The first chapter is fairy simple, to the point it's perfectly understandable for the general public, while the second is way more complex and supposed to be for graphic developpers, given it show bits of code/functions. Maybe it's why they have separate it in two short chapters...
I could rant a little about the explanations in the first chapter, (they don't explain what fragments are, the explanation about why tiles architectures have reduced total memory access is not wrong but hide importants things under the carpet, immediate-mode graphic card can also not render hidden polygons with earkly-Z,), but well, not today.