NeHe OpenGL Lesson07 - Texture filtering, lighting & keyboard control

This sample shows us how to bring texture mapping, different texture filtering, lighting and Widget keyboard control together.
Keyboard ‘page up’ or ‘page down’ to zoom out or zoom in. Key ‘L’ toggle light on or off. Key ‘F’ switch different texture filtering modes.
As you may notice that there is some artifact with the screen shot. When I press ‘Print SysRq’ key on my board, I found that the whole program pause for a while. That may be the reason, the ‘Print screen’ function could not finished during one frame, it last for several frames. So the print screen buffer cache several frame results, this cause such artifact.

 

OK, let’s dive into the key source code.
The following is the code set up the light source with OpenGL:

// set up the lighting
GLfloat lightAmbient[4] = {0.5f, 0.5f, 0.5f, 1.0f};
GLfloat lightDiffuse[4] = {1.0f, 1.0f, 1.0f, 1.0f};
GLfloat lightPosition[4] = {0.0f, 0.0f, 2.0f, 1.0f};

glLightfv(GL_LIGHT1, GL_AMBIENT, lightAmbient);
glLightfv(GL_LIGHT1, GL_DIFFUSE, lightDiffuse);
glLightfv(GL_LIGHT1, GL_POSITION, lightPosition);

glEnable(GL_LIGHT1);

As I remember there is a maximum number limit for the light sources, eight(8) may be the value.  So you may found that some program has a special class named “LightingManager”, this class will response for finding the most closest 8 light sources for each object. But this situation becomes much better when the additive lighting (blend with additive operation) came out.  All the lighting calculation happen in the vertex shader stage or vertex lighting, lighting calculated based on each vertex.

 

There is a super high command to switch on or off those light sources in OpenGL:

if ( !mUseLight )
    glDisable(GL_LIGHTING);
else
    glEnable(GL_LIGHTING);

 

The following code used to create texture objects with OpenGL:

// set up the second texture – use linear sampling
glBindTexture(GL_TEXTURE_2D, mDiffuse[1]);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexImage2D(GL_TEXTURE_2D, 0, 3, texture.width(), texture.height(),
        0, GL_RGBA, GL_UNSIGNED_BYTE, texture.bits());

// set up the third texture – use the mipmap texture
glBindTexture(GL_TEXTURE_2D, mDiffuse[2]);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR_MIPMAP_NEAREST);
gluBuild2DMipmaps(GL_TEXTURE_2D, GL_RGB, texture.width(), texture.height(),
        GL_RGBA, GL_UNSIGNED_BYTE, texture.bits());

As you see from the last passage of the source code, a texture object created with mip maps, and corresponding filter mode set. I have one questions here: how the OpenGL will generate the mip-map? Does it use point down sampling or linear down sampling or some sharp method to make the normal map become very strong? A good tool to compress texture and generate different kind of mip maps is nvCompress that from the nvdia web site.

And the program coded for the Ubuntu 10.4. To get it compiled ok and run you need to install gcc, cmake, opengl on your computer.

 

The full source code could be found here.