I'm creating a glBufferData using an array explicitly defined of vertex coordinates and everything works fine. When using a non explicitly defined array the result is totally different althought I'm using the same coordinates values. The drawcall renders things in a different vertexs order.
The vertexs are a grid of triangles.
Code 1 that works fine:
this->graphics->createBuffers(1, vboIndexsID);
this->graphics->bindBuffer(GL_ELEMENT_ARRAY_BUFFER, *vboIndexsID);
this->graphics->bufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(GLushort)*numIndexs, ind, GL_STATIC_DRAW);
this->graphics->createBuffers(1, vboVertexsID);
this->graphics->bindBuffer(GL_ARRAY_BUFFER, *vboVertexsID);
this->graphics->bufferData(GL_ARRAY_BUFFER, 300, quad_verts2, GL_STATIC_DRAW);
where quad_verts2 is:
GLfloat quad_verts2[75] =
{
0.0f, 0.0f, 0.0f,
0.0f, 0.0f, 100.0f,
0.0f, 0.0f, 200.0f,
0.0f, 0.0f, 300.0f,
0.0f, 0.0f, 400.0f,
100.0f, 0.0f, 0.0f,
100.0f, 0.0f, 100.0f,
100.0f, 0.0f, 200.0f,
100.0f, 0.0f, 300.0f,
100.0f, 0.0f, 400.0f,
200.0f, 0.0f, 0.0f,
200.0f, 0.0f, 100.0f,
200.0f, 0.0f, 200.0f,
200.0f, 0.0f, 300.0f,
200.0f, 0.0f, 400.0f,
300.0f, 0.0f, 0.0f,
300.0f, 0.0f, 100.0f,
300.0f, 0.0f, 200.0f,
300.0f, 0.0f, 300.0f,
300.0f, 0.0f, 400.0f,
400.0f, 0.0f, 0.0f,
400.0f, 0.0f, 100.0f,
400.0f, 0.0f, 200.0f,
400.0f, 0.0f, 300.0f,
400.0f, 0.0f, 400.0f
};
Code 2 that doesn't work:
this->graphics->createBuffers(1, vboIndexsID);
this->graphics->bindBuffer(GL_ELEMENT_ARRAY_BUFFER, *vboIndexsID);
this->graphics->bufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(GLushort)*numIndexs, ind, GL_STATIC_DRAW);
this->graphics->createBuffers(1, vboVertexsID);
this->graphics->bindBuffer(GL_ARRAY_BUFFER, *vboVertexsID);
this->graphics->bufferData(GL_ARRAY_BUFFER, 300, list, GL_STATIC_DRAW);
exactly the same as before, but the list data instead of quad_vert2.
where list is:
GLfloat list[75];
int j = 0;
for(int i=0;i<numVertexs;++i) {
list[j] = v->coord.x;j++;
list[j] = v->coord.y;j++;
list[j] = v->coord.z;j++;
++v;
}
the data in v, is exactly the same values than in quad_vert2.
glGetError, after glBufferData is always 0.
If I get the data in the buffer using glGetBufferSubData the values are the same.
Ideas?
thanks in advance.
I would be suspect of the value of numVertexs, since your buffer data upload call does not actually use it. You should be uploading sizeof (GLfloat) * 3 * numVertexs bytes of data and not using the magic number 300. If you replace 300 with that expression in both situations chances are good that it will expose your real problem.
Thank you all,
the problem here was:
I was rendering two objects, with different number of vertexs. When rendering, I forgot to assign the indexs buffer correctly, and openGL was trying to draw the second object with the indexs of the first one.
Related
I am trying to program some game with OpenGL and read a whole bunch of tutorials. Unfortunately I got a small problem who just interrupts my progress.
I created a "Mesh" class where I handover an array of GLfloats. These floats are included by an VAO and VBO.
