This is exactly why we say FPS plots are of limited usefulness in some cases: If you look at the 9600K’s stock result, it averages out to 59FPS for 0.1% lows, or 91FPS for the 1% lows. Those are both good numbers and are completely acceptable. In reality, that’s from averaging results that are all over the place and outside of normal error margins. It’s jumping around between 25FPS 0.1% and 100FPS 0.1%, resulting in an average that looks reasonable, but a gameplay experience that is 50% bad. If 50% of the experience is bad, it is, by definition, not a good experience. Averages have limited usefulness here, but the AVG FPS is about 139 and 152 when overclocked, comparable to the 8700K at both stock and overclocked levels. The difference is that the 8700K has better frametime consistency and is observably better in gameplay. As for the R7 2700, that ends up at around 111FPS AVG at 4.2GHz, with lows measuring within a range of 71FPS to 87FPS. Still a wide range of variability in the title, but nothing like what we saw a moment ago. The stock performance is about 102FPS AVG.
All of this said, before the AMD fans run to post absolutes about how Ryzen is more “fluid” or “smoother” 100% of the time, note that this is not true in every instance. In many games, as you’ll see today, the 9600K doesn’t have this issue and is technically superior in raw gaming performance. Far Cry seems to be somewhat anomalous in its behavior. That’s not to downplay the issue at all or downplay Ryzen’s positioning, but people often get carried away from a single chart and draw sweeping conclusions that are untrue when considering the broader scope.