Lighting GLSL 1.5

Graphics and GPU Programming Programming OpenGL

Started by Asem December 05, 2009 01:10 PM

5 comments, last by Asem 14 years, 4 months ago

154

Author

December 05, 2009 01:10 PM

I'm having a weird problem with my lighting. I think the cause is the normals which I feed a mat3x3 to the shader to multiply against the normal going into the shader. I'm trying to attempt just opengl 3.2 core functionality. My matrices are correct the only matrix that's most likely not correct is my normal matrix. I was reading that the normal matrix is the inverse of the model matrix transposed but doing that doesn't seem to change anything. PROBLEM: Whatever the case the cube is lighted on the inside instead of the outside. Yes the uniforms do get the data as I've tested each one to make sure. I tried switching the normals around but it did nothing. The cube is hard-coded so I set 36 vertices to get all normal directions. An image of what's going on: Here


//Vertex Shader

#version 150 core


uniform transform
{
	mat4 MVP;
	
} Transform;

in attrib
{
	vec3 Position;
} Attrib;

uniform vec3 lightPosition;

uniform mat3 NormalMatrix;

in  vec3 Normal;

out vec3 normal; 
out vec3 lightDir;
out vec3 eyeDir;



void main()
{

	vec4 vVertex = Transform.MVP * vec4(Attrib.Position, 1.0);
	vec4 LP = Transform.MVP * vec4(lightPosition, 1.0);
	vec3 N = NormalMatrix * (Normal);
	
	
	normal = N.xyz;
	lightDir = LP.xyz - vVertex.xyz;
	eyeDir   = -vVertex.xyz;
	
	gl_Position = vVertex;
}



//Fragment Shader

#version 150 core

out frag
{
	vec4 Color;
} Frag;

uniform vec4 color;
uniform vec4 specularcolor;
uniform vec4 ambient;
uniform float shine;


in vec3 normal; 
in vec3 lightDir;
in vec3 eyeDir;

void main()
{
	vec4 matColor = vec4(1.0, 1.0, 1.0, 1.0);

	vec4 finalColor = ambient;
	
	vec3 N = normalize(normal);
	vec3 L = normalize(lightDir);
	
	float lambertTerm = dot(N, L);
	
	if( lambertTerm > 0.0 )
	{
		finalColor = color * matColor * lambertTerm;
		
		vec3 E = normalize(eyeDir);
		vec3 R = reflect(-L, N);
		
		float specularpower = pow( max( dot(R, E), 0.0 ), shine);
		
		finalColor += specularcolor * vec4(0.8, 0.8, 0.8, 1.0) * specularpower;
	}


	Frag.Color = finalColor;

}


//Code

    Mat4X4lookAtCam = LookAtMatrix(movement[0], movement[1], movement[2], movement[4], 0.0f, 0.0f, 0.0f, 0.0f, 1.0f);
    gluLookAt(movement[0], movement[1], movement[2], movement[4], 0.0f, 0.0f, 0.0f, 0.0f, 1.0f);

    Mat4X4 MVP;
    Mat4X4 world;
    Mat4X4 scale;
    Mat4X4 trans;
    Mat4X4 rot;

    scale = MatrixScale(1.0f, 1.0f, 1.0f);
    trans = MatrixTranslation(0.0f, 0.0f, 0.0f);
    world = scale * trans;

    MVP =  world * lookAtCam * projection;
 
    Mat3X3 NormalMatrix;
    NormalMatrix.identityClear();

    NormalMatrix = MatrixInverse( MVP );
    NormalMatrix = MatrixTranspose( NormalMatrix );


    ////////////////////////
    ////////////////////////
    ////////////////////////

    glColor3f(0.0,1.0,0.0);
    glPointSize(10.0);
    glBegin( GL_POINTS );
        glVertex3f(2.0, 2.0, 3.0);
    glEnd();

    lightProp.color = Vec4(1.0f, 1.0f, 1.0f, 1.0f);
    lightProp.ambientColor = Vec4(0.2f, 0.2f, 0.2f, 1.0f);
    lightProp.specColor = Vec4(0.0f, 0.0f, 1.0f, 1.0f);
    lightProp.shine = 2.0f;

    glProgramUniformMatrix3fvEXT(pointlight_shader_15->GetHandle(), UniformNormalMatrix, 1, GL_FALSE, &NormalMatrix.matrixA[0]);

    glProgramUniform3fEXT(pointlight_shader_15->GetHandle(), uniformLP, 2.0f, 2.0f, 3.0f);

