Rumor: AMD Ryzen 3000 "Matisse" Launching July 7

Single thread performance is definitionally determined by nominal boost clock and effective IPC in single-thread-heavy workloads, the latter of which is heavily determined by cache structure and internal architecture. So different demands are going to highlight the strengths and weaknesses of respective architectures. When it comes down to brass tacks, you have to take a fine-grained look at the problem. (although AMD's and Intel's IPC in single-thread workloads appears to be converging)
 
nthexwn said:
Did you get a Skylake CPU 3 years ago? Have you played any games since then where it's obviously holding you back?
Click to expand...
Sorry for late reply. Only one game for me really uses CPU to full extent (along with the rest of the chipset) Arma 3. That was the only game I saw a decent jump in performance (I kept the same video card for the upgrade in this case). The game is notoriously hard on CPU. But it is also hard to measure performance with that game... you can get high FPS without a lot of units / AI in game (100+ FPS) But with a lot of units and scripting, the performance can drop into the teens. So as a point of reference I had in most situations Mid 30's-40's FPS before the upgrade, and high 40's to low 50's after the upgrade. So pretty significant. I think the faster (and more) memory and SSD's also played a role in improved FPS. Can't say I have seen more than a 1 or 2 FPS improvement in anything else though.
 
You must log in or register to reply here.
Back
Top