I am sure Intel's part will outperform in any form with 4 less core and less PCI lanes than Epyc, Intel knows well in what they are competing with. They are not going to lose their server market share as some speculators suspect.
You can't take Intel's perf claims as a fact and do note that those perf claims over the previous gen are distorted by better storage.
"Skylake's new AVX-512 vector processing extensions far outstrip Epyc's abilities in floating-point intensive jobs."
AVX-512 is not magic and you can't generalize to FP perf. Optimized code that can take advantage of it might run faster on Intel but that's not the case with most FP loads, look at 3rd party benchmarks. Zen has higher throughput than Skylake in anything but 256 and 512 bit vector code (http://www.agner.org/optimize/blog/read.php?i=838). Pair that with the extra memory bandwidth and the ability to utilize faster storage (more PCIe gives one that option) and Epyc will perform. Ofc there is the downside of the die to die latency in some applications.
To save this item to your list of favorite EE Times content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
28 
