It does need some explanation and sectioning in between the readme and code comments, for the purpose of readability. I started reading the C++ source and wasn't sure what was happening, then realised the file is humongous and stopped. Perhaps splitting it into separate files with individual explanations would help. Also worth keeping in mind that with complex or intricate code, no one has any idea what you're doing unless you explicitly tell them.

I used such a structure in a benchmark suite https://github.com/MC-DeltaT/cpu-performance-demos