Slide on Zero Cost Abstraction confusing

[Sundaram Ramaswamy]

This slide seems like it's about simd and gpus, when it's really about abstractions (eg. classes) not introducing runtime overhead.

The intent is to show that when other languages hide basic machine, C++ allows programming dedicated hardware too i.e. it's both zero cost abstraction and close to the metal. But it seems to be not very clear.

Originally reported in Reddit.