Advanced Micro Devices executives argued Thursday on the firm’s investor day that the chip maker is gaining floor on its competitors, beginning with two main wins for high supercomputer chip offers.

Mark Papermaster, chief expertise officer at Advanced Micro Devices, mentioned that AMD beat out “heated competition” for the Lawrence Livermore National Laboratory (LLNL) El Capitan mission, the world’s quickest supercomputer, described as an “exascale class” machine. AMD additionally received the Frontier supercomputer deal from Oak Ridge National Laboratory for supply in 2021.

“We love the fight,” mentioned Papermaster, talking at an occasion at AMD’s headquarters in Santa Clara, California. “That is what we’re made of, and so that fight drove us to improve our roadmap against our competitors to go head to head, and simply beat the competition.”

AMD’s revenues have grown from $5.2 billion in 2017 to $6.73 billion in 2019. Its debt has gone from $1.7 billion in 2017 to $600 million in 2019, and money has grown from $1.2 billion to $1.5 billion.

In the El Capitan deal, AMD will present Epyc central processing items (CPUs) and Radeon Instinct graphics processing items (GPUs) for HPE’s Cray Shasta supercomputer, to be delivered in early 2023. The machine is predicted to be the world’s quickest pc, promising two exaflops of double-precision efficiency. It will probably be used for making certain the security of the nation’s nuclear weapons stockpile.

AMD’s top supercomputer wins are a big moment in ‘heated competition’

AMD’s top supercomputer wins are a big moment in ‘heated competition’

Above: Mark Papermaster is CTO of AMD.

Image Credit: Dean Takahashi

AMD will present code-named Genoa Epyc processors that includes Zen four processor cores and next-generation Radeon Instinct GPUs. Those are anticipated to be produced on a 5-nanometer manufacturing course of.

Such offers prior to now have been received by rivals Intel and Nvidia. But AMD has been on a roll for the previous couple of years, with higher efficiency with its Zen and Zen 2 CPUs and RDNA GPUs. As a outcome, AMD has been gaining market share in CPUs.

Lisa Su, CEO of AMD, mentioned the corporate has a really robust roadmap for CPUs and GPUs, and she or he mentioned AMD has the quickest chips in the marketplace, each on the gaming facet and the server facet.

Su mentioned that the corporate is in an excellent place with its 7-nanometer chips, which have seven billionths of a meter width between circuits. 7-nanometer Zen three chips are below design. AMD has shipped greater than 260 million Zen cores on its chips since 2017.

“At the end of the day, we are playing our game,” Su mentioned.

But AMD has a variety of market share to achieve in industrial consumer PCs and client PCs. PCs are a $32 billion market alternative. Zen three CPUs code-named Milan are slated to ship as Epyc three processors in 2020. A successor chip, Genoa, is predicted to debut in 2022.

AMD’s top supercomputer wins are a big moment in ‘heated competition’

Above: AMD’s chips are being utilized in El Capitan supercomputer.

Image Credit: AMD

For the $35 billion server market, Su mentioned AMD expects to double the variety of accessible server platforms based mostly on Epyc this 12 months, with greater than 140 enterprise servers anticipated to ship with Epyc. Su famous that whereas the progress of Moore’s regulation — which predicts a doubling of efficiency each two years — is slowing, AMD is exploring new packaging methods similar to X3D, which may stack chips vertically.

Su mentioned AMD has shipped over 150 million items in gaming within the present technology of sport consoles for the Microsoft Xbox One and Sony PlayStation four machines. Now it is going to present RDNA 2-based chips for the upcoming PlayStation 5 and Xbox Series X. RDNA 2 will augur within the real-time ray tracing tech pioneered by Nvidia in GPUs. Gaming is a $12 billion market alternative for AMD.

Those chips are anticipated to ship late this 12 months, after which AMD is predicted to maneuver on to RDNA three GPUs by 2022 on an unspecified “advanced node.” AMD’s David Wang, head of the Radeon enterprise, mentioned that every technology will goal a 50% enchancment in efficiency per watt, a measure of effectivity.

AMD’s top supercomputer wins are a big moment in ‘heated competition’

Above: Lisa Su, CEO of AMD, speaks at firm’s 2020 investor day.

Image Credit: AMD

Rick Bergman, government vice chairman at AMD, mentioned, “This is really a watershed moment for AMD. In three major categories of performance, AMD is the leader.”

By 2023, AMD is taking pictures for 20% compound annual development price for revenues, 50%-plus gross margins, 20% for working margin, and 15%-plus at no cost money circulation margin. Those are all proxies for significantly better monetary efficiency than AMD has been capable of obtain prior to now.

AMD’s top supercomputer wins are a big moment in ‘heated competition’

Above: AMD’s RDNA 2 graphics.

Image Credit: Dean Takahashi/AMD

AMD confirmed off a picture of what RDNA 2 graphics with real-time ray tracing would appear to be (see the picture above).

And relatively than simply make graphics chips for the datacenter, AMD is pushing its CDNA structure for datacenter GPUs. AMD expects to begin delivery Radeon Instinct GPUs with CDNA structure for the Oak Ridge National Laboratory’s Frontier supercomputers in 2021. And CDNA-based GPUs will go into the El Capitan supercomputer in 2023.

“We are the new datacenter leader,” mentioned Forrest Norrod, senior vice presiden at AMD’s datacenter and embedded enterprise group.

Patrick Moorhead, analyst at Moor Insights & Strategy, mentioned in an e mail, “Overall, AMD looks to be going for a growth play with a 20% CAGR commitment, a massive uptick over the 14% CAGR the past 5 years. Most of that growth will come from datacenter CPUs and GPUs.”

He mentioned that splitting out CDNA for the datacenter (from RDNA for gaming and client markets) is an effective transfer as “as datacenter GPUs don’t need many of the features a consumer graphics card needs.”

Moorhead added, “This includes elements like display and pixel rendering engines, and ray tracing. His means AMD can save cost by removing those elements and add more gates that help datacenter performance like tensor OPS. AMD hadn’t done it until today because it couldn’t afford to have two architectures. I believe AMD can field a high performance datacenter GPU, but it will need to invest at least as much in software to complete the solution.”