Can computers understand sources of light well? Do they know how to project shadows/details efficiently? From my experience in video games, anti-aliasing on shadows becomes very expensive
Light and shadows are pretty tricky to capture correctly (especially when you factor in stuff like reflective/refractive surfaces). In order to do it correctly, you have to calculate how light bounces off of different parts of your image (which you'll see later in the course can be super expensive). This is actually what you'll do in projects 3-1 and 3-2!