How to proper position skybox camera using openGL - c++

Question

I created a skybox for my project and it looks the way I wanted it to; however, there are a few issues I cannot figure out how to fix and I have read some tutorials on this subject, but I was not able to find something that would help me.
The first problem is that I don't know how to get the box to always move with my camera. In the image below you can see that I am able to zoom out and see the whole box, instead of only zooming in/out of the solar system and always having the stars on the background.
The other issue I have is that when I zoom in too close my background disappears.The picture below illustrates what I mean
I know if I can get the camera working properly I can get this fixed, but it goes back to my first problem. I don't know how to access the camera info.
I believe I would have to modify glTranslatef() and glScalef() in my code from a fixed number to a number that changes with the camera position.
Here is my code:
void Skybox::displaySkybox()
{
Images::RGBImage test[6]; //6 pictures for 6 sides
test[0]=Images::readImageFile(fileName); //Top
//test[1]=Images::readImageFile(fileName);//Back
//test[2]=Images::readImageFile(fileName);//Bottom
//test[3]=Images::readImageFile(fileName);//Right
//test[4]=Images::readImageFile(fileName); //Left
//test[5]=Images::readImageFile(fileName); //Front
glEnable(GL_TEXTURE_2D);
glTexParameteri(GL_TEXTURE_2D,GL_TEXTURE_MAG_FILTER,GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D,GL_TEXTURE_MIN_FILTER,GL_LINEAR);
test[0].glTexImage2D(GL_TEXTURE_2D,0,GL_RGB);
// Save Current Matrix
glPushMatrix();
// Second Move the render space to the correct position (Translate)
glTranslatef(0,0,0);
// First apply scale matrix
glScalef(10000,10000,10000);
static const GLint faces[6][4] =
{
{5, 1, 2, 6}, // back
{5, 4, 0, 1}, // bottom
{0, 4, 7, 3}, // front
{4, 5, 6, 7}, // right ( 'left' in crinity's labeling )
{1, 0, 3, 2}, // left ( 'right' in crinity's labeling )
{2, 3, 7, 6} // top*/
};
GLfloat v[8][3];
GLint i;
v[0][0] = v[1][0] = v[2][0] = v[3][0] = -1; // min x
v[4][0] = v[5][0] = v[6][0] = v[7][0] = 1; // max x
v[0][1] = v[1][1] = v[4][1] = v[5][1] = -1; // min y
v[2][1] = v[3][1] = v[6][1] = v[7][1] = 1; // max y
v[0][2] = v[3][2] = v[4][2] = v[7][2] = -1; // min z
v[1][2] = v[2][2] = v[5][2] = v[6][2] = 1; // max z
for (i = 0; i < 7; i++)
{
//
glBegin(GL_QUADS);
glTexCoord2f(0,1); glVertex3fv(&v[faces[i][0]][0]);
glTexCoord2f(1,1); glVertex3fv(&v[faces[i][1]][0]);
glTexCoord2f(1,0); glVertex3fv(&v[faces[i][2]][0]);
glTexCoord2f(0,0); glVertex3fv(&v[faces[i][3]][0]);
glEnd();
}
// Load Saved Matrix
glPopMatrix();
}
How can I get access to these variables? Does openGL alreay have a function that takes care of that?

Answer 1

I believe I would have to modify glTranslatef() and glScalef() in my code from a fixed number to a number that changes with the camera position.
You're close, but there's a simpler solution:
Draw the skybox first, before translating the camera, so that you don't have to translate the box. Don't forget to clear your depth buffer for each new frame (you'll see why in a second).
Disable writes to the depth buffer (call glDepthMask(GL_FALSE)). This will cause every other object you render to draw over it, making it always appear "behind" everything else.
Assuming your transform matrices were reset at the start of the frame, apply only the rotation of the camera. This way the camera will always be "centered" inside the box.
Draw the skybox. Since writes to the depth buffer are off, it doesn't matter how small it is as long as it's larger than your camera's near clip plane.
Re-enable writes to the depth buffer (call glDepthMask(GL_TRUE))
Render your scene normally.

Answer 2

I haven't worked with skyboxes before, but it would make sense that the camera should always be at the center of the box. So start by translating the box to center around the camera coordinates, something like glTranslate(camera.x, camera.y, camera.z);
Then I'd think the box should stay infinitely distant, so maybe set the vertices to INT_MAX or something ridonculously big.
v[0][0] = v[1][0] = v[2][0] = v[3][0] = -INT_MAX; // min x
v[4][0] = v[5][0] = v[6][0] = v[7][0] = INT_MAX; // max x ...etc
Then probably get rid of the call to glScalef(). Try that out