices[boneIDs[2]] * Weights[2];
boneTransform += bone_matrices[boneIDs[3]] * Weights[3];
tCoords = UV;
gl_Position = projection * view * boneTransform * model * vec4(pos, 1.0);
}
fragment shader
#version 330 core
in vec2 tCoords;
out vec4 fragColour;
uniform sampler2D tex;
void main()
{
fragColour = texture(tex, tCoords);
}
The model is rendered properly, but I am not observing any movement. Again, not knowing much about skeletal animation, I can only assume that it's because I haven't included every bone that influences each vertex, and the corresponding weight. However, when buffering the data the shaders only accept up to vec4 aka 4 values per vertex. How can I pass 14 IDs and 14 weights? Could this be the cause of the animation not working?

How to update array of matrices to glsl shader

I'm currently working with skeletal animation and I'm really close to getting it working. Currently, I have a struct that has a matrix with 100 spots ( this is so that I can max have 100 joints ) like so :
struct skelShader {
glm::mat4 currentJointTrans[100];
};
The struct should be binded in the shader, I've done it like this:
glGenBuffers(1, &sksBuff);
glBindBuffer(GL_UNIFORM_BUFFER, sksBuff);
// bind buffer to work further with it...
// allocate memory for the buffer in the GPU
glBufferData(GL_UNIFORM_BUFFER, sizeof(skelShader), NULL, GL_STATIC_DRAW);
// because we hard-coded "binding=3" in the shader code we can do this:
// bind Uniform Buffer to binding point 3 (without caring about index of UBO)
glBindBufferBase(GL_UNIFORM_BUFFER, 4, sksBuff);
// good practice, unbind buffer
glBindBuffer(GL_UNIFORM_BUFFER, 0);
sksBuff is just an GLuint.
I fill this array with new values every render/frame that goes by, these values are the new transformations for the joints. I do it like this:
for (int i = 0; i < skeleton.size(); i++) {
globalSkelInfo.currentJointTrans[i] = skeleton[i]->transformMat[currentFrame - 1] * skeleton[i]->globalBindPosMat;
}
This is working correctly for the root joint, but the rest of the joints/mesh remains in bind pose. The problem should be located in where I update the array. Currently I do it like this in the render function after I've done the multiplication for each joint:
for (int i = 0; i < skeleton.size(); i++) {
glUniformMatrix4fv(glGetUniformLocation(aShaderProgram, ("currentJointTrans[" + std::to_string(i) + "]").c_str()),
1, GL_FALSE, glm::value_ptr(globalSkelInfo.currentJointTrans[i]));
}
After this I draw. The root joints values seem to be moving correctly, but the rest of the mesh is in bindpose and doesn't move. In the Vertex Shader I try to update the matrix like this:
#version 440
const int maxJoints = 100;
const int maxWeights = 4;
layout(location = 0) in vec3 vertex_position;
layout(location = 1) in vec2 vertex_UV;
layout(location = 2) in vec3 vertex_normal;
layout(location = 3) in vec4 vertex_weight;
layout(location = 4) in ivec4 vertex_controllers;
out vec2 outUVs;
out vec3 outNorm;
layout(binding = 3 , std140) uniform uniformBlock
{
vec3 camPos;
mat4 world;
mat4 LookAt;
mat4 projection;
mat4 MVP;
};
layout(binding = 4 , std140) uniform animationStruct
{
mat4 currentJointTrans[maxJoints];
};
void main() {
vec4 finalModelPos = vec4(0.0);
vec4 finalNormal = vec4(0.0);
for (int i = 0; i < 4; i++) {
mat4 jointTrans = currentJointTrans[vertex_controllers[i]];
vec4 posePos = jointTrans * vec4(vertex_position, 1.0);
finalModelPos += posePos * vertex_weight[i];
vec4 worldNormal = jointTrans * vec4(vertex_normal, 0.0);
finalNormal += worldNormal * vertex_weight[i];
}
gl_Position = MVP * finalModelPos;
outNorm = finalNormal.xyz;
outUVs = vertex_UV;
}
My theory is that the updating of the struct skelShader with my currentJointTrans array is incorrect. Any tips on how I should do this instead?
glUniform* calls cannot set data in uniform buffers. Indeed, the whole point of uniform buffers is that the uniform data comes from a buffer object. That's why you had to create one.
