HWUB: https://static.techspot.com/articles-info/1877/bench/CSGO.png https://static.techspot.com/articles-info/1877/bench/CSGO.pn... LTT (not OC'd): https://i

HWUB: https://static.techspot.com/articles-info/1877/bench/CSGO.png https://static.techspot.com/articles-info/1877/bench/CSGO.pn... LTT (not OC'd): https://i.redd.it/fnw15b6dxv831.png https://i.redd.it/fnw15b6dxv831.png The doubled cache really made an enormous difference in performance. Zen2 averaged about a 15% IPC improvement on average, in games the average was more like 30%, and in CS:GO it was about 42% improvement. It went from being pounded in this title to tying or beating the 9900K. And that's against a 2700X, the 1800X was significantly worse than that. Compared to the 1800X, CS:GO is more like 50-60% improved. The slightly higher clocks don't hurt, but Zen2 really made huge progress in handling the "unhappy cases" for their architecture. I tend to point the finger at cache for this specific case, cause I don't think of source engine titles when I think clean, well-optimized, highly threaded code.