Archive for March, 2006

Why Microsoft is not my favorite software company

Saturday, March 18th, 2006

The grade sheet that I posted a few minutes ago is rather hard to read in Internet Explorer. It looks like the bottom few pixels of each row are cut off.

The HTML file was generated by setting a print area and choosing Save As Web Page from the File menu in Excel 2004 for Mac.

I understand cross-browser differences. I even understand cross-platform differences. But if I can’t save a web page from the latest version of one Microsoft product and have it readable from the latest version of another Microsoft product, well…

You should all be using Mozilla Firefox anyway. It looks fine there.

Assignment 5

Saturday, March 11th, 2006

Assignment 5 is available. Note that this is the last assignment due before the Midterm Exam on March 24.

Pop-quizzes and pedagogy

Tuesday, March 7th, 2006

A study from Washington University in St. Louis finds that

quizzes — given early and often — may be a student’s best friend when it comes to understanding and retaining information for the long haul

Say, that gives me an idea…

Another schedule update

Monday, March 6th, 2006

Please note that I’ve updated the Course Outline again.

For those of you wondering why I’ve been making changes to the schedule, I’ve taught the course two different ways in the past:

  1. As a lab course with weekly assignments
  2. As a lecture course with 4 or 5 larger programming assignments

In courses of type (1), we cover the basic techniques of 3D graphics and you get a lot of experience implementing them. In courses of type (2) we talk about a broader range of topics, but for most of them you never get a chance to see the code.

I’d originally planned to teach this semester as a lecture course, but when it was scheduled for a single class meeting on Friday nights a couple of things became clear:

  1. The building would be nearly empty, and we could have our pick of labs
  2. In a lecture course, everyone would be asleep by 8:45

So I’m adjusting the outline as we go along. I think we’re on track for the rest of the semester, but don’t be surprised if there’s at least one more change before the end.

Vector Math Tutorial for 3D Computer Graphics

Monday, March 6th, 2006

If you need to brush up on your math before the midterm, the Vector Math for 3D Computer Graphics from the Computer Science department at Central Connecticut State University is a pretty good review.

A correction

Monday, March 6th, 2006

If you were paying attention over the weekend, you noticed that I updated the screenshot for Assignment 4.If you compare it with the original screenshot, you’ll notice that the original wasn’t really a parallel projection.

The problem is that (as you should recall from your reading) OpenGL is a state machine. When you make a function call like gluPerspective() to update the current projection matrix, that change stays in effect until you reset the matrix (e.g., with glLoadIdentity().

My mistake was to draw the parallel projections (calling glOrtho()) without remembering to reset the projection matrix, which had already been set with gluPerspective. The result was that we took a cube in perspective, then drew a parallel projection of the perspective rendering. Oops. I didn’t notice because it looks “sort of” right — there was, in fact, a parallel projection happening.

To avoid making silly mistakes like me, do the following whenever you go to render a model (e.g., in your display callback):

  1. Set the viewport.
    glViewport(0, 0, width, height);
  2. Set up the projection.
    glMatrixMode(GL_PROJECTION);
    glLoadIdentity();
    glOrtho(-2.0, 2.0, -2.0, 2.0, 0.0, 10.0);
  3. Set your viewing parameters.
    glMatrixMode(GL_MODELVIEW);
    glLoadIdentity();
    gluLookAt(0.0, 0.0, 3.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0);
  4. Draw the scene.
    glClearColor(0.0, 0.0, 0.0, 1.0);
    glClear(GL_COLOR_BUFFER_BIT);
    
    glColor(0.0, 0.0, 1.0);
    glutWireSphere(1.0, 10, 10);

Don’t cross the streams

Friday, March 3rd, 2006

If you’re still working on Assignment 3, note that the coordinate axes (the lines in gray) should not be moving when you transform the cube. The axes should remain at the center of the world, not at the center of your model.

By the way, Assignment 4 is available.