AMD has made the silent announcement that AI processors would subsequently take precedence over flagship GPUs for gaming. The company recently unveiled its new business plan, which calls for combining its RDNA gaming graphics and CNDA data center operations into a single, universal “UDNA” that prioritizes artificial intelligence.
Jack Huynh, the head of AMD’s computer and graphics division, is blunt in two interviews with Tom’s Hardware (which you should read). He clarifies that gaming graphics’ objective is to grow market share and scalability at lower price points, not to outsell Nvidia with its flagship “King of the Hill” GPUs.
The salient point from the initial interview is as follows:
Tom’s Hardware: Price point-wise, you have leadership, but you won’t go after the flagship market?
Jack Huynh: One day, we may. But my priority right now is to build scale for AMD. Because without scale right now, I can’t get the developers. If I tell developers, ‘I’m just going for 10 percent of the market share,’ they just say, ‘Jack, I wish you well, but we have to go with Nvidia.’ So, I have to show them a plan that says, ‘Hey, we can get to 40% market share with this strategy.’ Then they say, ‘I’m with you now, Jack. Now I’ll optimize on AMD.’ Once we get that, then we can go after the top.
However, he asserts that AMD wants to construct “King of the Hill” specifically for the AI data center, as it is the market segment in which the company now holds a sizable market share thanks to its EPYC CPUs and where it hopes to expand by offering the finest data center GPUs as well.
“Don’t worry, I love gaming. When I present to the board, I say gaming is a strategic pillar in my strategy,” Huynh told Tom’s Hardware later in the interview.
Based only on its financial reports, which revealed that half of the firm’s sales are currently data center products, it was reported in July that AMD is becoming an AI chip manufacturer, similar to Nvidia. Now, more quickly than ever before, both corporations have accelerated every aspect of their operations to produce new silicon designs each year. Huynh has now revealed that he hopes to follow the success of Nvidia’s own universal CUDA platform in the Tom’s Hardware interview regarding AMD’s UDNA.
While it makes sense to believe that speedier development will benefit everyone who wants a chip since new developments and maybe whole GPUs will trickle down, AMD has recently — predictably — admitted that it is unable to create all of the different types of chips at once. Nvidia may also prefer data center processors over gaming chips. Nvidia has a strong incentive to sell every graphics chip it can into a more profitable market since its next generation of GPUs is coming later than normal.
However, I do enjoy AMD’s return to the top of the value-oriented GPU market. Because Nvidia’s GPUs had some specific ugly in the $300–$400 range that most PC gamers end up picking from, and because GPUs had been growing absurdly costly compared to their performance advances even before the AI mania. If AMD can manage to convince me to give up my Nvidia 3060 Ti, that is.
AMD is shifting its focus from high-end gaming GPUs to AI processors, merging its RDNA and CNDA operations into a new “UDNA” to prioritize artificial intelligence. While aiming for market share growth with affordable gaming graphics, AMD seeks to lead AI data centers with advanced GPUs.
Stay updated here at Tech Exposed, and if you want, Buy us a coffee!