So if you want to set the uniform data for a uniform block, you set that data into the buffer object.

GLSL/OpenGL shader tessellation flickering and failure

I just started with OpenGL tessellation and have run into a bit a trouble. I am tessellating series of patches formed by one vertex each. These vertices/patches are structured in a gridlike fashion to later form a terrain generated by Perlin Noise.
The problem I have run into is that starting from the second patch, and every 5th patch after that, sometimes have a lot of tessellation (not the way i configured) but most of the time it doesn't get tessellated at all.
Like so:
The two white circles mark the highly/over tessellated patches. Also note the pattern of untessellated patches.
The strange thing is that it works on my Surface Pro 2 (Intel HD4400 graphics) but bugs on my main desktop computer (AMD HD6950 graphics). Is it possible the hardware is bad?
The patches are generated with the code:
vec4* patches = new vec4[m_patchesWidth * m_patchesDepth];
int c = 0;
for (unsigned int z = 0; z < m_patchesDepth; ++z) {
for (unsigned int x = 0; x < m_patchesWidth; ++x) {
patches[c] = vec4(x * 1.5f, 0, z * 1.5f, 1.0f);
c++;
}
}
m_fxTerrain->Apply();
glGenBuffers(1, &m_planePatches);
glBindBuffer(GL_ARRAY_BUFFER, m_planePatches);
glBufferData(GL_ARRAY_BUFFER, m_patchesWidth * m_patchesDepth * sizeof(vec4), patches, GL_STATIC_DRAW);
GLuint loc = m_fxTerrain->GetAttrib("posIn");
glEnableVertexAttribArray(loc);
glVertexAttribPointer(0, 4, GL_FLOAT, GL_FALSE, sizeof(vec4), nullptr);
delete(patches);
And drawn with:
glPatchParameteri(GL_PATCH_VERTICES, 1);
glBindVertexArray(patches);
glPolygonMode(GL_FRONT_AND_BACK, GL_LINE);
glDrawArrays(GL_PATCHES, 0, nrOfPatches);
Vertex Shader:
#version 430 core
in vec4 posIn;
out gl_PerVertex {
vec4 gl_Position;
};
void main() {
gl_Position = posIn;
}
Control shader:
#version 430
#extension GL_ARB_tessellation_shader : enable
layout (vertices = 1) out;
uniform float OuterTessFactor;
uniform float InnerTessFactor;
out gl_PerVertex {
vec4 gl_Position;
} gl_out[];
void main() {
if (gl_InvocationID == 0) {
gl_TessLevelOuter[0] = OuterTessFactor;
gl_TessLevelOuter[1] = OuterTessFactor;
gl_TessLevelOuter[2] = OuterTessFactor;
gl_TessLevelOuter[3] = OuterTessFactor;
gl_TessLevelInner[0] = InnerTessFactor;
gl_TessLevelInner[1] = InnerTessFactor;
}
gl_out[gl_InvocationID].gl_Position = gl_in[gl_InvocationID].gl_Position;
}
Evaluation shader:
#version 430
#extension GL_ARB_tessellation_shader : enable
layout (quads, equal_spacing, ccw) in;
uniform mat4 ProjView;
uniform sampler2D PerlinNoise;
out vec3 PosW;
out vec3 Normal;
out vec4 ColorFrag;
out gl_PerVertex {
vec4 gl_Position;
};
void main() {
vec4 pos = gl_in[0].gl_Position;
pos.xz += gl_TessCoord.xy;
pos.y = texture2D(PerlinNoise, pos.xz / vec2(8, 8)).x * 10.0f - 10.0f;
Normal = vec3(0, 1, 0);
gl_Position = ProjView * pos;
PosW = pos.xyz;
ColorFrag = vec4(pos.x / 64.0f, 0.0f, pos.z / 64.0f, 1.0f);
}
Fragment shader:
#version 430 core
in vec3 PosW;
in vec3 Normal;
in vec4 ColorFrag;
in vec4 PosH;
out vec3 FragColor;
out vec3 FragNormal;
void main() {
FragNormal = Normal;
FragColor = ColorFrag.xyz;
}
I have tried to hardcode the different tessellation levels but that did not help. I recently started out with OpenGL so please let me know if i am doing something stupid.
