In this depiction, some of the tiny triangles modeling the surface are smaller than the size of a single pixel. How is this possible? Is the pixel not the smallest level of granularity you can get when modeling an image (i.e. either filled in or not)?

Does this illustration depict a projection of a model rendered on a higher resolution screen onto a lower resolution, resulting in different pixel sizes?

I'm assuming this is a zoomed-in, high resolution depiction of the screen.

I have the same question as @lwzt