How does OpenGL color4 alpha works? - opengl

I am testing alpha in glColor4f, but the result of using glColor4f(1.0, 1.0, 0, 0); is same as using glColor4f(1.0, 1.0, 0, 1);.
I've tried changing glClearColor(0.0, 0.0, 0.0, 0.0); to glClearColor(0.0, 0.0, 0.0, 1.0);,it changes nothing.
Here's my code:
#include <GL/glut.h>
void init();
void display();
int main(int argc, char* argv[])
{
glutInit(&argc, argv);
glutInitDisplayMode(GLUT_RGB | GLUT_SINGLE);
glutInitWindowPosition(0, 0);
glutInitWindowSize(300, 300);
glutCreateWindow("OpenGL 3D View");
init();
glutDisplayFunc(display);
glutMainLoop();
return 0;
}
void init()
{
glClearColor(0.0, 0.0, 0.0, 0.0);
glMatrixMode(GL_PROJECTION);
glOrtho(-5, 5, -5, 5, 5, 15);
glMatrixMode(GL_MODELVIEW);
gluLookAt(0, 0, 10, 0, 0, 0, 0, 1, 0);
}
void display()
{
glClear(GL_COLOR_BUFFER_BIT);
glColor4f(1.0, 1.0, 0, 0.01);
glutWireTeapot(3);
glFlush();
}
I am wondering how did the transparency works. Thank you~

So the first thing I see is that you aren't initializing your display to use blending. this is done with the:
glEnable(GL_BLEND);
By Default, GL blending is disabled. See the description here:
http://lmb.informatik.uni-freiburg.de/people/reisert/opengl/doc/glBlendFunc.html
The next thing would be to setup your blend mode. The above link shows the routine for setting up your blend function.
Also if you look at the glutInitDisplayMode routine description here:
https://www.opengl.org/resources/libraries/glut/spec3/node12.html
it shows that "Note that GLUT_RGBA selects the RGBA color model, but it does not request any bits of alpha (sometimes called an alpha buffer or destination alpha) be allocated. To request alpha, specify GLUT_ALPHA. The same applies to GLUT_LUMINANCE."

I suppose the result of setting alpha to zero in glClearColor() you expected to be the windows with transparent background?No,it won't happen.The windows transparency is controlled by OS .See this answer.Then why glClearColor() has alpha channel clear option you would ask?That's indeed useful and sometimes critical feature when rendering between multiple offscreen frame buffers.Sometimes,when you blit (or blend)a content of one FBO into another you must clear target FBO attachment alpha as well to zero to prevent edge bleeding (actual for transparent graphics) and other nasty artifacts.But yeah,changing that value on default FBO as you do would have no effect.Hope it is clear now.
Also,you must not forget that when using such FBO texture attachments later in the pipeline, which have alpha information in them, you must use hardware alpha blending(expressed by OGL API blending functions).Otherwise the alpha info will be lost in the final result.

Related

Can't make glViewport() to work

I'm starting to work in a 2D game, but I don't know the way I'm supposed to use viewPort(). This is my Init() code:
void init(void)
{
glClear(GL_COLOR_BUFFER_BIT);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
glViewport(100, 100, 800, 600);
gluOrtho2D(0, 800, 600, 0);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glClearColor(0.0, 0.0, 0.0, 0.0);
}
What I want to know is that, after drawing an image to the window, how can I look only at a part of it, and not the whole world?
Everything you did there belongs into the drawing code. Loading textures and shaders, setting up framebuffer objects would be initialization, but you're not doing that up there.
Also glClear must come after glClearColor.
Last but not least, the viewport does not affect clear operations (you have to use scissor testing for clear operations to be limited to a certain rectangle).

What's going on with my program (openGL in c++)?

I am starting with openGL and c++, and I was wondering why I don't see anything on the window. Here is my code:
#include <GLUT/GLUT.h>
#include <stdlib.h>
void init() {
glClearColor(0, 0, 1, 0);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
gluPerspective(60.0, 1.0, 1.0, 100.0);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glTranslatef(0, 0, -10);
}
void display() {
glClear(GL_COLOR_BUFFER_BIT);
glBegin(GL_TRIANGLES);
glVertex3i(-0.5, -0.5, 0);
glVertex3i(0, 0.5, 0);
glVertex3i(0.5, -0.5, 0);
glEnd();
}
int main(int argc, char **argv) {
glutInit(&argc, argv);
glutInitDisplayMode(GLUT_SINGLE| GLUT_RGBA);
glutInitWindowPosition(200, 200);
glutInitWindowSize(400, 400);
glutCreateWindow("Window");
init();
glutDisplayFunc(display);
glutMainLoop();
}
I have a few questions:
If I run the program like this all I see is a white window... Didn't I set the color to blue?
When I do glutSwapBuffers() at the end of display function and run the program, I see the blue window without the triangle. So, I thought glutSwapBuffers() function only worked with double buffering.
And the most important, where the hell is my triangle? O.o Didn't I translate the camera with glTranslatf() function to -10 in the z-axes? If you are wondering why I used gluPerspective, I have to say that I am trying out new things, but neither works if I use gluOrtho2D().
I don't know if I am missing something or what. Maybe I need to search more information about this, but I think most of the code is correct.
1 & 2) Well you don't have to call glutSwapBuffers() when using single buffer. But you have to call glFlush(), so the draw commands are executed on the GPU.
3) I noticed that you are creating vertices with double coordinates, but you are calling integer version of glVertex** function (decimal part will be truncated) - it means that you will be drawing triangle with zero size.
Use glVertex3d() or glVertex3f() instead of glVertex3i().
Small note: intermediate mode is deprecated in the latest OpenGL.