So does anyone have any idea what could be causing this "flickering" of certain patches?
Update: I had a friend run the project and he got the same pattern of flickering tessellation but the failing patches were not drawn at all except when being overly tessellated. He has the same graphics card as I do (AMD HD6950).
You should use triangle/quad tessellation, in which each patch has 3 or 4 vertices. As I can see, you use quads (I use them too). In that case, you can set it like this:
glPatchParameteri(GL_PATCH_VERTICES,4);
glBindVertexArray(VertexArray);
(TIP: use drawelements for your terrain, much better performance for 2D-displacement based mesh.)
In the control shader, use
layout (vertices = 4) out;
since your patch has 4 control points. The ordering is still important (CCW/CW).
Personally I don't like to use built-in variables, so for the vertex shader you can send your vertex data to the tesscontrol like this:
layout (location = 0) out vec3 outPos;
....
outPos.xz = grid.xy;
outPos.y = noise(outPos.xz);
Tess control:
layout (location = 0) in vec3 inPos[]; //outPos (location = 0) from vertex shader
//'collects' the 4 control points to an array in the order they're sended
layout (location = 0) out vec3 outPos[]; //send the c.points to the ev. shader
...
gl_TessLevelOuter[0] = outt[0];
gl_TessLevelOuter[1] = outt[1];
gl_TessLevelOuter[2] = outt[2];
gl_TessLevelOuter[3] = outt[3];
gl_TessLevelInner[0] = inn[0];
gl_TessLevelInner[1] = inn[1];
outPos[ID] = inPos[ID];//gl_invocationID = ID
Note that both in and out vertex data is an array.
The tessev is simple:
layout (location = 0) in vec3 inPos[]; //the 4 control points
layout (location = 0) out vec3 outPos; //this is no longer array, next is the fragment shader
...
//edit: do not forgot to add the next line
layout (quads) in;
vec3 interpolate3D(vec3 v0, vec3 v1, vec3 v2, vec3 v3) //linear interpolation for x,y,z coords on the quad
{
return mix(mix(v0,v1,gl_TessCoord.x),mix(v3,v2,gl_TessCoord.x),gl_TessCoord.y);
};
...main{...
outPos = interpolate3D(inPos[0],inPos[1],inPos[2],inPos[3]); //the four control points of the quad. Every other point is linearly interpolated between them according to the TessCoord.
gl_Position = mvp * vec4(outPos,1.0f);
A good representation of the quad domain: http://ogldev.atspace.co.uk/www/tutorial30/tutorial30.html.
I think the problem is with your one-vertex patch. I cannot imagine how a one vertex path can be divided into triangles, I don't know how it works on another hardware. The tessellation is for divide primitives into other simple primitives, to triangles in case of OGL, since it can be handled by a GPU easily (3 points always lie in a plane). So, the minimum number of patch vertices should be 3, for a triangle. I like quads, because it simplier to index, and the memory cost is less. It will be divided into triangles too during tessellation. http://www.informit.com/articles/article.aspx?p=2120983
Also, there is another type, the isoline tessellation. (check out the links, the second is pretty good.)
All in all, try it with quads or triangles, and set the control vertices to 4 (or 3). My (pretty complex) terrain shader is here with frustum culling, tessellation shader culling for a geoclipmap based terrain. Also, without tessellation it works with vertex morph in vertex shader. Maybe some part of this code will be useful. http://speedy.sh/TAvPR/gshader.txt
A scene with tessellation at about 4 pixels/triangle runs at 75 FPS (with fraps) with runtime normal calculation and bicubic smoothing and other things. I'm using AMD HD 5750. It still could be much faster with better code and pre-baked normals:D. (runs at max 120 w/o normal calc.)
