But the biggest benefit is if you use a Core Affinity app and set the game to just run on set cores with no HT. Latency can have an impact as well as 元 cache in fact 元 cache is likely the biggest factor anything that doesn't fit into cache will get pushed to the system memory so Ryzen X3D chips will likely have bigger gains. So yeah it scales with a memory to a certain point but not so much now. Going from 5 GHz on my 8700K down to 4.6 now had a bigger impact than the ram. It did scale quite well with memory back in the day but memory bandwidth on modern systems is high enough that its back to being thread limited. with DDR4 it stopped scaling around DDR4 3000 MHz on my 8700K from DDR4 3000 which I run now vs 3666 MHz there is zero FPS difference. Faster memory helps but will never compensate for that. Game wants to nom nom on more 元 cache to keep the CPU thread properly fed with data. If you can't scale clock speed then scaling bandwidth helps but suffice to say the game is still a buggy mess on ancient code.Īs it is with the 7950X3D with same settings etc vs a 7950X with higher clocks but same memory the extra cache shows a 20-30% improvement in performance. So while it will distribute load the 1 important thread is the main culprit. tends to smooth things out and stops the game from swapping between so many cores on which its single thread bound anyway. Click to expand.That was back in the DDR3 era.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |