How does anisotropic filtering differ from bi/tri-linear filtering?
Here is another example of anisotropic filtering: https://www.bit-tech.net/reviews/tech/aliasing_filtering/4/
From my understanding, anisotropic filtering helps insure that farther away textures are sharper, especially when considering angled surfaces.
Found a source through wikipedia, https://research.cs.wisc.edu/graphics/Courses/559-f2003/Articles/texturing.pdf#page=16 and it described some things on this slide pretty well. Anisotropic filtering algorithms are schemes that "can retrieve texel values over areas which are not square". An example would be how while mip maps (isotropic) would downsize an image of 128x128 to 64x64 and then 32x32, etc. a ripmap (an anisotropic algorithm) would also be downsized to 128x64, 128x32, etc.
One thing I am curious about though, is that I haven't been able to figure out what diagonal footprints are though (google seems to have failed me), so if anyone knows please let me know!