Oh, and you can only send the x and z coords if you displace the vertex in the shader. It will be faster too.
Lots of vertices.

Skeletal Animation ussing ASSIMP and GLSL: bone uniform array size

I'm working on an ASSIMP skeletal animation loader and renderer and right now all the data is correctly loaded and interpolated at its current timeframe. However, there is still one part that isn't working as it should and that's the vertex shader stage.
Via a VBO I pass in two vec4s that contain the bone IDs and the weights for each vertex (up to a maximum of 4 bones/weights per vertex) and the vertex shader has a matrix array of 100 bone transformations (pre-calculated per frame) that are indexed via the bone IDs.
However, it seems that the bones uniform doesn't contain the proper transformations. For debugging purposes I colored the model with the weight values and the bone IDs value and they contain a color (and thus valid values). However, when I transform my vertex via the bone transformation and color the model with the result, the entire model is colored black, meaning the transformation matrices are all 0.0. So they're not initialized properly.
I think the problem is with passing the matrices to the uniform array, or perhaps the maximum size of uniforms allowed (I also tried setting the number of uniform matrices to 32 (number of bones on current model) but without effect)?
Before passing the information to the shader, the transformation matrices are indeed valid matrices (not identity/empty matrices) so the fault should probably be in the GLSL shader or the passing of the uniforms.
The following code is from the vertex shader:
#version 330
layout (location = 0) in vec3 position;
layout(location = 1) in vec3 normal;
layout(location = 2) in vec3 tangent;
layout(location = 3) in vec3 color;
layout(location = 4) in vec2 texCoord;
layout(location = 5) in ivec4 boneIDs;
layout(location = 6) in vec4 weights;
uniform mat4 projection;
uniform mat4 view;
uniform mat4 model;
uniform mat4 bones[100];
out vec2 TexCoord;
out vec4 colorz;
void main()
{
vec4 newPos = vec4(position, 1.0);
colorz = vec4(0.0, 1.0, 0.0, 1.0);
if (weights != vec4(0.0, 0.0, 0.0, 0.0))
{
mat4 boneTransform = bones[boneIDs[0]] * weights[0];
boneTransform += bones[boneIDs[1]] * weights[1];
boneTransform += bones[boneIDs[2]] * weights[2];
boneTransform += bones[boneIDs[3]] * weights[3];
// newPos = boneTransform * vec4(position, 1.0);
vec4 test = vec4(1.0);
colorz = boneTransform * test;
// newPos = boneTransform * newPos;
}
TexCoord = texCoord;
gl_Position = projection * view * model * newPos;
}
The following snippet of code pass the matrix data to the GLSL shader:
// Sets bone transformation matrices
void Shader::SetBoneMatrix(GLint index, aiMatrix4x4 matrix)
{
glm::mat4 mat = glm::transpose(glm::make_mat4(&matrix.a1));
glUniformMatrix3fv(boneLocations[index], 1, GL_FALSE, glm::value_ptr(mat));
}
Also the code that gets all the uniform locations of the bones array:
for(unsigned int i = 0; i < 100; i++)
{
string name = "bones[";
string number;
stringstream ss;
ss << i;
ss >> number;
name += number;
name += ']';
boneLocations[i] = glGetUniformLocation(this->program, name.c_str());
}
Oké, via glslDevil I came across a continous GL_INVALID_OPERATION error when setting the bone matrix to the shader via glUniformMatrix. The origin of the problem was indeed at the stage where the program passes the information along to the shader.
It is quite a stupid mistake actually since I'm using glUniformMatrix3f instead of glUniformMatrix4f. Changing this did indeed solve the problem and the animations are working perfectly right now.

glLineStipple deprecated in OpenGL 3.1

glLineStipple has been deprecated in the latest OpenGL APIs.
What is it replaced with?
If not replaced, how can I get a similar effect?
(I don't want to use a compatibility profile of course...)