As long as I create the array inside the constructor (with the whole initialization-functions) it works fine. But if I want to handover the array as an argument OpenGL just won't draw. Did I forget something?
Here is my main code:
Mesh.cpp
Mesh::Mesh(GLfloat* vertices)
{
glGenVertexArrays(1, &m_vertexArray);
glBindVertexArray(m_vertexArray);
glGenBuffers(1, &m_vertexBuffer);
glBindBuffer(GL_ARRAY_BUFFER, m_vertexBuffer);
glBufferData(GL_ARRAY_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
glEnableVertexAttribArray(0);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (void*)0);
glBindVertexArray(0);
}
Mesh::~Mesh()
{
glDeleteVertexArrays(1, &m_vertexArray);
}
void Mesh::draw()
{
glBindVertexArray(m_vertexArray);
glDrawArrays(GL_TRIANGLES, 0, 6);
glBindVertexArray(0);
}
main.cpp
[...]
GLfloat vertices[] = {
-0.5f, -0.5f, 0.0f,
0.5f, -0.5f, 0.0f,
0.0f, 0.5f, 0.0f,
-1.0f, -1.0f, 0.0f,
0.0f, -1.0f, 0.0f,
-0.5f, 0.0f, 0.0f
};
Mesh mesh(vertices);
while (!mainWindow->isClosed())
{
glClearColor(0.0f, 0.0f, 1.0f, 1.0f);
glClear(GL_COLOR_BUFFER_BIT);
shaders->bind();
// here main functions:
mesh.draw();
mainWindow->update();
}
I handover an array of GLfloats
No, you do not "handover" an array of anything. You passed a pointer. Pointers and arrays are different things in C++. Most importantly, arrays can "decay" into pointers (that's what allows you to pass them to functions that take pointers), but when they do, all sizing information is lost.
sizeof(vertices) is the size of a GLfloat*. In all likelihood, that's either 4 or 8. It is most assuredly not the size of the array you had in the caller of the function.
The preferred method for handling this would be to pass the pointer and a size to that function. However, sizeof(vertices) will be the number of bytes in the array, not the number of array elements.
One way to handle this would be to get a std::vector rather than an array, through C++11:
std::vector<GLfloat> vertices = {
-0.5f, -0.5f, 0.0f,
0.5f, -0.5f, 0.0f,
0.0f, 0.5f, 0.0f,
-1.0f, -1.0f, 0.0f,
0.0f, -1.0f, 0.0f,
-0.5f, 0.0f, 0.0f
};
Mesh mesh(vertices.data(), vertices.size());
Alternately, you can calculate the size with some clever macros:
#define ARRAY_COUNT(arr) ( sizeof(arr) / sizeof(arr[0]))
GLfloat vertices[] = {
-0.5f, -0.5f, 0.0f,
0.5f, -0.5f, 0.0f,
0.0f, 0.5f, 0.0f,
-1.0f, -1.0f, 0.0f,
0.0f, -1.0f, 0.0f,
-0.5f, 0.0f, 0.0f
};
Mesh mesh(vertices, ARRAY_COUNT(vertices));
Or, if you want to use cleverer C++11 features:
template<typename T, size_t N>
constexpr size_t array_count(T (&arr)[N]) {return N;}
GLfloat vertices[] = {
-0.5f, -0.5f, 0.0f,
0.5f, -0.5f, 0.0f,
0.0f, 0.5f, 0.0f,
-1.0f, -1.0f, 0.0f,
0.0f, -1.0f, 0.0f,
-0.5f, 0.0f, 0.0f
};
Mesh mesh(vertices, array_count(vertices));
Or you can just skip the middle-man and employ the C++ Core Guidelines support library class span:
Mesh::Mesh(gsl::span<GLfloat> vertices)
{
glGenVertexArrays(1, &m_vertexArray);
glBindVertexArray(m_vertexArray);
glGenBuffers(1, &m_vertexBuffer);
glBindBuffer(GL_ARRAY_BUFFER, m_vertexBuffer);
glBufferData(GL_ARRAY_BUFFER, vertices.size_bytes(), vertices.data(), GL_STATIC_DRAW);
glEnableVertexAttribArray(0);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (void*)0);
glBindVertexArray(0);
}
GLfloat vertices[] = {
-0.5f, -0.5f, 0.0f,
0.5f, -0.5f, 0.0f,
0.0f, 0.5f, 0.0f,
-1.0f, -1.0f, 0.0f,
0.0f, -1.0f, 0.0f,
-0.5f, 0.0f, 0.0f
};
Mesh mesh(vertices);
I'm new to OpenGL and theres a small project I'm working on. A part of this project is create a grid which allows for some sections of the grid to have different colors.