    glProgramUniform4fvEXT(pointlight_shader_15->GetHandle(), LColor, 1, lightProp.color.getXYZW());
    glProgramUniform4fvEXT(pointlight_shader_15->GetHandle(), LSpec , 1, lightProp.specColor.getXYZW());
    glProgramUniform4fvEXT(pointlight_shader_15->GetHandle(), LAmb  , 1, lightProp.ambientColor.getXYZW());
    glProgramUniform1fEXT (pointlight_shader_15->GetHandle(), LShine, lightProp.shine);

    pointlight_shader_15->UseShader();

    glBindBuffer( GL_UNIFORM_BUFFER, UniformBufferTransPHandle );
        glBufferSubData( GL_UNIFORM_BUFFER, 0, sizeof(MVP), &MVP.matrixA[0]);

    glBindBuffer(GL_UNIFORM_BUFFER, 0);

    myCube->Draw(); //Calls a VAO. Also though it's not shown I did make sure that the attributes were set correctly for position and normal

    pointlight_shader_15->StopShader();

Erik Rufelt

5,903

December 05, 2009 01:39 PM

Perhaps your vertices are in the wrong order?
Try switching between glFrontFace(GL_CW) and glFrontFace(GL_CCW), or glDisable(GL_CULL_FACE) to disable face culling.

Asem

154

Author

December 05, 2009 02:39 PM

I thought that would do the trick but it doesn't affect anything. Does glFrontFace affect some matrix?

Erik Rufelt

5,903

December 05, 2009 04:05 PM

No, it decides which faces are visible. Triangles are usually considered having a front and a back side, and your image seems to indicate that you see the faces on the opposite side of the cube (the inside should be back-facing, if the cube is supposed to be viewed from the outside). The problem could also be with depth-testing, in case the further faces are drawn after the first ones, without face-culling. Do you enable depth-testing with glEnable(GL_DEPTH_TEST), and clear the depth buffer with glClear(GL_DEPTH_BUFFER_BIT)?

bubu LV

1,436

December 05, 2009 04:37 PM

Original post by Asem

    glColor3f(0.0,1.0,0.0);    glPointSize(10.0);    glBegin( GL_POINTS );        glVertex3f(2.0, 2.0, 3.0);    glEnd();

What is this? This is definitely not 3.2 core.

Asem

154

Author

December 05, 2009 07:41 PM

I know what glFrontFace does. I kinda made it seem like I didn't but for whatever reason it didn't work. GL_DEPTH_TEST is enabled too.

Yes, I know that is not opengl 3.2 core code. I did it to be clear on where the light is positioned.

I have it on compatibility mode and I avoid it in pretty much all my code except that one bit of code. I didn't feel like trying to create some type of buffer for one point and etc although I probably could have just used the geometry shader to pull that off. Even though I'm using the compatibility mode this doesn't stop the shader from working. Also if I try to use the core opengl crashes if I use glSubBuffer which I find really weird. (nvidia 9500 gt)

I made 2 separate buffers. 1 for the position data and another for the normal data because I didn't want to think about offsetting the data but I do make sure they match.

This should be okay right? I might have missed something that has completely skipped my mind since I don't really run into this situation.

