I'm experimenting with applying textures to objects. I have a 256x256 .bmp file that I'm using as a texture. I can apply this texture to a simple quad using: glTexCoord2f(0.0, 0.0); glVertex3f(0.0, 0.0, 0.0); glTexCoord2f(0.0, 1.0); glVertex3f(0.0, 256.0, 0.0); glTexCoord2f(1.0, 1.0); glVertex3f(256.0, 256.0, 0.0); glTexCoord2f(1.0, 0.0); glVertex3f(256.0, 0.0, 0.0); And I can make a cube by repeating this 6 times for each face. But my question is how do I go about applying this texture to more complex shapes such as a pyramid, sphere or torus? (these would be drawn using the GLUT function rather than by hand). I want to 'stretch' the texture to wrap around the entire object rather than repeat the texture. Is there a simple function that will do this or would I have to do it manually by stepping through the 2D array storing the texture and altering it by hand? If there is no simple function I will be sticking to a 2D shape (because I'm running out of time!), so I would want to just apply the texture to various 2D shapes; triangles, circles or random polygons for example. Again, is there a simple function that can acheive this or do I have to alter the data manually? Thanks for any advice.

Hi,

as far as I know all "glut" primitives like glutSolidSphere ... don´t have predefined texture coordinates but I might be wrong.
The only exception as far as I know is the glutSolidTeapot.
So I the only way to texture more complex objects is to apply them by hand which might end up in a real pain, or to use a (simple) file format and an editor to place textures onto your objects.
A rather simple format for static objects might be the wavefront *.obj format.
Hope that helps,

Ralph

Thanks Ekast.

So I wont be using and of the pre-defined glut primitives! What about the posibility of using the texture to "stretch around" a simple 3D object, like a cube or pyramid that I have defined? How can that be acheived?

I cant really use any special file formats or editors because this assignment was all to do with creating a .bmp file and using that as a texture, and texture mapping it onto an object and I'm already going a bit off course!!

Thanks again.

Hi,

For your defined primitives like a sphere or easier a cylinder it´s quite easy to define texcoords. Think of a cylinder defined by 3 variables (r, theta, z)
Where r is the radius of your cylinder, theta is the angle of the current point on the cylinder and z is the current height of the point.
For further explanation on cylindrical coordinates look here:

Cylindrical Coordinates
For describing coordinates on a sphere another coordinate system might be more suitable:
Spherical Coordinates

For mapping the texture onto the cylinder just the angle theta and the height z have to be taken into account.
If you want to map the texture around the cylinder exaclty on time you have to calculate the u-coordinate like that:

u = theta / 360

Assuming the cylinder "stands" vertically with theta given in degrees.
Retrieving v is rather easier, v is zero at the bottom and 1 at the top of the cylinder. For any vertices between bottom and top just interpolate the values.

A Cube can be seen as a cylinder with "just" 4 faces on the side so just treat it as a cylinder.

For mapping a pyramid just think of the pyramid as a "special" kind of cylinder with all top vertices merged into one vertex.

For any other arbitrary surfaces it is probably the easiest to use an editor though.

I hope that answer was sufficient and helped you a bit.

Regards

Ralph

Thanks again Ekast.

The problem I have is more to do with the actual coding, rather than the maths behind it. As in, how do I use the functions like:
glTexCoord2f(0.0, 0.0); glVertex3f(0.0, 0.0, 0.0);
glTexCoord2f(0.0, 1.0); glVertex3f(0.0, 256.0, 0.0); etc

I think I understand how I could calculate the vertex data for the glVertex3f(0.0, 0.0, 0.0); part - I could just replace the 3 variables with r, z and theta (in the cylinder example), but how does the TexCoord part vary? I dont really understand how this function works.

Would it be easier to use the physical pixel data from the 2D array or is the above method best?

Thanks