freeglut/ glu 3d drawing - c++

Question

I'm trying to draw a blue quad across the bottom of a room.
This is the the code I've been attempting to use, but there is no quad, just the green clear color.
#include <iostream>
#include <vector>
#include <cmath>
#include <GL/glu.h>
#include <GL/freeglut.h>
bool keystates[256];
bool speckeystates[256];
// manage key events
void keys(unsigned char key, int x, int y) {
keystates[key]=true;
}
void keyup(unsigned char key, int x, int y) {
keystates[key]=false;
}
void skeys(int key, int x, int y) {
speckeystates[key]=true;
}
void skeyup(int key, int x, int y) {
speckeystates[key]=false;
}
void draw(void) {
glLoadIdentity();
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
gluLookAt(0,0,0,640,480,640,0,1,0); // from near-top-left to far-bottom-right, floor should be visible
glBegin(GL_QUADS);
glColor3ub(15,15,255); // visible blue color
glVertex3i(0,480,0); // draw quad across bottom
glVertex3i(640,480,0);
glVertex3i(640,480,640);
glVertex3i(0,480,640);
glEnd();
glFlush(); // flush drawing
glutSwapBuffers(); // swap the buffers
glutPostRedisplay();
}
int main(int argc, char** argv) {
glutInit(&argc,argv);
glutInitDisplayMode(GLUT_DOUBLE | GLUT_RGB | GLUT_DEPTH);
glutInitWindowSize(640,480);
glutInitWindowPosition(0,0);
glutCreateWindow("Window!"); // init window stuff
glutDisplayFunc(draw); // our frame/draw event
glutKeyboardFunc(keys);
glutKeyboardUpFunc(keyup);
glutSpecialFunc(skeys);
glutSpecialUpFunc(skeyup); // init key event stuff
glEnable(GL_DEPTH_TEST); // init opengl stuff
glClearColor(0.2f,0.3f,0.2f,1.0f);
gluPerspective(45.0,640/480,0,640); // x: 0-640 y: 0-480 z: 0-640
glDepthFunc(GL_LEQUAL);
glutMainLoop(); // begin the loop
}

Answer 1

I am going to answer your question, but first I'd advise you to try and forget about the glBegin function as well as glVertex/color etc. This is a very outdated API from the time when a draw call (e.g. a transfer of data from RAM to graphics memory) was less expensive than the rendering itself, so it didn't matter that you were inefficiently sending data because it took the graphics card more time to process whatever you sent it the last frame than it took the CPU and DMA to get the data from one memory to the other.
A few years later, the exact opposite was true, so the standard expanded to include VBOs so you don't need to send each vertex via a separate draw call, but they kept the immediate mode (which is what you are using) for compatibility reasons.
Nowadays the immediate mode is (finally) deprecated, so it's a bad idea to use it, you can find some decent tutorials about the new API here
Anyway your problem is the way you setup your camera, you stand in the point 0,0,0 and look at the 640,480,640 point which means that the vertices will be defined in the wrong order, causing them to get culled, either change the glulookat call and inverse the target and the eye positions, or change the culling mode