Jump to content
  • Advertisement
Sign in to follow this  
Sethcran

OpenGL 3D Programming Bug

This topic is 2827 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I'm working on a project for one of my classes, and I'm running into an issue, and am just really unsure where to ask this. For my project, I need to create a program that reads in a .ssd file to create a 3d scene. For this project in particular, it's all about matrices, and so we are not allowed to use any matrix related functions in openGL. As it is, I've written a fair bit, and I'll try to post the segments I think are probably most relevant, but I'll upload the entire folder I'm working with as well, in case the problem lies elsewhere.

The program takes in a command line argument for a .ssd file. The .ssd files that begin with lab2_ in particular are used for this project. Some of the other files are from the first project in the course.

The main problem I'm having is that the output images I am getting are not the same as the examples, and they appear as if the camera were in the wrong position ( although I've checked, and it is reading it in from the files correctly ). From what I can see, the parsing of the file seems to be going just fine, but the problem seems to lie in the matrices themselves, or one of the functions I use to manipulate them. At this point, I am simply talking about the world -> screen transformations, and not model transformations.

Any pointers on where I am going wrong, hopefully with a hint on how to fix it, would be greatly appreciated!

Matrices are stored as [4][4] arrays, and vectors are basically just a single array of [4].

Matrix * Matrix function:
Matrix Matrix::operator*( const Matrix& rhs )
{
Matrix temp;
for( int i = 0; i < 4; i++ )
{
for( int j = 0; j < 4; j++ )
{
temp.m[j] = ( m[0] * rhs.m[j][0] ) + ( m[1] * rhs.m[j][1] ) + ( m[2] * rhs.m[j][2] )
+ ( m[3] * rhs.m[j][3] );
}
}
return temp;
}



Matrix * Vector function:
Vertex Matrix::operator*( Vertex& rhs )
{
Vertex temp;
temp.SetX( ( m[0][0] * rhs.GetX() ) + ( m[1][0] * rhs.GetY() ) + ( m[2][0] * rhs.GetZ() ) +
( m[3][0] * rhs.GetW() ) );
temp.SetY( ( m[0][1] * rhs.GetX() ) + ( m[1][1] * rhs.GetY() ) + ( m[2][1] * rhs.GetZ() ) +
( m[3][1] * rhs.GetW() ) );
temp.SetZ( ( m[0][2] * rhs.GetX() ) + ( m[1][2] * rhs.GetY() ) + ( m[2][2] * rhs.GetZ() ) +
( m[3][2] * rhs.GetW() ) );
temp.SetW( ( m[0][3] * rhs.GetX() ) + ( m[1][3] * rhs.GetY() ) + ( m[2][3] * rhs.GetZ() ) +
( m[3][3] * rhs.GetW() ) );
temp.SetColor( rhs.GetColor() );
return temp;
}


Vector cross product, division with abs value, and unary - functions:
Vertex Vertex::operator*( Vertex& rhs )
{
Vertex temp;
temp.SetX( ( y * rhs.GetZ() ) - ( z * rhs.GetY() ) );
temp.SetY( ( z * rhs.GetX() ) - ( x * rhs.GetZ() ) );
temp.SetZ( ( x * rhs.GetY() ) - ( y * rhs.GetX() ) );
temp.SetW( 1 );
return temp;
}

Vertex Vertex::operator/( Vertex& rhs )
{
Vertex temp;
if( rhs.GetX() != 0 )
temp.SetX( x / abs( rhs.GetX() ) );
else
temp.SetX( 0 );
if( rhs.GetY() != 0 )
temp.SetY( y / abs( rhs.GetY() ) );
else
temp.SetY( 0 );
if( rhs.GetZ() != 0 )
temp.SetZ( z / abs( rhs.GetZ() ) );
else
temp.SetZ( 0 );
temp.SetW( 1 );
return temp;
}

Vertex Vertex::operator-()
{
Vertex temp;
temp.SetX( -x );
temp.SetY( -y );
temp.SetZ( -z );
temp.SetW( -w );
return temp;
}