E.g. The entire grid is green but a block on the grid is lit red and maybe another one yellow.
What I have done to draw this grid is with GL_TRIANGLE_STRIP while using indicies. After that stage, I have also included colors in the same vertex data array. But the output isn't how i want it to be.
Firstly there was an interpolation which i tried to remove by adding the 'flat' flag for the color. But there seem to be a overlapping problem. Which resulted this picture.
Is there anyway to create a grid. Where a block of the grid can be of a different color than the grid.
Update
Here is my code for GL_TRIANGLES
short* Grid::Indicies()
{
const int X_GRID_SIZE = X_GRID_SIZE_;
const int Y_GRID_SIZE = Y_GRID_SIZE_;
const int INDICIES_SIZE = (((X_GRID_SIZE * 4) + ((X_GRID_SIZE_ - 3) * 2)) * Y_GRID_SIZE);
short* indicies = new short[INDICIES_SIZE];
int index = 0;
for (size_t y = 0; y < Y_GRID_SIZE_; y++)
{
// Current, Down, Right, Down
indicies[index++] = (short)(y * X_GRID_SIZE_);
indicies[index++] = (short)((y + 1) * X_GRID_SIZE_);
indicies[index++] = (short)((y * X_GRID_SIZE_) + 1);
indicies[index++] = (short)((y + 1) * X_GRID_SIZE_);
for (size_t x = 1; x < X_GRID_SIZE_ - 1; x++)
{
// Current, Down, Current, Down, Right, Down
for (size_t i = 0; i < 2; i++)
{
indicies[index++] = (short)((y * X_GRID_SIZE_) + x);
indicies[index++] = (short)(((y + 1) * X_GRID_SIZE_) + x);
}
indicies[index++] = (short)((y * X_GRID_SIZE_) + x + 1);
indicies[index++] = (short)(((y + 1) * X_GRID_SIZE_) + x);
}
// Current, Down
indicies[index++] = (short)(((y + 1) * X_GRID_SIZE_) - 1);
indicies[index++] = (short)(((y + 2) * X_GRID_SIZE_) - 1);
}
indicies_size_ = index;
return (indicies);
}
GLfloat vertices[] = {
// Position // Color
-1.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f,
-0.5f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f,
0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f,
0.5f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f,
1.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f,
-1.0f, 0.5f, 0.0f, 1.0f, 0.0f, 0.0f,
-0.5f, 0.5f, 0.0f, 1.0f, 0.0f, 0.0f,
0.0f, 0.5f, 0.0f, 1.0f, 0.0f, 0.0f,
0.5f, 0.5f, 0.0f, 1.0f, 0.0f, 0.0f,
1.0f, 0.5f, 0.0f, 1.0f, 0.0f, 0.0f,
-1.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f,
-0.5f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f,
0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f,
0.5f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f,
1.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f,
-1.0f, -0.5f, 0.0f, 0.0f, 1.0f, 0.0f,
