1)Posts link to an article with just one gaming fps benchmark in it, GTA V. If you don't know what the problem is with a single game benchmark...
2)Said benchmark yields a 14% improvement, so it must apply to all games... yeah, that's not misleading at all.
3)All the benchmark results are jokes and misleading in themselves: a gamer isn't going to actually notice these differences in a game shooting and blowing stuff up, unless*...
-Cinebench R20: the single thread scores a literally the same, which, by the way, many games still are, and they don't care at all about multi-thread performance.
-Timespy: they're all borderline 7000pts
-Civ 6 AI turn times: they're all 7 seconds. Who's going to notice 7.1 or 7.5 seconds?
-GTA V: a 20fps difference! I'm ~SO going to notice that when I'm already getting 120+fps on a 144hz monitor.
*Oh wait, some people these days like to have these fps counters in the corners of their screens 24/7, like they're some holy grail of performance, or some junk, unawares that it is the primary cause of ruining their gaming experiences.
Those fps monitors/counters are a stealth poison.