Liabe Brave
Member
Indeed. As I pointed out above, even seeing the games will not ground the discussion. Because without direct developer input, we'll not be able to tell which elements of the final output may or may not be due to RPM (or any other technique). It will still all be armchair speculation, as BriareosGAF so pointedly averred.I think the discussion is so much theoretical. We should postpone it until we actually have a game that allows us to switch between usage of FP16/FP32 mixed and FP32 only to measure the differences on frame times. Otherwise it will always a back and forth of "opinions".
I responded mainly to correct LukasTaves' mistake about the sources of the discussion.