LockeNess Monster posted:

The only thing missing is good code formatter and not freezing.

LockeNess Monster posted:

At least its not an overcomplicated confusing mess that is VS.

Avenging Dentist posted:

How is VS confusing? You basically install it and press a couple buttons and start typing. Once you learn the build/run/debug shortcuts you don't even need to look at the menus 99% of the time.

Of course, the real question then is why not pick something decent like Eclipse (ok, I'm being charitable here, but at least Eclipse doesn't usually crash), or emacs/vim?

LockeNess Monster posted:

If I need to develop in C++ i code in linux (using gedit or Eclipse, I know I should probably learn vim or emacs). If I need to port it to Windows I get it to point where I can compile it using cl.exe or in DevC++ and continue writing actual code in linux.

Avenging Dentist posted:

Then why do you even have an IDE? Why not just cross-compile?

LockeNess Monster posted:

For syntax highlighting, code folding and code formatting? Also I occasionally use refactoring tool in Eclipse.

Meh, I should probably pick up Visual Studio, but right now all of my C++ development needs are satisfied by using Eclipse in Ubuntu running in my VMWare.

OddObserver posted:

You don't need an IDE for the first 2, and real men don't write code that needs reformatting ;-). But why not use Eclipse on Windows too, if you like it?

LockeNess Monster posted:

I don't really understand all the hate for DevC++. Sure it freezes occasionally, but it is a lot simpler and more straightforward to use and learn how to use than Visual Studio. It is pretty easy to set up C++ projects in it and have them compile and run fine. The only thing missing is good code formatter and not freezing.

floWenoL posted:

If you can't figure out how to use Visual Studio, then you have no business programming C++.

niteice posted:

Also, Dev-C++ comes with GCC 3.4 and GDB 5.2, which are amazingly out of date. Like, 5 years out of date.

Whilst farting I posted:

I'm trying to do something extremely simple with OpenGL and every tutorial I can find on this is written way beyond my level of understanding and doesn't involve simple shapes. All I want to do is rotate a square (currently on the right side of the screen) around the center of its axis while moving it 40 units to the left until it hits a certain point. I can get it to move, but it's through using a very "hack"ish way of modifying the x-coordinates when I feel like transform might need to be used.

Also please don't mind the name "rect" because it's a square and I'm a bit OCD and don't like to do what would be cosmetic changes until I actually have it semi-functional.

code:

// called whenever the rectangle needs to move
void rectMove()
{
   if (rectStage != RECT_STATIC)
   {
	glBegin(GL_QUADS);
		glColor3f(0.5, 1.0, 0.5);

		glVertex3f(rectX1, 110.0, 0.0);
		glVertex3f(rectX2, 110.0, 0.0);
		glVertex3f(rectX2, 245.0, 0.0);
		glVertex3f(rectX1, 245.0, 0.0);
	glEnd();

	rectX1 += -40.0f;
	rectX2 += -40.0f;
   }

   if (rectX1 && rectX2 < 240)
         rectStage = RECT_STATIC;
}

I've tried some pushing and popping of matrices to use transform and rotate but they just give me an unmoving square. I've also read that I need to shift the object down to the center of where its center would be because it will otherwise shift from a corner? I've tried Red Book, Nehe, and countless other tutorials but nothing clearly explaining any of this - at least how I can understand.

functional posted:

If I have a void * ptr to some data is there a way to distinguish whether it points to mmaped data or malloced data? Will madvise work for this? There is also an int size that represents the number of allocated bytes.

ShoulderDaemon posted:

No. malloc might use mmap behind the scenes.

functional posted:

Is this actually true, or is munmap in practice more resilient than free()? The mapped data I am concerned about unmapping will always represent files on disk.

(In some cases I might have control over whether malloc uses mmap. What might change in the above scenarios when this is the case?)

litghost posted:

Your problem is you are trying to move the rectangle by changing the statements within you glBegin/glEnd block. This will become a problem when your geometry is nontrival and you end up using display lists or vertex buffers.

You instead want to use glTranslate and glRotate (or glMultMatrix once you know what you are doing) to transform the drawn shape.

This code example assumes the matrix mode is GL_MODELVIEW.

And FYI, my base for testing this code was NeHe Lesson 4. All you need to do is place this within the DrawGLScene().

code:

	glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);	// Clear Screen And Depth Buffer

        glPushMatrix(); // Save current matrix
	glLoadIdentity();


	float xpos = 0.5f, ypos = 0.5f, depth = -6.0f;
        float rquad = 0.0f; // Amount to rotate rectangle in degrees

	// Translate within scene
	glTranslatef(xpos,ypos,depth);

	// Rotate at translated position about the normal of the rectangle	
        glRotatef(rquad,0.0f,0.0f,1.0f);


	glColor3f(0.5f,0.5f,1.0f);

