View Single Post
Staro 19.03.2017., 09:04   #1043
Sn4k3
Premium
Moj komp
 
Sn4k3's Avatar
 
Datum registracije: Jan 2008
Lokacija: Zd
Postovi: 4,681
Quote sa PcPer:

Citiraj:
In fact, though we are waiting for official comments we can attribute from AMD on the matter, I have been told from high knowledge individuals inside the company that even AMD does not believe the Windows 10 scheduler has anything at all to do with the problems they are investigating on gaming performance.

In the process, we did find a new source of information in our latency testing tool that clearly shows differentiation between Intel's architecture and AMD's Zen architecture for core to core communications. In this way at least, the CCX design of 8-core Ryzen CPUs appears to more closely emulate a 2-socket system. With that, it is possible for Windows to logically split the CCX modules via the Non-Uniform Memory Access (NUMA), but that would force everything not specifically coded to span NUMA nodes (all games, some media encoders, etc) to use only half of Ryzen. How does this new information affect our expectation of something like Naples that will depend on Infinity Fabric even more directly for AMD's enterprise play?
Moguce da ovo boldano buni razne reviewere a s njima i mase ljudi koji ih prate...


Jos jedan zanimljiv quote sa Techpowerup, vezan uz DDR4 i Infinity Fabric:

Citiraj:
Infinity Fabric, a successor to HyperTransport, is AMD's latest interconnect technology that connects the various components on the Ryzen "Summit Ridge" processor, and on the upcoming "Vega" GPU family. According to AMD, it is a 256-bit wide bi-directional crossbar. Think of it as town-square for the chip, where tagged data and instructions change hands between the various components. Within the CCX, the L3 cache performs some inter-core connectivity. The speed of the Infinity Fabric crossbar on a "Summit Ridge" Ryzen processor is determined by the memory clock. When paired with DDR4-2133 memory, for example, the crossbar ticks at 1066 MHz (SDR, actual clock). Using faster memory, according to AMD, hence has a direct impact on the bandwidth of this interconnect.
__________________

Zadnje izmijenjeno od: Sn4k3. 19.03.2017. u 09:29.
Sn4k3 je offline   Reply With Quote