nah, modern games really like more threads these days...There are only going to be more DX12 and Vulkan games, not less.
Is there definitive proof of that yet? In the gaming benchmarks that I've looked at, the Ryzen 1800X barely beats the Ryzen 1500X even with double the cores, double the threads, faster base clock, faster boost clock, and more power.
Take Anandtech's benches as an example.
http://www.anandtech.com/show/11244...x-vs-core-i5-review-twelve-threads-vs-four/13
At 4K resolution, it is GPU limited and there is hardly any difference so more threads didn't help. But even at 1080p resolution, the difference in average FPS was 23% (GTX 1080) and 11% (GTX 1060). Doubling the cores and threads gave no where near double the performance. The performance difference is almost the same as the 8% difference in the boost clock rate. Only if you look at some metric like time under 90 FPS was there a really big difference in double the threads with Ryzen.
Same goes for the Intel tests that I have seen. More threads doesn't seem to correlate with more performance as much as more cache does. That is, unless you are talking about gaming + other simultaneous tasks (such as live streaming).
But, I have been out of the loop for a while on gaming benchmarks, so if there is good proof that it is more threads and not more cache, then please let me know.