Hi, I have been reading the OpenGl programmers guide v1.1. In my program I am just drawing a simple rectangle, and experimenting with gluLookAt and glTranslatef. Here's my code :

	gluLookAt(-0.25,-0.25,0.0, 0.0,0.0,0.0, 0.0,0.0,0.0);

	//glTranslatef(0.25,0.25,0.0);

	glBegin(GL_POLYGON);
		glVertex3f(0.0,0.0,0.0);
		glVertex3f(0.5,0.0,0.0);
		glVertex3f(0.5,0.5,0.0);
		glVertex3f(0.0,0.5,0.0);
	glEnd();

I have read in the book that glTranslatef can be used inplace of gluLookAt . But here when I use gluLookAt in place of glTranslatef it does not give me the same effect. Can somebody pls explain me where I am going wrong? Thanks. --eminemence.

Well I tried this:
gluLookAt(-0.25,-0.25,0.0, 0.0,0.0,1.0, 0.0,1.0,0.0);
And the result is that now i can see the rect but it seems to have scaled
instead of translating..
Thanks.
--eminemence.

gluLookAt combines a few rotations and translation depending on the parameters. It is used to position and rotate the camera, not the polygons. The first 3 params are a vector that represents where the camera is position. The second is a vector that represents a point that the camera is looking at. In front of the camera. The third is the vector representing UP usually up is 0, 1, 0 but if you roll in 3d up may very well be 0, -1, 0 if you are actually upside down. The function then translates and rotates the camera to get what you are looking for.

You can use your matrices or use rotation/scale etc.. functions within OpenGL. However, combining them isn't always pretty. I usually see it one way or the other.

Quote:Is there any benefit to doing the matrix math yourself?

speedwise no, though its useful sometimes to know what matrix the scene is viewing (hmm exmplained bad)

Quote:I have read in the book that glTranslatef can be used inplace of gluLookAt

no, with translate u dont 'really' supply a difrection

**no, with translate u dont 'really' supply a difrection**
translate can be used instead of gluLookAt, but you have to glRotate as well. gluLookAt basicly does this anyway.About using your own matrices, the only thing I ever need to add my own matrix for is shadows(I think the technique is called projective shadows). You create a shadow matrix based on the position of a light and a plane. Then you translate like normal, glMultMatrix(the matrix), then draw like normal(black, no texture, blended, maybe offset a little from the plane), this matrix basicly plasters all the vertices onto the plane according to the light position. This is the only technique I use my own matrices for. In order to use quaternions, you would have to use your own matrices as well, but I haven't needed quats yet. For either of these cases, you don't have to retrieve the current matrix. Just MultMatrix and OpenGL does it for you. If there is a great need to get the matrix, it may be faster to implement your own, but I dn't have a use for that as of yet, maybe for physics calculations or something like that.