-0.5f, -0.5f, 0.0f, 0.0f, 1.0f, 0.0f,
0.0f, -0.5f, 0.0f, 0.0f, 1.0f, 0.0f,
0.5f, -0.5f, 0.0f, 0.0f, 1.0f, 0.0f,
1.0f, -0.5f, 0.0f, 0.0f, 1.0f, 0.0f,
-1.0f, -1.0f, 0.0f, 0.0f, 1.0f, 0.0f,
-0.5f, -1.0f, 0.0f, 0.0f, 1.0f, 0.0f,
0.0f, -1.0f, 0.0f, 0.0f, 1.0f, 0.0f,
0.5f, -1.0f, 0.0f, 0.0f, 1.0f, 0.0f,
1.0f, -1.0f, 0.0f, 0.0f, 1.0f, 0.0f,
};
GLuint v_buffer_object, v_array_object, e_buffer_object;
glGenVertexArrays(1, &v_array_object);
glGenBuffers(1, &v_buffer_object);
glGenBuffers(1, &e_buffer_object);
glBindVertexArray(v_array_object);
glBindBuffer(GL_ARRAY_BUFFER, v_buffer_object);
glBufferData(GL_ARRAY_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, e_buffer_object);
glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(grid.Indicies()) * grid.IndiciesSize(), grid.Indicies(), GL_STATIC_DRAW);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 6 * sizeof(GLfloat), (GLvoid*)0);
glEnableVertexAttribArray(0);
glVertexAttribPointer(1, 3, GL_FLOAT, GL_FALSE, 6 * sizeof(GLfloat), (GLvoid*)(3 * sizeof(GLfloat)));
glEnableVertexAttribArray(1);
glBindBuffer(GL_ARRAY_BUFFER, 0);
glBindVertexArray(0);
glPolygonMode(GL_FRONT_AND_BACK, GL_LINE);
Don't use GL_TRIANGLE_STRIP. Use GL_TRIANGLES and draw each block with 2 triangles and 4 vertices. Each vertex belongs to only 1 block. Thanks to that you don't have to use 'flat' flag, just change color for 4 vertices at once. You can try to draw each block separate, with space between each other. Then you can connect them by translating vertices and create the grid.
With GL_TRIANGLE_STRIP, flat shading will use the color from vertex i+2 to color traingle i (this is called the provoking vertex). This doesn't get you what you want. An easy option is to switch to GL_TRIANGLES, and make sure to specify your element indexes so that the same vertex is chosen third for both triangles in each quad (because the third vertex is the provoking vertex for triangles, by default). For example, suppose these are the vertexes:
0 1 2 3
4 5 6 7
8 9 10 11
12 13 14 15
Let's look at one of the quads:
0---1
| |
4---5
Let's choose vertex 0 to hold the color for the entire quad. What we then do is make sure that both triangles include vertex 0, and that vertex 0 is last in both triangles.
0---1
| \ |
4---5
So our index array will be:
5 1 0 4 5 0 ...
This uses positive (anticlockwise) winding order.
I am loosely following the very handy tutorial at opengl-tutorial.org. I've been able to create a mesh, draw a sprite to it, and rotate and scale that mesh perfectly fine.
