… In a blog post titled "Agentic AI Needs Rack-Scale CPU Performance – AMD EPYC Delivers It Today," AMD argues that thread-, core-, or even socket-level performance comparisons aren't really meaningful because they don't describe what a customer actually deploys. …
… According to testing published at Phoronix , performance results for NVIDIA's Vera CPU were striking: across a geometric mean of enterprise workloads, Vera outpaced AMD’s top-tier high-frequency EPYC 9575F server chip by roughly 11%. …
… Linux kernel patches for HighestFreq from May. Finally, PQOS Global Bandwidth Enforcement is really more of an EPYC server CPU feature than a desktop Ryzen delight, but basically it allows the firmware to limit how much memory bandwidth individual threads are allowed to use. …
… 384 threads = 288 1.333..., which is larger than the 30% performance claim. So the EPYC part is actually more efficient overall, interestingly. …
… Spotting Nova Lake in the CI lab is interesting, but the really fascinating part is what the Linux boot logs say about the chips' capabilities: Indeed, not only does this confirm support for AVX10 vector states on both the P-cores and E-cores as otherwise there would be no need to inform the operat… …