Sign in to follow this  
tmason

OpenGL 'Holes' with Procedurally Generated Sphere for OpenGL

Recommended Posts

Hello,

Hopefully this is a simple question; I have the following code that I am using to generate a sphere object for OpenGL. The problem is, as the picture will show, is that there are "holes" in the spheres. I placed a few boxes in the scene to show that overall my buffers/shaders seem to be OK and I believe the problem may be the indices.
 
Anyway, here is the picture:
 
Terrible_Sphere.png

And here is how I am generating the sphere:
 

float ShapeRadius = 3.0f;

	GLuint NumberOfRings = 50;
	GLuint NumberOfSectors = 50;

	float const R = 1. / (GLfloat)(NumberOfRings - 1);
	float const S = 1. / (GLfloat)(NumberOfSectors - 1);
	
	int r, s;
	
	Temp_Vertices = new std::vector<GLfloat>;
	Temp_Normals = new std::vector<GLfloat>;
	Temp_TextCoords = new std::vector<GLfloat>;

	Temp_Indices = new std::vector<GLuint>;

	Temp_Vertices->resize(NumberOfRings * NumberOfSectors * 3);
	Temp_Normals->resize(NumberOfRings * NumberOfSectors * 3);
	Temp_TextCoords->resize(NumberOfRings * NumberOfSectors * 2);

	std::vector<GLfloat>::iterator v = Temp_Vertices->begin();
	std::vector<GLfloat>::iterator n = Temp_Normals->begin();
	std::vector<GLfloat>::iterator t = Temp_TextCoords->begin();

	for (r = 0; r < NumberOfRings; r++) for (s = 0; s < NumberOfSectors; s++) {

		float const y = sin(-M_PI_2 + M_PI * r * R);
		float const x = cos(2 * M_PI * s * S) * sin(M_PI * r * R);
		float const z = sin(2 * M_PI * s * S) * sin(M_PI * r * R);

		*t++ = s*S;
		*t++ = r*R;

		*v++ = x * ShapeRadius;
		*v++ = y * ShapeRadius;
		*v++ = z * ShapeRadius;

		*n++ = x;
		*n++ = y;
		*n++ = z;
	}

	Temp_Indices->resize(NumberOfRings * NumberOfSectors * 4);

	std::vector<GLuint>::iterator i = Temp_Indices->begin();

	for (r = 0; r < NumberOfRings; r++) for (s = 0; s < NumberOfSectors; s++) {
		*i++ = r * NumberOfSectors + s;
		*i++ = r * NumberOfSectors + (s + 1);
		*i++ = (r + 1) * NumberOfSectors + (s + 1);
		*i++ = (r + 1) * NumberOfSectors + s;
	}
Vertices = Temp_Vertices->data();
    NumOfVertices = Temp_Vertices->size();

    Indices = Temp_Indices->data();
    NumOfIndices = Temp_Indices->size();

    Normals = Temp_Normals->data();
    NumOfNormals = Temp_Normals->size();

    TextureCoordinates = Temp_TextCoords->data();
    NumOfTextureCoordinates = Temp_TextCoords->size();

What can I do to get rid of the holes?

 

Thank you.

Share this post


Link to post
Share on other sites

Start by breaking out of the loop early, first after the first iteration, then after the second, and so on, for both the sectors and the ring.

That should go from no polygons to a single triangle at the top of the sphere, and then if that looks OK then go to a full ring, and when that looks OK go to 2 rings.

It should make it much easier to isolate the problem, than trying to see why the full sphere looks like it does.

 

In this case I'm reasonably sure your problem is in building the indices. Are you drawing with GL_TRIANGLES?

If so you need 6 indices per quad, as it's built from 2 separate triangles. From your indices you need to draw a separate quad or triangle-strip for each sector.

If it still looks strange, do glDisable(GL_CULL_FACE) and see if it fixes it, which would mean face-winding is inverted.

Share this post


Link to post
Share on other sites

The problem definitely is the indices (all the points seem to be in the right place)

 

It looks like you are generating quads, but are drawing it as triangles, as Erik suspects.

Edited by Olof Hedman

Share this post


Link to post
Share on other sites

After executing your code I think the problem is indeed in drawing 'GL_QUADS' geometry with 'GL_TRIANGLES'. Take a look at the following screenshots of your geometry.

[attachment=23463:Triangles_BackfaceCulling.jpg] Fig.1. Drawing with GL_TRIANGLES and GL_CULL_FACE enabled.

[attachment=23464:Triangles.jpg] Fig.2. Drawing with GL_TRIANGLES and GL_CULL_FACE disabled.

[attachment=23465:Quads_BackfaceCulling.jpg] Fig. 3. Drawing with GL_QUADS and GL_CULL_FACE enabled.

 

Share this post


Link to post
Share on other sites

Start by breaking out of the loop early, first after the first iteration, then after the second, and so on, for both the sectors and the ring.
That should go from no polygons to a single triangle at the top of the sphere, and then if that looks OK then go to a full ring, and when that looks OK go to 2 rings.
It should make it much easier to isolate the problem, than trying to see why the full sphere looks like it does.
 
In this case I'm reasonably sure your problem is in building the indices. Are you drawing with GL_TRIANGLES?
If so you need 6 indices per quad, as it's built from 2 separate triangles. From your indices you need to draw a separate quad or triangle-strip for each sector.
If it still looks strange, do glDisable(GL_CULL_FACE) and see if it fixes it, which would mean face-winding is inverted.


Changing to GL_QUADS per @Rootus's suggestion worked! Thanks.

One problem though; on Intel-based systems I get the sphere with no problems but on AMD based graphics cards nothing shows up.

I haven't tried the code on nVidia based systems.

Any reason why this may be?

Share this post


Link to post
Share on other sites

After executing your code I think the problem is indeed in drawing 'GL_QUADS' geometry with 'GL_TRIANGLES'. Take a look at the following screenshots of your geometry.

attachicon.gifTriangles_BackfaceCulling.jpg Fig.1. Drawing with GL_TRIANGLES and GL_CULL_FACE enabled.

attachicon.gifTriangles.jpg Fig.2. Drawing with GL_TRIANGLES and GL_CULL_FACE disabled.

attachicon.gifQuads_BackfaceCulling.jpg Fig. 3. Drawing with GL_QUADS and GL_CULL_FACE enabled.

Thank you for trying this out!

 

Hope to get this fully working on all systems like AMD.

 

So far I only see stuff on Intel...

Share this post


Link to post
Share on other sites

One problem though; on Intel-based systems I get the sphere with no problems but on AMD based graphics cards nothing shows up.

If you're using 'core' profile, GL_QUADS is deprecated. That might be why AMD driver refuses to draw anything. The solution would be to add the indices for the missing triangle for each quad you're generating (and draw with GL_TRIANGLES).

Edited by TheChubu

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this