Intel Confirms That AMD Is Major Threat To Its Enterprise Business

During last week’s Computex, the topic of EPYC came up in conversation more than once – not because it had anything to do with the event itself, but because speculation was running rampant about AMD’s potential growth. Everyone I chatted to acknowledged that no enterprise customer was going to take EPYC too seriously in the first-gen, but that some have no doubt been testing the waters, in anticipation of a more fleshed-out and even more capable EPYC 2.

We learned a couple of weeks ago that EPYC’s second release wouldn’t follow Ryzen to become a 12nm part, but instead be pushed back until 2019 (testing in late 2018) to have it release on 7nm. The company has even tapped two different foundries (GloFo and TSMC) to make sure that its execution is as smooth as possible. Meanwhile, Intel’s managed to run into a wall, having had to push its 10nm mainstream releases to 2019.

Unfortunately, while 28-core machines OC’d to 5GHz, and also the nostalgic-sounding Core i7-8086K are interesting in their own right, they’re not important to Intel like its core products are. The 28-core chip won’t be available until the end of the year, and even at its release, no vendor I talked to expected it to sell in huge quantity, but that’s to be expected with any expensive part. It’s also expected with the i7-8086K, which is purposefully limited in stock.

While Intel’s been focusing on products like that, and also things like VR demos and drones, AMD has managed to strike back hard. When EPYC first released last summer, it was of course hard to predict exactly where it’d go, but based on what we knew of Ryzen, there was a lot of potential. Apparently, that potential is now converting itself to real gains.

Intel’s 28-core desktop enthusiast chip in action

In a new interview with analyst Romit Shah, Intel CEO Brian Krzanich dropped a bombshell: An excerpt: “we thought it was interesting that Mr. Krzanich did not draw a firm line in the sand as it relates to AMD’s potential gains in servers; he only indicated that it was Intel’s job to not let AMD capture 15-20% market share.”

Intel’s bread and butter is the enterprise, and for years, it’s owned virtually all of it. AMD hasn’t even registered, something the company itself has admitted during presentations this past year. With as massive a market as the enterprise is for Intel, losing even a single percent of share is going to be noticed, and here, Krzanich admits that AMD is enough of a threat to claw back up to 15% – or at least, his company’s goal is to make sure AMD can’t grab more than that.

A little more than a sick burn

While losing market share would be detrimental to Intel, it’d prove an absolute boon for AMD if this plays out like Krzanich himself alludes. Ultimately, with whatever happens, this is good for the end consumer, even if that end consumer is dealing with enterprise hardware. As all technology generally trickles down, the better the competition, the quicker better products will reach everyone.

When Ryzen launched last spring, it felt likely that the company was going to be successful in taking back some desktop market share from Intel, and it has, even if the results are still modest for the moment. But now, we’re also seeing the same kind of disruption in the enterprise, the final proof we need that AMD really does have something special with Zen. Likewise, AMD might do well to make sure it retains Lisa Su for the long-haul, because something’s clearly going well with this setup.