	// Draw your object
	glBegin(GL_QUADS);									// Draw A Quad
		glVertex3f(-1.0f, 1.0f, 0.0f);					// Top Left
		glVertex3f( 1.0f, 1.0f, 0.0f);					// Top Right
		glVertex3f( 1.0f,-1.0f, 0.0f);					// Bottom Right
		glVertex3f(-1.0f,-1.0f, 0.0f);					// Bottom Left
	glEnd();											// Done Drawing The Quad

	glPopMatrix(); // Restore matrix state

:edit: More details:

glTranslate is a simple transformation, it effectively moves the draw stuff from the last applied transform to a new location. glTranslate defines what x,y,z mean via the the rotation state, which is modified by glRotate.

glRotate changes the orientation of the coordinate axis of the last applied transformation. glRotate takes 4 arguments. The first is simplest to understand, it is the amount to rotate about in degrees.

The next three arguments define a vector to rotate about. In this example, it rotating about the z axis (ie unit vector [0,0,1]), which happens to be pointing towards the user's eye. However, if you had applied another rotation first, the z-axis of the rotation would be the new z-axis defined by that transformation. However no matter where the final z-axis points, it will be the normal of your rectangle, because the quad is defined in the x-y plane.

Whilst farting I posted:

sonic bed head posted:

What exactly is g++? I know that it is a compiler. But is it also a linker? Is it also a loader? I'm pretty sure that it is not a loader, but I think that it could be a loader. I am studying for a midterm in computer architecture 1 and this is a question that is coming up when reading through the "Translating and Starting a Program" chapter. I was also under the impression that loading and linking happened at the same time because linking needs to know the exact addresses. Does that mean that bash or the OS takes care of that when you actually type "./a.out"?

litghost posted:

You should learn to read the man pages, see glTranslate, and glRotate.

You also need to realize that the coordinates in OpenGL have no direct relations to pixels except in extremely specific situations, like an orthographic projection with the width and height equal to your view port size. See glOrth. Generally speaking OpenGL coordinates can be united based on physical units if you want (ft, in, m, etc.), or whatever you want. The projection matrix is how you get from model coordinates (your rectangle) to screen coordinates, kind of. I am simplifying this explaination, you can get more details here.

Whilst farting I posted:

Whilst farting I posted:

I understand how translation and rotation work, I think, now - I just don't understand why my code is not working. It'll be a lot easier for me to understand projection matrices though if I can get it moving besides manually changing the x coordinates of the quad.

For instance, I don't get why this code works, from Red Book chapter 3:

code:

draw_wheel_and_bolts()
{
   long i;

   draw_wheel();
   for(i=0;i<5;i++){
      glPushMatrix();
         glRotatef(72.0*i,0.0,0.0,1.0);
         glTranslatef(3.0,0.0,0.0);
         draw_bolt();
      glPopMatrix();
   }
}

But the code I posted above, which is virtually identical, does not.

code:

// called whenever the rectangle needs to move
void rectMove()
{
   // only animate when actually doing something
   if (rectStage != RECT_STATIC)
   {
        // commented out because the entire screen 
        // would go blank whenever it was called
        // glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);	

	glPushMatrix(); // saves the matrix as it is
	glLoadIdentity(); // loads the matrix as it is

        // moves the square 0.5 pixels to the right and up
        // and zoom out at a scale of -6
	glTranslatef(0.5f, 0.5f, -6.0f);

        // rotates on the x axis at -.75 units 
	glRotatef(-0.75f, 0.0f, 0.0f, 1.0f);

		glBegin(GL_QUADS);
			glColor3f(0.5, 1.0, 0.5);
		        glVertex3f(rectX1, 110.0, 0.0);
		        glVertex3f(rectX2, 110.0, 0.0);
		        glVertex3f(rectX2, 245.0, 0.0);
		        glVertex3f(rectX1, 245.0, 0.0);
		glEnd();
	glPopMatrix();
      }
}

covener posted:

It compiles and links, but is not a loader. On linux "man ld.so" explains runtime linking and loading and "man exec" for what happens when you're asking your shell to run an executable.

g++ posted:

lo1-189:~ ntaylor$ g++ -Wall foo.cpp
foo.cpp: In function ‘void populate()’:
foo.cpp:50: warning: the address of ‘bool checkNull(char (*)[10])’ will always evaluate as ‘true’
foo.cpp: In function ‘bool checkNull(char (*)[10])’:
foo.cpp:60: warning: NULL used in arithmetic
foo.cpp: In function ‘void checkSides(char (*)[10], int, int)’:
foo.cpp:72: warning: NULL used in arithmetic
foo.cpp:77: warning: NULL used in arithmetic
foo.cpp: In function ‘void checkDiag(char (*)[10], int, int)’:
foo.cpp:86: warning: NULL used in arithmetic
foo.cpp:89: warning: NULL used in arithmetic
foo.cpp:92: warning: NULL used in arithmetic
foo.cpp:95: warning: NULL used in arithmetic