However, I'm running into some issues when trying to translate the mesh. (pictures below)
Here's an update function on the sprite:
Transform* transform = static_cast<Transform*>(owner->GetComponent(CID_TRANSFORM));
glUseProgram(shaderID_);
glm::mat4 projection = glm::perspective(45.0f , 4.0f / 3.0f, 0.1f, 100.0f);
glm::mat4 view = glm::lookAt(
glm::vec3(3, 3, 3),
glm::vec3(0, 0, 0),
glm::vec3(0, 1, 0)
);
glm::mat4 model = transform->GetModelMatrix();
glm::mat4 mvp = projection * view * model;
GLuint matrixID = glGetUniformLocation(shaderID_, "MVP");
glUniformMatrix4fv(matrixID, 1, GL_FALSE, &mvp[0][0]);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, texture_);
glUniform1i(samplerID_, 0);
glEnableVertexAttribArray(0);
glBindBuffer(GL_ARRAY_BUFFER, vertexBuffer_);
glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (void*)0);
glEnableVertexAttribArray(1);
glBindBuffer(GL_ARRAY_BUFFER, uvBuffer_);
glVertexAttribPointer(1, 2, GL_FLOAT, GL_FALSE, 0, (void*)0);
glDrawArrays(GL_TRIANGLES, 0, 3 * 2);
glDisableVertexAttribArray(0);
glDisableVertexAttribArray(1);
And here's the getModelMatrix function:
glm::mat4 Transform::GetModelMatrix()
{
glm::mat4 trans = glm::mat4(
1.0f, 0.0f, 0.0f, translation.x,
0.0f, 1.0f, 0.0f, translation.y,
0.0f, 0.0f, 1.0f, translation.z,
0.0f, 0.0f, 0.0f, 1.0f);
float xCos = glm::cos(rotation.x);
float xSin = glm::sin(rotation.x);
float yCos = glm::cos(rotation.y);
float ySin = glm::sin(rotation.y);
float zCos = glm::cos(rotation.z);
float zSin = glm::sin(rotation.z);
glm::mat4 xRotation = glm::mat4(
1.0f, 0.0f, 0.0f, 0.0f,
0.0f, xCos, -xSin, 0.0f,
0.0f, xSin, xCos, 0.0f,
0.0f, 0.0f, 0.0f, 1.0f);
glm::mat4 yRotation = glm::mat4(
yCos, 0.0f, ySin, 0.0f,
0.0f, 1.0f, 0.0f, 0.0f,
-ySin, 0.0f, yCos, 0.0f,
0.0f, 0.0f, 0.0f, 1.0f);
glm::mat4 zRotation = glm::mat4(
zCos, -zSin, 0.0f, 0.0f,
zSin, zCos, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 0.0f,
0.0f, 0.0f, 0.0f, 1.0f);
glm::mat4 rot = xRotation * yRotation * zRotation;
glm::mat4 sca = glm::mat4(
scale.x, 0.0f, 0.0f, 0.0f,
0.0f, scale.y, 0.0f, 0.0f,
0.0f, 0.0f, scale.z, 0.0f,
0.0f, 0.0f, 0.0f, 1.0f);
return trans * rot * sca;
}
Here is the sprite at the origin viewed from (3, 3, 3).
Here is the sprite translated to (1, 0, 0) viewed from (3, 3, 3).
Matching OpenGL, GLM stores matrices in column major order. The constructors also expect elements to be specified in the same order.
However, your translation matrix is specified in row major order:
glm::mat4 trans = glm::mat4(
1.0f, 0.0f, 0.0f, translation.x,
0.0f, 1.0f, 0.0f, translation.y,
0.0f, 0.0f, 1.0f, translation.z,
0.0f, 0.0f, 0.0f, 1.0f);
To specify the matrix in the correct column major order, this needs to be:
glm::mat4 trans = glm::mat4(
1.0f, 0.0f, 0.0f, 0.0f,
0.0f, 1.0f, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 0.0f,
translation.x, translation.y, translation.z, 1.0f);
First of all here are the important parts of my code.
Creating the vertices.
D3DVertexTexture Vertices[] =
{
{-1.0f, 1.0f, 0.0f, 0.0f, 0.0f, },
{ 1.0f, 1.0f, 0.0f, 1.0f, 0.0f, },
{ 1.0f, -1.0f, 0.0f, 1.0f, 1.0f, },
{-1.0f, -1.0f, 0.0f, 0.0f, 1.0f, },
};
Creating the vertex buffer.
D3DDevice->CreateVertexBuffer(sizeof(Vertices),
0,
D3DFVF_CUSTOMVERTEXTEXTURE,
D3DPOOL_MANAGED,
&vb,
NULL);
Memory crap.
void* pVoid;
vb->Lock(0, sizeof(pVoid), (void**) &pVoid, 0);
memcpy(pVoid, Vertices, sizeof(Vertices));
vb->Unlock();
Loading the texture.
