Tuesday, May 16th 2017

AMD Announces High Performance Computing Platform - "Naples" is EPYC
Today on their Financial Analyst Day 2017, AMD has taken the lid off their "Naples" Zen implementation. The balanced Zen core in its unrestrained, server-grade level has become EPYC, with AMD CEO Lisa Su holding the silicon in her bare hands. The new EPYC platform with its I/O performance improvements allows more GPUs to be connected to a CPU than any other platform, with up to 128 PCIe lanes being expected on these server-grade chips.The Naples-based EPYC chips as announced by Lisa Su will carry up to 45% more cores, leveraging 122% greater memory bandwidth, and 60% more I/O bandwidth than their competitor's (read, Intel's) solutions. Expect more information to be available shortly.AMD is also looking at EPYC to usher in virtualization environments...Thought it would seem that what the company is actually looking to do is out-muscle Intel in the server market by offering not only greater performance, lowered power consumption and greater flexibility...While tearing Intel a new one in the platform density equation, announcing a denser platform that simultaneously offers more cores and more I/O, while taking up less space and consuming less power, which should also improve thermal characteristics and overall operating costs, which are frequently much higher throughout the lifespan of the hardware than the initial investment.AMD also promises a much simplified machine intelligence architecture compared to Intel's by fully integrating platform requirements in their EPYC processors.But this is not all: AMD is committing to a road-map of excellence, with further upgrades to its platform linearly lined-up so as to allow businesses to perform long-term planning and investment.
Source:
AMD Financial Day 2017 Webcast
40 Comments on AMD Announces High Performance Computing Platform - "Naples" is EPYC
And in regards to your last answer; Anyone choosing their 1000-core serverfarm etc. based of the NAME of the CPU should get fired instantaneously.
But OK, I'll try differently. Do you like the name Ryzen? Don't you think it was a great choice considering what we see now... even in comments on TPU: "AMD has Ryzen" and so on.
This is the same marketing idea after all. Risen->Ryzen, Epic->Epyc.
It's just that "risen" is a normal word. "Epic" is a word used by people that say "dude".
Technical person picking one CPU over the other, because one has a BETTER NAME is an incompetent idiot. This has nothing to do whatsoever with his/her private life, way he/she is dressing and pretty much anything else. There are no sysadmins in large companies (we might, perhaps, have different understanding of what "large company" means) that are choosing the brand of the server equipment, as in large companies it's not their job to do it.
Product are evaluated by a number of focused groups, including infrastructure architects and at some point purchasing department also weighs in.
CPU code name is NEVER EVER EVER taken into account as it has absolutely no practical impact.