Lecture 5: Texture Mapping (42)

I think this policy is related to antialiasing because if we only took one sample of information (a comment on a lecture) from each lecture, we would miss everything that happens in between. The third comment helps yield the true shape of this curriculum woohoo!


I think that antialiasing is needed because our participation is a high-frequency signal (can vary from 0 to 1 each day). The weekly participation credit samples this signal at a low rate (1 sample per 7 days). It does supersampling by considering each day within the week, not just a single day. However, additional antialiasing is still beneficial to avoid some artifacts. For example, if I were to earn 2 points on a Sunday and 2 points 8 days later, on Monday, it would not be very accurate to represent this as a week with 2 points, a week with 0 points, then a week with 2 points. This would also make it more stressful for students, since Sunday each week would become a hard deadline. Applying a low-pass filter blurs out these high-frequency details, which makes participating on a Monday almost as good as participating on a Sunday.

You must be enrolled in the course to comment