D3DXCreateTextureFromFile(D3DDevice, "images/tex.png", &t);
Rendering.
D3DDevice->SetFVF(D3DFVF_CUSTOMVERTEXTEXTURE);
D3DDevice->SetTexture(0, t);
D3DDevice->SetStreamSource(0, vb, 0, sizeof(D3DVertexTexture));
D3DDevice->DrawPrimitive(D3DPT_TRIANGLESTRIP, 0, 2);
Where is my problem.
It shows a square but the left side of the side is missing in a triangular shape like this.
Vertices A,B,C,D in a triangle strip will produce two triangles: A,B,C and B,C,D
A -- B A--B B
| | \ | /|
| | \| / |
D -- C C D--C
Look at that diagram and picture those two triangles...
Then go and put your vertices in the right order - triangle strips should 'zig-zag', not proceed in clockwise or anti-clockwise order.
If you order them: A,B,D,C - the quad will draw correctly.
Have you tried defining you vertices in this order:
D3DVertexTexture Vertices[] =
{
{ 1.0f, -1.0f, 0.0f, 1.0f, 1.0f, },
{-1.0f, -1.0f, 0.0f, 0.0f, 1.0f, },
{-1.0f, 1.0f, 0.0f, 0.0f, 0.0f, },
{ 1.0f, 1.0f, 0.0f, 1.0f, 0.0f, },
};
I believe the order in wich vertices are drawn is by default the clockwise order. You are defining in an incorrect order.
I can't work out why this code is seg faulting:
AxesMarker::AxesMarker(float size)
: size_(size), vbo_vertices_(0), vbo_elements_(0)
{
Vertex vertices[6] = {
Vertex(Color4f::RED, Vector3f::ZERO, Vector3f::ZERO),
Vertex(Color4f::RED, Vector3f::ZERO, Vector3f(size_, 0.0f, 0.0f)),
Vertex(Color4f::BLUE, Vector3f::ZERO, Vector3f::ZERO),
Vertex(Color4f::BLUE, Vector3f::ZERO, Vector3f(0.0f, size_, 0.0f)),
Vertex(Color4f::GREEN, Vector3f::ZERO, Vector3f::ZERO),
Vertex(Color4f::GREEN, Vector3f::ZERO, Vector3f(0.0f, size_, 0.0f)) };
GLuint elements[6] = { 0, 1, 2, 3, 4, 5 };
fprintf(stderr, "sizeof(vertices): %d, sizeof(Vertex): %d", (int) sizeof(vertices), (int) sizeof(Vertex));
/* create buffers */
glGenBuffers(1, &vbo_vertices_);
glGenBuffers(1, &vbo_elements_);
/* bind buffers */
glBindBuffer(GL_ARRAY_BUFFER, vbo_vertices_);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, vbo_elements_);
/* buffer data */
glBufferData(GL_ARRAY_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(elements), elements, GL_STATIC_DRAW);
/* unbind buffers */
glBindBuffer(GL_ARRAY_BUFFER, 0);
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, 0);
}
Compiles with no warnings, but appears to be seg faulting on the first call to glBufferData().
I can post more code if necessary, I'm not familiar enough with GL to know what might be relevant.
Thanks!
GLfloat vertices[60] = {
1.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
1.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f,
0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f,
0.0f, 0.0f, 1.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f };
yields same seg fault.
Is your Vertex class a plain old data type? Does it have any virtual functions which might mean it also has a vtable? Can you try re-writing this code using an array of plain floats,(just to test your calls to glBufferData are working). From what I can tell though, it looks like you are using glBufferData correctly, but then again I might have missed something.
EDIT:
Did you make absolutely sure that your OpenGL context is fully initialised before you call this code. Is this a global object, because it's constructor might be called before main?