Sorry, it hasn't been replaced with anything. The first idea coming to my mind for emulating it would be the geometry shader. You feed the geometry shader with a line, compute its screen space length and based on that you generate a variable number of sub lines between its start and end vertex.
EDIT: Perhaps you could also use a 1D texture with the alpha (or red) channel encoding the pattern as 0.0 (no line) or 1.0 (line) and then have the lines texture coordinate go from 0 to 1 and in the fragment chader you make a simple alpha test, discarding fragments with alpha below some threshold. You can facilitate the geometry shader to generate your line texCoords, as otherwise you need different vertices for every line. This way you can also make the texCoord dependent on the screen space length of the line.
The whole thing get's more difficult if you draw triangles (using polygon mode GL_LINE). Then you have to do the triangle-line transformation yourself in the geometry shader, putting in triangles and putting out lines (that could also be a reason for deprecating polygon mode in the future, if it hasn't already).
EDIT: Although I believe this question abandomned, I have made a simple shader triple for the second approach. It's just a minimal solution, feel free to add custom features yourself. I haven't tested it because I lack the neccessary hardware, but you should get the point:
uniform mat4 modelViewProj;
layout(location=0) in vec4 vertex;
void main()
{
gl_Position = modelViewProj * vertex;
}
The vertex shader is a simple pass through.
layout(lines) in;
layout(line_strip, max_vertices=2) out;
uniform vec2 screenSize;
uniform float patternSize;
noperspective out float texCoord;
void main()
{
vec2 winPos0 = screenSize.xy * gl_in[0].gl_Position.xy / gl_in[0].gl_Position.w;
vec2 winPos1 = screenSize.xy * gl_in[1].gl_Position.xy / gl_in[1].gl_Position.w;
gl_Position = gl_in[0].gl_Position;
texCoord = 0.0;
EmitVertex();
gl_Position = gl_in[1].gl_Position;
texCoord = 0.5 * length(winPos1-winPos0) / patternSize;
EmitVertex();
}
In the geometry shader we take a line and compute its screen space length in pixels. We then devide this by the size of the stipple pattern texture, which would be factor*16 when emulating a call to glLineStipple(factor, pattern). This is taken as 1D texture coordinate of the second line end point.
Note that this texture coordinate has to be interpolated linearly (noperspective interpolation specifier). The usual perpective-correct interpolation would cause the stipple pattern to "squeeze together" on farther away parts of the line, whereas we are explicitly working with screen-space values.
uniform sampler1D pattern;
uniform vec4 lineColor;
noperspective in float texCoord;
layout(location=0) out vec4 color;
void main()
{
if(texture(pattern, texCoord).r < 0.5)
discard;
color = lineColor;
}
The fragment shader now just performs a simple alpha test using the value from the pattern texture, which contains a 1 for line and a 0 for no line. So to emulate the fixed function stipple you would have a 16 pixel 1-component 1D texture instead of a 16bit pattern. Don't forget to set the pattern's wrapping mode to GL_REPEAT, about the filtering mode I'm not that sure, but I suppose GL_NEAREST would be a good idea.
But as said earlier, if you want to render triangles using glPolygonMode, it won't work this way. Instead you have to adapt the geometry shader to accept triangles and generate 3 lines for each triangle.
EDIT: In fact OpenGL 3's direct support for integer operations in shaders allows us to completely drop this whole 1D-texture approach and work straight-forward with an actual bit-pattern. Thus the geometry shader is slightly changed to put out the actual screen-size pattern coordinate, without normalization:
texCoord = 0.5 * length(winPos1-winPos0);
In the fragment shader we then just take a bit pattern as unsigned integer (though 32-bit in contrast to glLineStipple's 16-bit value) and the stretch factor of the pattern and just take the texture coordinate (well, no texture anymore actually, but nevermind) modulo 32 to get it's position on the pattern (those explicit uints are annoying, but my GLSL compiler says implicit conversions between int and uint are evil):
uniform uint pattern;
uniform float factor;
...
uint bit = uint(round(linePos/factor)) & 31U;
if((pattern & (1U<<bit)) == 0U)
discard;
To answer this question, we've to investigate first, what glLineStipple actually does.