//Cube Data//Bottom    vertices[0]  = vector_math::Vec3<float>(-1.0f,   1.0f, -1.0f); normals[0]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);    vertices[1]  = vector_math::Vec3<float>( 1.0f,  -1.0f, -1.0f); normals[1]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);    vertices[2]  = vector_math::Vec3<float>(-1.0f,  -1.0f, -1.0f); normals[2]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);    vertices[3]  = vector_math::Vec3<float>(-1.0f,   1.0f, -1.0f); normals[3]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);    vertices[4]  = vector_math::Vec3<float>( 1.0f,   1.0f, -1.0f); normals[4]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);    vertices[5]  = vector_math::Vec3<float>( 1.0f,  -1.0f, -1.0f); normals[5]  = vector_math::Vec3<float>(0.0f, 0.0f, -1.0f);//Top    vertices[6]  = vector_math::Vec3<float>(-1.0f,   1.0f, 1.0f);  normals[6]  = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);    vertices[7]  = vector_math::Vec3<float>( 1.0f,  -1.0f, 1.0f);  normals[7]  = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);    vertices[8]  = vector_math::Vec3<float>(-1.0f,  -1.0f, 1.0f);  normals[8]  = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);    vertices[9]  = vector_math::Vec3<float>(-1.0f,   1.0f, 1.0f);  normals[9]  = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);    vertices[10] = vector_math::Vec3<float>( 1.0f,   1.0f, 1.0f);  normals[10] = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);    vertices[11] = vector_math::Vec3<float>( 1.0f,  -1.0f, 1.0f);  normals[11] = vector_math::Vec3<float>(0.0f, 0.0f, 1.0f);//Left    vertices[12] = vector_math::Vec3<float>(-1.0f,  -1.0f,  1.0f); normals[12] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);    vertices[13] = vector_math::Vec3<float>( 1.0f,  -1.0f, -1.0f); normals[13] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);    vertices[14] = vector_math::Vec3<float>(-1.0f,  -1.0f, -1.0f); normals[14] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);    vertices[15] = vector_math::Vec3<float>(-1.0f,  -1.0f,  1.0f); normals[15] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);    vertices[16] = vector_math::Vec3<float>( 1.0f,  -1.0f,  1.0f); normals[16] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);    vertices[17] = vector_math::Vec3<float>( 1.0f,  -1.0f, -1.0f); normals[17] = vector_math::Vec3<float>(0.0f, -1.0f, 0.0f);//Right    vertices[18] = vector_math::Vec3<float>(-1.0f,  1.0f,  1.0f);  normals[18] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);    vertices[19] = vector_math::Vec3<float>( 1.0f,  1.0f, -1.0f);  normals[19] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);    vertices[20] = vector_math::Vec3<float>(-1.0f,  1.0f, -1.0f);  normals[20] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);    vertices[21] = vector_math::Vec3<float>(-1.0f,  1.0f,  1.0f);  normals[21] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);    vertices[22] = vector_math::Vec3<float>( 1.0f,  1.0f,  1.0f);  normals[22] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);    vertices[23] = vector_math::Vec3<float>( 1.0f,  1.0f, -1.0f);  normals[23] = vector_math::Vec3<float>(0.0f, 1.0f, 0.0f);//Back    vertices[24] = vector_math::Vec3<float>(-1.0f,  -1.0f,  1.0f); normals[24] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);    vertices[25] = vector_math::Vec3<float>(-1.0f,   1.0f, -1.0f); normals[25] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);    vertices[26] = vector_math::Vec3<float>(-1.0f,  -1.0f, -1.0f); normals[26] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);    vertices[27] = vector_math::Vec3<float>(-1.0f,  -1.0f,  1.0f); normals[27] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);    vertices[28] = vector_math::Vec3<float>(-1.0f,   1.0f,  1.0f); normals[28] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);    vertices[29] = vector_math::Vec3<float>(-1.0f,   1.0f, -1.0f); normals[29] = vector_math::Vec3<float>(-1.0f, 0.0f, 0.0f);//Front    vertices[30] = vector_math::Vec3<float>( 1.0f,  -1.0f,  1.0f); normals[30] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    vertices[31] = vector_math::Vec3<float>( 1.0f,   1.0f, -1.0f); normals[31] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    vertices[32] = vector_math::Vec3<float>( 1.0f,  -1.0f, -1.0f); normals[32] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    vertices[33] = vector_math::Vec3<float>( 1.0f,  -1.0f,  1.0f); normals[33] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    vertices[34] = vector_math::Vec3<float>( 1.0f,   1.0f,  1.0f); normals[34] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    vertices[35] = vector_math::Vec3<float>( 1.0f,   1.0f, -1.0f); normals[35] = vector_math::Vec3<float>( 1.0f, 0.0f, 0.0f);    for(int i=0; i<36; i++)        indices = i;    glBindBuffer(GL_ARRAY_BUFFER, vertexBuffer);        glBufferData(GL_ARRAY_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);    glBindBuffer(GL_ARRAY_BUFFER, normalBuffer);        glBufferData(GL_ARRAY_BUFFER, sizeof(normals), normals, GL_STATIC_DRAW);    glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, iBuffer);        glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(indices), indices, GL_STATIC_DRAW );    glBindVertexArray(vaoBuffer);        glBindBuffer(GL_ARRAY_BUFFER, vertexBuffer);        glVertexAttribPointer(attribVertex, 3, GL_FLOAT, GL_FALSE, sizeof(vector_math::Vec3<float>), 0);        glBindBuffer(GL_ARRAY_BUFFER, normalBuffer);        glVertexAttribPointer(attribNormal, 3, GL_FLOAT, GL_FALSE, sizeof(vector_math::Vec3<float>), 0);        glEnableVertexAttribArray(attribVertex);        glEnableVertexAttribArray(attribNormal);        glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, iBuffer);    glBindVertexArray(0);

[Edited by - Asem on December 5, 2009 7:41:12 PM]

Asem

154

Author

December 06, 2009 02:12 PM

solved it. I thought the vertices were in the right order as I've checked over them many times but I was wrong oops. Thank you guys for the suggestions though. [freaking duh on my part!]

Lighting GLSL 1.5

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Lighting GLSL 1.5

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Reticulating splines