Your first section doesn't seem to contribute to an answer. The second however relies on the idea of AMD dev's out optimizing Intel or Nvidia -- I don't consider this realistic given the huge resources each company gives to their software development in contract to AMD. Furthermore, Intel is sinking in huge resources to development and its open source -- they still outperform AMD's like offerings.
A case where Intel's open development model outperforms AMD's models are when Catalyst is made to work on linux (which is a challenge, mind you) the GPU on the a10-7850K get 9k fps on glxgears, and ~4k fps on gallium while Intel's tiny little HD 4000 gets 7.5k fps.
Despite having 512 processing elements vs. Intel's 20, Intel on any stable setup out performs AMD's with less than 1/20 processing power ratio.