|07 July 2013||#31|
Performance Technology Supervisor
Join Date: Jul 2002
That's not quite how it works.
You can decide to compile to a sequential assembly, and good c++ compilers will automatically target assembly when they can for the sake of performance (although it takes a fair bit of domain knowledge to tweak things so they will in an optiomal fashion, and it's compiler dependent, but it's not uncommon for billion run loops to be structured so that they will be compiled optimally), but most C++ compilers will actually push to object files for the linker to piece together into a deliverable binary.
Assembly isn't the same as binary, and it's not a mandatory target for compilers. Plenty languages/compilers don't ever even get remotely close to it.
Even when it's an option (IE: GCC's S flag) you receive an asm file that then takes another step to make it to contact with the actual metal.
It's probably not that ironic, but yeah, if you want to take it philosophically at the end of the day the end target is switching transistors. The human managed abstraction part is important though and it's why we have so many languages and the whole eco system with enough variety for all the different tasks.
Come, Join the Cult http://www.cultofrig.com - Rigging from First Principles
|Thread Closed share thread|