Projection matrices are partially read from file, and then constructed:
else if( keyword == "perspective" )
{
file >> angle >> near >> far;
angle = angle * ( pi / 180 );
top = abs( near ) * tan( angle / 2 );
bottom = -top;
right = top * ( screenWidth / screenHeight );
left = -right;
Matrix tempMatrix;
tempMatrix.SetMatrix(
2 / ( right - left ), 0, 0, -( ( right + left ) / ( right - left ) ),
0, 2 / ( top - bottom ), 0, -( ( top + bottom ) / ( top - bottom ) ),
0, 0, 2 / ( near - far ), -( ( near + far ) / ( near - far ) ),
0, 0, 0, 1 );
Matrix p;
p.SetMatrix(
1, 0, 0, 0,
0, 1, 0, 0,
0, 0, ( near + far ) / near, -far,
0, 0, 1 / near, 0 );
projectionMatrix = tempMatrix * p;
}
else if( keyword == "ortho" )
{
file >> right >> top >> near >> far;
left = -right;
bottom = -top;
Matrix tempMatrix;
tempMatrix.SetMatrix(
2 / ( right - left ), 0, 0, -( ( right + left ) / ( right - left ) ),
0, 2 / ( top - bottom ), 0, -( ( top + bottom ) / ( top - bottom ) ),
0, 0, 2 / ( near - far ), -( ( near + far ) / ( near - far ) ),
0, 0, 0, 1 );
projectionMatrix = tempMatrix;
}


Constructing and composing final matrix:
Matrix vp, m, cam, temp;
Vertex w, u, v;

w = -( camera.GetGazeVector() / camera.GetGazeVector() );
u = ( camera.GetUpVector() * w ) / ( camera.GetUpVector() * w );
v = w * u;

cout << w.GetX() << " " << w.GetY() << " " << w.GetZ() << endl;
cout << u.GetX() << " " << u.GetY() << " " << u.GetZ() << endl;
cout << v.GetX() << " " << v.GetY() << " " << v.GetZ() << endl;

cam.SetMatrix(
u.GetX(), u.GetY(), u.GetZ(), 0,
v.GetX(), v.GetY(), v.GetZ(), 0,
w.GetX(), w.GetY(), w.GetZ(), 0,
0, 0, 0, 1 );
temp.SetMatrix(
1, 0, 0, -camera.GetEyeX(),
0, 1, 0, -camera.GetEyeY(),
0, 0, 1, -camera.GetEyeZ(),
0, 0, 0, 1 );
cam = cam * temp;

vp.SetMatrix(
screenWidth / 2, 0, 0, ( screenWidth - 1 ) / 2,
0, screenHeight / 2, 0, ( screenHeight - 1 ) / 2,
0, 0, 1, 0,
0, 0, 0, 1 );

m = vp * projectionMatrix * cam;


Drawing the lines:
for ( int i = 0; i < (int)polyLines.size(); i++ ) 
{
glLineWidth( polyLines->GetWidth() );
glBegin( GL_LINE_STRIP );
for ( int j = 0; j < polyLines->GetNumberOfLines(); j++ )
{
Vertex temp = m * polyLines->GetVertex( j );
glColor3f( polyLines->GetVertex( j ).GetColor().GetRed(),
polyLines->GetVertex( j ).GetColor().GetGreen(),
polyLines->GetVertex( j ).GetColor().GetBlue() );
glVertex2d( temp.GetX() / temp.GetW(), temp.GetY() / temp.GetW() );
}
glEnd();
}
glFlush ();




For the full code, in case the problem doesn't actually lie in one of these bits, it can be downloaded from
http://www.dreamincode.net/forums/topic/196461-3d-programming-bug/
at the bottom of the first post as an attachment.

Share this post


Link to post
Share on other sites
Advertisement
My suggestion would be to compare your matrices to OpenGL generated ones for debugging. There's too much there to quickly spot a missed negative sign or incorrect variable.

Maybe along side your matrix calculations, use opengl equivalent functions and then use glGetFloatv() to retrieve the opengl computed matrices, and compare them element by element vs your own matrices. You should be able to spot where you make a mistake pretty quickly.

Since you're not allowed to use opengl functions, just strip them all out once your own code is working.

Share this post


Link to post
Share on other sites
Well, thank you for the suggestion, I definitely liked it, but now I'm running into a bit of trouble with the suggestion. Namely, I don't know much of anything about openGL and it's functions. After some time spent google searching, I did manage to feed in some of my matrices and test my multiplication function against openGL's, and got the same results, but how to really compare some of the other crucial parts, I really have no idea :(.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!