See the image, where the quad at the left is drawn by 4 separated line segments using the primitive type GL_LINES.
The circle at the right is drawn by a consecutive polygon line, using the primitive type GL_LINE_STRIP.
When using line segments, the stipple pattern started at each segment. The pattern is restarted at each primitive.
When using a line strip, then the stipple pattern is applied seamless to the entire polygon. A pattern seamlessly continuous beyond vertex coordinates.
Be aware that the length of the pattern is stretched at the diagonals. This is possibly the key to the implementation.
For separate line segments, this is not very complicated at all, but for line strips things get a bit more complicated. The length of the line cannot be calculated in the shader program, without knowing all the primitives of the line. Even if all the primitives would be known (e.g. SSBO), then the calculation would have to be done in a loop.
See also Dashed lines with OpenGL core profile.
Anyway, it is not necessary to implement a geometry shader. The trick is to know the start of the line segment in the fragment shader. This easy by using a flat interpolation qualifier.
The vertex shader has to pass the normalized device coordinate to the fragment shader. Once with default interpolation and once with no (flat) interpolation. This causes that in the fragment shade, the first input parameter contains the NDC coordinate of the actual position on the line and the later the NDC coordinate of the start of the line.
#version 330
layout (location = 0) in vec3 inPos;
flat out vec3 startPos;
out vec3 vertPos;
uniform mat4 u_mvp;
void main()
{
vec4 pos = u_mvp * vec4(inPos, 1.0);
gl_Position = pos;
vertPos = pos.xyz / pos.w;
startPos = vertPos;
}
Additionally the varying inputs, the fragment shader has uniform variables. u_resolution contains the width and the height of the viewport. u_factor and u_pattern are the multiplier and the 16 bit pattern according to the parameters of glLineStipple.
So the length of the line from the start to the actual fragment can be calculated:
vec2 dir = (vertPos.xy-startPos.xy) * u_resolution/2.0;
float dist = length(dir);
And fragment on the gap can be discarded, by the discard command.
uint bit = uint(round(dist / u_factor)) & 15U;
if ((u_pattern & (1U<<bit)) == 0U)
discard;
Fragment shader:
#version 330
flat in vec3 startPos;
in vec3 vertPos;
out vec4 fragColor;
uniform vec2 u_resolution;
uniform uint u_pattern;
uniform float u_factor;
void main()
{
vec2 dir = (vertPos.xy-startPos.xy) * u_resolution/2.0;
float dist = length(dir);
uint bit = uint(round(dist / u_factor)) & 15U;
if ((u_pattern & (1U<<bit)) == 0U)
discard;
fragColor = vec4(1.0);
}
This implementation is much easier and shorter, then using geometry shaders. The flat interpolation qualifier is supported since GLSL 1.30 and GLSL ES 3.00. In this version geometry shaders are not supported.
See the line rendering which was generated with the above shader.
The shader gives a proper result line segments, but fails for line strips, since the stipple pattern is restarted at each vertex coordinate.
The issue can't even be solved by a geometry shader. This part of the question remains still unresolved.
