# OpenGL Creating an infinite grid in OpenGL 2.1

This topic is 893 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hello everyone,

I am very new to coding with Opengl. I am currently creating a view where it is an infinte grid. At least, infinte in the the sense that everytime the user pans, the view will update to show the grid.

For the first version, I was thinking that I would have dots to indicate the grid markings. I have some code but I am not sure what is precisely wrong with it. Maybe I am thinking about it the wrong way.

What should be happening is that when the code is executed, dots indicating the grid markings will appear. The idea is that everytime the user pans across the scene, the program will redraw the dots

void geometryEditorCanvas::drawGrid()
{
// Clears the color buffer
//	glClear(GL_COLOR_BUFFER_BIT);

//Reset to modelview matrix
//	glMatrixMode(GL_MODELVIEW);
//	glPopMatrix();

for(double i = 0; i < factor; i++)
{
for(double j = 0; j < factor; j++)
{
glBegin(GL_POINT);
glColor3d(0.0d, 0.0d, 0.0d);
glVertex2d(j * coordinateFactorWidth, i * coordinateFactorHeight);
glEnd();
}
}
}

void geometryEditorCanvas::onGeometryPaint(wxPaintEvent &event)
{
wxGLCanvas::SetCurrent(*geometryContext);// This will make sure the the openGL commands are routed to the wxGLCanvas object
wxPaintDC dc(this);// This is required for drawing

render();

drawGrid();
glFlush();
SwapBuffers();// Display the output
}



For those wondering, I am developing a 2D scene and the UI library that I am using is wxWidgets. If you need more of the code please let me know, thank you

Edited by philm001

##### Share on other sites

You say "I am not sure what is precisely wrong" but you're not describing what's going on here at all.

What are you expecting to happen?  What's not happening?  You really need to describe your problem a lot better - help us to help you!

##### Share on other sites

So.. does it not work?

If there are any dots displayed at all, post a screenshot.

If not, post the initialization of "factor" and "coordinateFactorWidth/Height" variables, as well as your matrices.

You can start troubleshooting with glLoadIdentity() for both projection and modelview, which should at least show one single dot in the center, as long as factor > 0. Once you see a dot it should just be a matter of getting the matrices and scaling right to draw a grid.

##### Share on other sites

No, it does not work. Factor is a constant number which tells me how many coordinate markings there should be in the default view for both x and y.

here is the coordinateFactorHeight/Width initilization:

coordinateFactorWidth = canvasWidth / (double)factor;
coordinateFactorHeight = canvasHeight / (double)factor;


##### Share on other sites

So factor is some number, canvasWidth is some number, and coordinateFactorWidth is some number divided by some other number.
The running theme here is that you keep not providing enough information for anyone to do anything but ask for more information.
Would it be so hard to provide the equation, and then say, “factor is 10, canvasWidth is 800”?

Did you try ::glLoadIdentity() as you were told to do?  If you are calling render() before drawGrid() it is fairly likely that the matrices are set to something strange.
Did you disable depth testing?
Are the dots a different color from the clear/background color?
Are the dots too close to the camera and being culled?  Why not push the dots back a unit?
How big are the dots?  Are you sure they are at least a pixel large?
Is coordinateFactorWidth an integer?  Is it rounding down to 0?
Did you set an orthographic projection matrix?
The first thing geometryEditorCanvas::drawGrid() should be doing is setting the model and view matrices to identity and setting an orthographic projection matrix.

glColor3d(0.0d, 0.0d, 0.0d);

This is not valid code. 0.0d = compiler error. If you want to type a constant double literal, use 0.0.
This suggests very clearly that this is not your actual code.
If you want help, post your exact actual code.

L. Spiro

Edited by L. Spiro

##### Share on other sites

No, it does not work.

This means: nothing draws?

One thing you should do is always set back to the identity when drawing the grid.

//Reset to modelview matrix
glMatrixMode(GL_MODELVIEW);

##### Share on other sites
glColor3d(0.0d, 0.0d, 0.0d);
This is not valid code. 0.0d = compiler error. If you want to type a constant double literal, use 0.0.
This suggests very clearly that this is not your actual code.
If you want help, post your exact actual code.

On the same theme, "glBegin (GL_POINT)" is also not valid code, and will also produce a compiler error.

##### Share on other sites

Yes, and some off-topic advice from a beginner like you, who made the same mistake one month ago. Don't start with old OpenGL!!!

1. First, all the matrix transformations are done using high-level functions, and you just use them like a monkey and you barely even know how 3d stuff works.

2. Literally NONE of what you learn will help you, all your code is rubbish.

3.  In my previous posts, Bregma said that the code is rubbish, but what you learn and the knowledge you get is not rubbish.   I say: That is rubbish, too!! ( or the slowest possible way of learning openGL )

4. Better get the proper math book first, so you can have a good vector background, and then you will be able to make all the matrixes on your own, or even better, use GLM.

(GLM is header only, copy/paste the stuff from the "include" are you are ready to go )

5. There was a university lecturer in my last post who said that the learning curve of new programmable pipeline( OpenGL 3.3+) is impossibly big for a beginner to understand and that he taught his students to use old openGL in the first year, and programmable pipeline in the next year. I'm a student and I say: What a waste of time!!!

