The Scorpio x4.5 figure is based on raw flops comparisons right and disregards additional problems devs faced in using ESRAM?
Is it fair to say that ESRAM isn't factored into this comparison, i.e. without proper optimisation games won't run as well as if they were written for unified memory, and therefore, despite the flops differences some games run worse than the xbox one flop number indicates?
e.g. cod ghosts ran at 720p, compared with the 1080 ps4 output but this is because of esram?
Is it fair to say that ESRAM isn't factored into this comparison, i.e. without proper optimisation games won't run as well as if they were written for unified memory, and therefore, despite the flops differences some games run worse than the xbox one flop number indicates?
e.g. cod ghosts ran at 720p, compared with the 1080 ps4 output but this is because of esram?