Corrupt-looking Accumulation Buffer output

I am in the process of building a simple 3D game engine that is built on top of OpenGL, and for windowing and I/O, GLUT. I have run into a problem with the OpenGL accumulation buffer when trying to build a motion-blur option into the engine. Essentially, here is the small block of code that is supposed to do this for me:
glAccum(GL_MULT, 0.99f);
glAccum(GL_ACCUM, 1.0f - 0.99f);
glAccum(GL_RETURN, 1.0f);
I first tried this block of code by planting it in my Render() method, but it showed a corrupt-looking view where only a select few pixels were visible. So, I then tried it with the rest of the source from the website from which I found the code. I still got the same issue. Below is an image of the issue:
Then, I just took out the accumulation buffer portion (the three lines that are supposed to achieve the motion blur), and here is what I got:
Of course, there would be no motion blur since I removed the glAccum() lines, but that at least told me there is either a problem with my graphics card (it doesn't like accumulation buffers?) or those lines of code don't work.
I don't know if it matters, but I am running the code through NetBeans 7.2 (C++) on a MacBook Pro from 2011. Also, I did request an accumulation buffer in the following line:
glutInitDisplayMode(GLUT_DEPTH | GLUT_ACCUM | GLUT_DOUBLE | GLUT_RGBA);
Here is a sample piece of code I just threw together. I'm not sure if something is wrong in the code, and I know I probably didn't use best practices either, but it gets the point across. I still experienced the error with this code:
#include <iostream>
#include <GLUT/GLUT.h>
using namespace std;
float Rotation = 0.0f;
void Reshape(int width, int height)
{
glViewport(0, 0, width, height);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
glOrtho(-1, 1, -1.0f * ((float)height / (float)width), 1.0f * ((float)height / (float)width), 0.1f, 200.0f);
glMatrixMode(GL_MODELVIEW);
}
void Update(int value)
{
Rotation++;
glutPostRedisplay();
glutTimerFunc(17, Update, 0);
}
void InitGL()
{
glEnable(GL_DEPTH_TEST);
glEnable(GL_CULL_FACE);
glEnable(GL_COLOR_MATERIAL);
glClearDepth(100.0f);
}
void Render(void)
{
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glLoadIdentity();
gluLookAt(0, 0, 5.0f, 0, 0, 0, 0, 1, 0);
glPushMatrix();
{
glRotatef(Rotation, 0.0, 1.0, 0.0);
/* Render Icosahedron */
glColor3f(0.5f, 0.5f, 0.5f);
glutSolidIcosahedron();
/* Render wireframe */
glColor4f(1.0, 1.0, 1.0, 1.0);
glLineWidth(2.0);
glutWireIcosahedron();
}
glPopMatrix();
/* Blur */
glAccum(GL_MULT, 0.99);
glAccum(GL_ACCUM, 0.01);
glAccum(GL_RETURN, 1.0);
glFlush();
glutSwapBuffers();
}
int main(int argc, char** argv)
{
glutInit(&argc, argv);
glutInitDisplayMode(GLUT_DEPTH | GLUT_DOUBLE | GLUT_RGBA | GLUT_ACCUM);
glutInitWindowSize(400, 400);
glutCreateWindow("Test");
glutDisplayFunc(Render);
glutReshapeFunc(Reshape);
InitGL();
Reshape(400, 400);
glutTimerFunc(17, Update, 0);
glutMainLoop();
return 0;
}

