1 of 1 people found this helpful
What I'm still curious about is this: AMD's Infinity Fabric detailed
Ryzen will have machine learning integrated into the processor design so it will get modestly better at recurring tasks.
The question being is that is the reason Ryzen isn't the greatest out of the box because the processors are still "dumb" and need to "learn" a game's calls and such, akin to how a SSHD's algorithms take time to learn what to load into the SSD cache.
Does Ryzen have a log of thread activity pertaining to specific, communicating threads to ensure less CCX "crosstalk"?
Is this not Microsoft or whoever Operating System manufacturer's responsibility to prevent incurred latency of CCX intercommunicating?
Since testing this review, and waiting a few days to even write the conclusion, there has been much going on about ways in which AMD’s performance is perhaps being neutered. This list includes discussions around:
- Windows 10 RTC disliking 0.25x multipliers, causing timing issues,
- Software not reading L3 properly (thinking each core has 8MB of L3, rather than 2MB/core),
- Latency within a CCX being regular, but across CCX boundaries having limited bandwidth,
- Static partitioning methods being used shows performance gains when SMT is disabled,
- Ryzen showing performance gains with faster memory, more so than expected,
- Gaming Performance, particularly towards 240 Hz gaming, is being questioned,
- Microsoft’s scheduler not understanding the different CCX core-to-core latencies,
- Windows not scheduling threads within a CCX before moving onto the next CCX,
- Some motherboards having difficulty with DRAM compatibility,
- Performance related EFIs being highly regularly near weekly since two weeks before launch."
This shows a significant performance gap between Windows 7 and 10 via Ryzen, and it could be attributed to scheduling:
He will upload the benchmark map next...
I wish Anandtech would have benchmarked it on Windows 7 like they did Kaby Lake...
Then, pcper writes this!
Here's an AMD quote from the article with my response below:
"Above all, we would like to thank the community for their efforts to understand the Ryzen processor and reporting their findings. The software/hardware relationship is a complex one, with additional layers of nuance when preexisting software is exposed to an all-new architecture. We are already finding many small changes that can improve the Ryzen performance in certain applications, and we are optimistic that these will result in beneficial optimizations for current and future applications."
Hopefully, they will resolve induced latency via CCX intercommunicating, as it is easily reproducible.
Project Mercury: Thread affinities to CCXs, SMT etc optimizations. Very light weight/efficient.
AMD Ryzen Processor Optimization added to Cacheman 10.10:
Bitsum's Process Lasso: Optimize and automate process CPU affinities:
If I were AMD I would be having a good look at these apps, benchmarking like hell with these apps and perhaps speaking to the devs to get their heads together and get an ...'official' app onto every Ryzen computer out there, as well as to all the review sites.
I have seen rumours of 50 fps increases with Project mercury in some older games and it seems to be the most automated and light weight of these apps.
I want AMD to succeed and stick it to Intel! Especially with the 12 and 16 core X399 machines.