Why is the ground noisy but the blocker perfectly smooth? Is it not also texture by random sampling?
I think professor mentioned we were only sampling the ground. Also I was wondering if there was any way we could use nearby location's information to calculate a the sample's lighting. It seems inefficient to independently calculate the lighting of each location, when in reality the lighting of points close to each other should be similar!
That sounds like a blurring operation maybe, to use information from neighboring areas, though this may be less accurate when there are sharp edges in our shadow.