OpenGL depth buffer isn't working

I am attempting to make a simple drawing using openGL. However, the depth buffer doesn't appear to be working.
Other people with a similar problem are typically doing one of two things wrong:
Not including glEnable(GL_DEPTH_TEST)
Bad clipping values
However, my code does not have either of these problems.
...
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
gluPerspective(25.0,1.0,10.0,200.0);
// Set the camera location
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
gluLookAt(20.0, 10.0, 50.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0);
// Enable depth test
glEnable(GL_DEPTH_TEST);
// Cull backfacing polygons
glCullFace(GL_BACK);
glEnable(GL_CULL_FACE)
drawCoordinateAxis();
drawBox(5.0,2.0,5.0,0.8,0.0,0.0);
glTranslated(1.0,-1.0,1.0); //The box is 5x2x5, it is shifted 1 unit down and 1 in the x and z directions
drawBox(5.0,2.0,5.0,0.0,1.0,1.0);
...
When I execute my code, this is drawn. http://imgur.com/G9y41O1
Note that the blue box and the red box collide, so the red box should be covering part of the blue box.
The functions drawCoordinateAxis() and drawBox() just draw a few primitives, nothing fancy inside.
I am running this on Debian squeeze.
void reshape(GLint width, GLint height)
{
g_Width = width;
g_Height = height;
glViewport(0, 0, g_Width, g_Height);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
gluPerspective(65.0, (float)g_Width / g_Height, g_nearPlane, g_farPlane);
glMatrixMode(GL_MODELVIEW);
}
So set Matrix Mode to GL_PROJECTION first, then gluPerspective.... and then back to MODELVIEW mode.

Calling glRasterPos2i and glutBitmapString in the presence of ModelView transforms

I'm trying to display a text-overlay (basically a help screen which shows my keyboard shortcuts) on top of a 3D Texture I'm rendering. The texture works great and I've got some east-to-use rotations and translations for the user.
My thought was to use
const unsigned char tmp[100] = "text to render";
glRasterPos2i(x, y);
glColor4b(255, 255, 255, 255);
glutBitmapString(GLUT_BITMAP_HELVETICA_18, tmp);
As recommended in How do I use glutBitmapString() in C++ to draw text to the screen? .
This works great except that the text now rotates with the object instead of remaining in a static location on the screen. I read some documentation and found that the glRasterPos functions are manipulated when you manipulate the model view matrix:
The object coordinates presented by glRasterPos are treated just like those of a glVertex command: They are transformed by the current modelview and projection matrices and passed to the clipping stage.
-Source
I then found via another post that you could push and pop the current matrix with glPushMatrix and glPopMatrix.
-Source
When I do this, the text disappears all together. At first I thought I might have had the coordinates wrong for the text, but I tried x=y=0 through x=y=25 in intervals of .01 and never saw the text. It's still possible I'm misunderstanding where this should be drawn, but I'm not sure what to try next.
My drawing function is calling something akin to:
glLoadIdentity();
glPushMatrix();
glTranslatef(0,0,-sdepth);
glRotatef(-stheta, 1.0, 0.0, 0.0);
glRotatef(sphi, 0.0, 0.0, 1.0);
glRotatef(rotateX,0,1,1);
glRotatef(rotateY,1,0,0);
glTranslatef(-0.5,-0.5,-0.5);
glPopMatrix();
glRasterPos2i(2, 2);
glColor4b(255, 255, 255, 255);
glutBitmapString(GLUT_BITMAP_HELVETICA_18, tmp);
Anyone have any recommendations for debug/troubleshooting steps to try to get this text to display in a single, static location on the screen?
Well, if glRasterPos is treated the same way as glVertex, then you need to set up proper projection (GL_PROJECTION) matrix (using gluOrtho2D) before calling glRasterPos.
Give this a shot:
#include <GL/glut.h>
#include <string>
using namespace std;
void display()
{
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
glColor3ub(255,0,0);
glPushMatrix();
glScalef(5,5,5);
glBegin(GL_QUADS);
glVertex2f(-1,-1);
glVertex2f(1,-1);
glVertex2f(1,1);
glVertex2f(-1,1);
glEnd();
glPopMatrix();
glColor3ub(0,255,0); // A
glRasterPos2i(0,0); // B
string tmp( "wha-hey!" );
for( size_t i = 0; i < tmp.size(); ++i )
{
glutBitmapCharacter(GLUT_BITMAP_HELVETICA_18, tmp[i]);
}
glutSwapBuffers();
}
void reshape(int w, int h)
{
glViewport(0, 0, w, h);
glMatrixMode(GL_PROJECTION);
glLoadIdentity();
double aspect_ratio = (double)w / (double)h;
glOrtho(-10*aspect_ratio, 10*aspect_ratio, -10, 10, -1, 1);
}
int main(int argc, char **argv)
{
glutInit(&argc, argv);
glutInitDisplayMode(GLUT_RGBA | GLUT_DEPTH | GLUT_DOUBLE);
glutInitWindowSize(800,600);
glutCreateWindow("Text");
glutDisplayFunc(display);
glutReshapeFunc(reshape);
glutMainLoop();
return EXIT_SUCCESS;
}
Oddly enough swapping lines A and B causes the glColor3ub() call to not take effect. I think that's what you were running into with the code sequence you posted.
As an aside glColor4b() takes chars which max out at 127. You should switch to glColor4ub() if you want to persist in passing in 255.
Documented here ("The sequence of glRasterPos(), glColor(), glBitmap() doesn't result in the desired bitmap color"), but no explanation given :(
EDIT: Ah ha! The current raster position contains its own color state, which is only updated during a glRasterPos() call.