My opinion: Better download a sample starter program for OpenGL 3.3+ from some tutorial, ignore the vertex and fragment shader files in the beginning, buy 2-3 books for Opengl 4.3 and use them for reference ( read and practice simultaneously ), go through all the functions, learn what they do, experiment, change, enjoy, and when you are ready, return to shaders, and trust me, it's not complicated at all once you get it.

6. You will learn really bad habits with old OpenGL, and even the shading language you will learn in the old books is deprecated. And most of the books make a whole fat chapter ( 100 or more pages ) only to display lists and other deprecated stuff, which is kind of outdated, too, at least from what I see.

All of this is just a beginner's advice, don't take it seriously if you don't feel like trusting me, I can't help someone who doesn't want to be helped, sry.
But if you do: 1 more advice from me. When you start a book, don't read all the pages, just read what you need, use it for reference, and practice a lot. If you read 800 pages at once you are going to get frustrated by all of the material and give up.

Edited by Heelp

##### Share on other sites

Hello all,

thank you for your patience and your replies. I can see that my posts are very unclear. I am a little frustrated at the code and the pressure to get things working was amounting. I very apologize to the community and I hope that you all can forgive.

If I may, please let me try again on a clean slate.

I am attempting to create a 2D CAD editor with some very basic functions to draw shapes take measurements, etc. I will not be using shaders since everything is literally going to be black and white. I know that the older versions of opengl have a number of shortcomings but since I will not be using the majority of the featurs, I figured I could get aaway with using an older version. I was also hoping that the older version would be "simplier". I am also looking for a very wide range of device support so I went with GL 2.1 (I heard that 1.1 is a very bad idea so 2.1 seemed to suffice). This CAD editor will be used in a finite Element simulator (Just to give you some context of where I am going). Thankfully, I am not going to be getting into 3D at this time and I am sticking to only 2D.

Below is the infinite grid space. I hope that someone will find this beneficial later on. Although, I am still very new to openGL programming and this was found on a basis of guess and check. So I do have some questions on why something work.

void geometryEditorCanvas::drawGrid()
{
//Reset to modelview matrix
glMatrixMode(GL_MODELVIEW);
glPopMatrix();
glPushMatrix();
glTranslatef(canvasWidth / 2.0f - (float)cameraX, canvasHeight / 2.0f - (float)cameraY, 0.0f);
for(double i = 0; i < factor; i++)
{
for(double j = 0; j < factor; j++)
{
glBegin(GL_POINTS);
glColor3d(0.0d, 0.0d, 0.0d);
glVertex2d((j * coordinateFactorWidth - canvasWidth / 2.0d ), (i * coordinateFactorHeight - canvasHeight / 2.0d));
glEnd();
}
}
}


In the function, I am selecting the model view matrix, then I am popping and pushing a matrix to the stack. I am kinda of borrowing code from an example and when they were drawing, they were poping and pushing the matrix as such. The author did not go into great detail but what is the purpose of doing this? (Also, if you see any issues with my terminology or my interpretation of anything, please point it out and correct me so that I can make sure that my understanding is correct)

Continuing on, I set the view matrix to the center of the screen (I think that what this is happening is that I am actually setting the eniter model to the center of the screen). Now we go into the whole double for loop. My idea is that one loop controls the marking for the x-axis and the other controls the markings for the y-axis. CoordinateFactorWidth is a number which describes the number of pixels between grid markings and factor is a number controlling the number of grid markings that the user will see. The glColor3d(0.0d, 0.0d, 0.0d) is crucial because this will set the points to be black (yes, this compiles and this is legit. At least, it compiles on my linux machine).

CameraX and CameraY is a variable which describes the number of pixels that the model is shifted by. The code for this is found below:

void geometryEditorCanvas::onKeyDown(wxKeyEvent &event)
{
if(event.GetKeyCode() == LETTER_W || event.GetKeyCode() == LETTER_w)
{
cameraY -= 16.0f;
}
else if(event.GetKeyCode() == LETTER_S || event.GetKeyCode() == LETTER_s)
{
cameraY += 16.0f;
}
else if(event.GetKeyCode() == LETTER_A || event.GetKeyCode() == LETTER_a)
{
cameraX -= 16.0f;
}
else if(event.GetKeyCode() == LETTER_d || event.GetKeyCode() == LETTER_D)
{
cameraX += 16.0f;
}

glMatrixMode(GL_MODELVIEW);
glPopMatrix();

glTranslated(cameraX, cameraY, 0.0f);
glPushMatrix();

this->Refresh();// This will force the canvas to experience a redraw event
}


And for those curious, yes, this is my own code that I handwritten. Some parts, I saw they were doing in the tutorial which I emulated but, I do not have much understanding why they did as such.

Above, I have an initialize routine which is setting the projection by glOrtho.

Overall, I am wondering, using openGL 2.1, is there a better method of doing this from what i have? What improvements do you see that I can implement? (improvements could be it is more efficient or the code could be structured better, etc.

Edited by philm001

1. 1
2. 2
Rutin
21
3. 3
4. 4
frob
14
5. 5

• 12
• 9
• 17
• 19
• 9
• ### Forum Statistics

• Total Topics
632598
• Total Posts
3007333

×