By: Hassan Mujtaba
AMD has said that it has plans to expand Ryzen AI across the entire product family but will be investigating where it adds the most value.
Adding AI To Ryzen Chips Is Expensive & That Might Be The Reason Why There Aren’t A Whole Lot of Mobile & Desktop Chips Integrating It (Yet!)
At CES 2023, AMD announced its dedicated AI engine from Ryzen CPUs known as Ryzen AI. As a part of its XDNA AI Engine, the Ryzen AI co-processor is an on-die and dedicated chip that can be used to boost AI capabilities. AMD said that the Ryzen AI Engine can drive up to 4 concurrent AI streams while multi-tasking and delivering up to 35% higher responsiveness compared to just a single AI stream.
But building chips with Ryzen AI has some caveats and the major one is the cost and overall value proposition. In an interview with PCWorld, AMD’s DVP & GM of Client Channel Business, David McAfee, stated that Ryzen AI is just the beginning of AMD’s AI ventures. CEO, Dr. Lisa Su, has also committed to AI as the number one strategic priority of the company.
The first chips to make use of the AMD Ryzen AI engine are the Ryzen 7040 “Phoenix” mobility processors. Not all Phoenix chips house the dedicated AI engines and there’s a good reason why AMD has made this choice. Based on the Xilinx IP, these chips can make experiences in software and operating systems such as Windows 11 better by delivering higher efficiency and better use of the hardware rather than simply burdening the CPU and GPU. Some of the features you can expect from Ryzen AI include:
Beyond this, David states that AMD is already discussing plans on how to expand its Ryzen AI stack across its Ryzen products. The major barrier to implementing Ryzen AI is the cost & that there has to be a good enough and actual reason to put Ryzen AI within budget chips & even desktop SKUs. David also raises the possibility of adding Ryzen AI to a Threadripper chip with high core counts & even then, while it might be used for training purposes, it may not necessarily use it.
With Ryzen AI deployed to just a pair of Ryzen laptop processors, the natural question is how AMD will begin distributing it to the rest of the CPU lineup. That, too, is being discussed, McAfee said. “I think we’re we’re having the AI conversation all the way across the Ryzen product line,” McAfee said.
Because of the extra cost attached to the manufacturing the Ryzen AI core, AMD is evaluating what value Ryzen AI adds, especially in its budget processors. McAfee said that the end-user benefit “has to be a lot more concrete” before AMD would add Ryzen AI to its low-end mobile Ryzen chips.
Will AMD add Ryzen AI to its desktop Ryzens? That, somewhat surprisingly, is less sure. McAfee considered the question in terms of the desktop’s power efficiency. Because of the power of the desktop, “maybe Ryzen AI becomes more of a testing tool as opposed to something that is driving the everyday value of the device,” he said. It’s possible that a high-core-count Threadripper could be used to train AI, but not necessarily use it, he added.
“I really do believe there will be a point in time in the not-too-distant future where yes, as people think about the value of their system, it’s not just my CPU and my GPU, this, this becomes a third element of compute that that does add significant value to their platform,” McAfee said.
It is pointed out that the main driver that overcomes the value & cost barrier will be software. As software evolves and makes use of AI better and adds more value, then there definitely becomes a good reason to have dedicated AI hardware blocks on your chips.
In a recent rumor, it has been pointed out that AMD is working on a future generation of EPYC CPUs that will incorporate a dedicated AI chiplet within it. Though being an EPYC chip means that it’s definitely not going to use the same design structure as the Ryzen AI block but it will definitely be based on a similar Xilinx IP. In the end, AMD definitely has a lot of AI stuff planned for the future & we can expect similar advances on the Radeon side of things.