For the following simple demo program I've used the GLFW API for creating a window, GLEW for loading OpenGL and GLM -OpenGL Mathematics for the math. I don't provide the code for the function CreateProgram, which just creates a program object, from the vertex shader and fragment shader source code:
#include <vector>
#include <string>
#include <glm/glm.hpp>
#include <glm/gtc/matrix_transform.hpp>
#include <glm/gtc/type_ptr.hpp>
#include <gl/gl_glew.h>
#include <GLFW/glfw3.h>
std::string vertShader = R"(
#version 330
layout (location = 0) in vec3 inPos;
flat out vec3 startPos;
out vec3 vertPos;
uniform mat4 u_mvp;
void main()
{
vec4 pos = u_mvp * vec4(inPos, 1.0);
gl_Position = pos;
vertPos = pos.xyz / pos.w;
startPos = vertPos;
}
)";
std::string fragShader = R"(
#version 330
flat in vec3 startPos;
in vec3 vertPos;
out vec4 fragColor;
uniform vec2 u_resolution;
uniform uint u_pattern;
uniform float u_factor;
void main()
{
vec2 dir = (vertPos.xy-startPos.xy) * u_resolution/2.0;
float dist = length(dir);
uint bit = uint(round(dist / u_factor)) & 15U;
if ((u_pattern & (1U<<bit)) == 0U)
discard;
fragColor = vec4(1.0);
}
)";
GLuint CreateVAO(std::vector<glm::vec3> &varray)
{
GLuint bo[2], vao;
glGenBuffers(2, bo);
glGenVertexArrays(1, &vao);
glBindVertexArray(vao);
glEnableVertexAttribArray(0);
glBindBuffer(GL_ARRAY_BUFFER, bo[0] );
glBufferData(GL_ARRAY_BUFFER, varray.size()*sizeof(*varray.data()), varray.data(), GL_STATIC_DRAW);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, 0);
return vao;
}
int main(void)
{
if ( glfwInit() == 0 )
return 0;
GLFWwindow *window = glfwCreateWindow( 800, 600, "GLFW OGL window", nullptr, nullptr );
if ( window == nullptr )
return 0;
glfwMakeContextCurrent(window);
glewExperimental = true;
if ( glewInit() != GLEW_OK )
return 0;
GLuint program = CreateProgram(vertShader, fragShader);
GLint loc_mvp = glGetUniformLocation(program, "u_mvp");
GLint loc_res = glGetUniformLocation(program, "u_resolution");
GLint loc_pattern = glGetUniformLocation(program, "u_pattern");
GLint loc_factor = glGetUniformLocation(program, "u_factor");
glUseProgram(program);
GLushort pattern = 0x18ff;
GLfloat factor = 2.0f;
glUniform1ui(loc_pattern, pattern);
glUniform1f(loc_factor, factor);
//glLineStipple(2.0, pattern);
//glEnable(GL_LINE_STIPPLE);
glm::vec3 p0(-1.0f, -1.0f, 0.0f);
glm::vec3 p1(1.0f, -1.0f, 0.0f);
glm::vec3 p2(1.0f, 1.0f, 0.0f);
glm::vec3 p3(-1.0f, 1.0f, 0.0f);
std::vector<glm::vec3> varray1{ p0, p1, p1, p2, p2, p3, p3, p0 };
GLuint vao1 = CreateVAO(varray1);
std::vector<glm::vec3> varray2;
for (size_t u=0; u <= 360; u += 8)
{
double a = u*M_PI/180.0;
double c = cos(a), s = sin(a);
varray2.emplace_back(glm::vec3((float)c, (float)s, 0.0f));
}
GLuint vao2 = CreateVAO(varray2);
glm::mat4(project);
int vpSize[2]{0, 0};
while (!glfwWindowShouldClose(window))
{
int w, h;
glfwGetFramebufferSize(window, &w, &h);
if (w != vpSize[0] || h != vpSize[1])
{
vpSize[0] = w; vpSize[1] = h;
glViewport(0, 0, vpSize[0], vpSize[1]);
float aspect = (float)w/(float)h;
project = glm::ortho(-aspect, aspect, -1.0f, 1.0f, -10.0f, 10.0f);
glUniform2f(loc_res, (float)w, (float)h);
}
glClear(GL_COLOR_BUFFER_BIT);
glm::mat4 modelview1( 1.0f );
modelview1 = glm::translate(modelview1, glm::vec3(-0.6f, 0.0f, 0.0f) );
modelview1 = glm::scale(modelview1, glm::vec3(0.5f, 0.5f, 1.0f) );
glm::mat4 mvp1 = project * modelview1;
glUniformMatrix4fv(loc_mvp, 1, GL_FALSE, glm::value_ptr(mvp1));
glBindVertexArray(vao1);
glDrawArrays(GL_LINES, 0, (GLsizei)varray1.size());
glm::mat4 modelview2( 1.0f );
modelview2 = glm::translate(modelview2, glm::vec3(0.6f, 0.0f, 0.0f) );
modelview2 = glm::scale(modelview2, glm::vec3(0.5f, 0.5f, 1.0f) );
glm::mat4 mvp2 = project * modelview2;
glUniformMatrix4fv(loc_mvp, 1, GL_FALSE, glm::value_ptr(mvp2));
glBindVertexArray(vao2);
glDrawArrays(GL_LINE_STRIP, 0, (GLsizei)varray2.size());
glfwSwapBuffers(window);
glfwPollEvents();
}
glfwTerminate();
return 0;
}
See also
Dashed line in OpenGL3?
OpenGL ES - Dashed Lines
Since I struggled a bit (no pun intended) to get it right, I thought it could be useful to others if I shared my implementation of a set of stippling shaders based on Christian Rau's version.
To control pattern density, the fragment shader requires the number of patterns nPatterns per unit length of the viewport - instead of setting a factor. Also included is an optional clipping plane feature.
The rest is mainly commenting and cleaning.
Free to use to all intents and purposes.
The vertex shader:
#version 330
in vec4 vertex;
void main(void)
{
// just a pass-through
gl_Position = vertex;
}
The geometry shader:
#version 330
layout(lines) in;
layout(line_strip, max_vertices = 2) out;
uniform mat4 pvmMatrix;
uniform mat4 mMatrix;
uniform mat4 vMatrix;
out vec3 vPosition; // passed to the fragment shader for plane clipping
out float texCoord; // passed to the fragment shader for stipple pattern
void main(void)
{
// to achieve uniform pattern density whatever the line orientation
// the upper texture coordinate is made proportional to the line's length
vec3 pos0 = gl_in[0].gl_Position.xyz;
vec3 pos1 = gl_in[1].gl_Position.xyz;
float max_u_texture = length(pos1 - pos0);
// Line Start
gl_Position = pvmMatrix * (gl_in[0].gl_Position);
texCoord = 0.0;
// depth position for clip plane
vec4 vsPos0 = vMatrix * mMatrix * gl_Position;
vPosition = vsPos0.xyz / vsPos0.w;
EmitVertex(); // one down, one to go
// Line End
gl_Position = pvmMatrix * (gl_in[1].gl_Position);
texCoord = max_u_texture;
// depth position for clip plane
vec4 vsPos1 = vMatrix * mMatrix * gl_Position;
vPosition = vsPos0.xyz / vsPos0.w;
EmitVertex();
// done
EndPrimitive();
}
The fragment shader:
#version 330
uniform int pattern; // an integer between 0 and 0xFFFF representing the bitwise pattern
uniform int nPatterns; // the number of patterns/unit length of the viewport, typically 200-300 for good pattern density
uniform vec4 color;
uniform vec4 clipPlane0; // defined in view-space
in float texCoord;
in vec3 vPosition;
layout(location=0) out vec4 fragColor;
void main(void)
{
// test vertex postion vs. clip plane position (optional)
if (vPosition.z > clipPlane0.w) {
discard;
return;
}
// use 4 bytes for the masking pattern
// map the texture coordinate to the interval [0,2*8[
uint bitpos = uint(round(texCoord * nPatterns)) % 16U;
// move a unit bit 1U to position bitpos so that
// bit is an integer between 1 and 1000 0000 0000 0000 = 0x8000
uint bit = (1U << bitpos);
// test the bit against the masking pattern
// Line::SOLID: pattern = 0xFFFF; // = 1111 1111 1111 1111 = solid pattern
// Line::DASH: pattern = 0x3F3F; // = 0011 1111 0011 1111
// Line::DOT: pattern = 0x6666; // = 0110 0110 0110 0110
// Line::DASHDOT: pattern = 0xFF18; // = 1111 1111 0001 1000
// Line::DASHDOTDOT: pattern = 0x7E66; // = 0111 1110 0110 0110
uint up = uint(pattern);
// discard the bit if it doesn't match the masking pattern
if ((up & bit) == 0U) discard;
fragColor = color;
}