AMD Unveils Ryzen AI 300 Chips with Zen 5, Better GPU and Vastly Improved NPU

AMD’s next-generation laptop processors will arrive later this year, joining the new Ryzen 9000 desktop processors and ushering in another revamp in how AMD handles laptop CPU model numbers. .

But the biggest thing the company wants to boost is the new chips’ performance in generative AI and machine learning workloads: it’s putting “Ryzen AI” in the name and emphasizing the presence of an improved neural processing unit (NPU). that meets and exceeds Microsoft performance requirements for Copilot+ PCs. The new Ryzen AI 300 series follows the Ryzen 8040 chips from earlier this year, which themselves were a relatively mild upgrade to the Ryzen 7040 processors less than a year earlier.

AMD promises performance of up to 50 trillion operations per second (TOPS) with its new third-generation NPU, a significant increase over the 10 to 16 TOPS offered by the Ryzen 7000 and 8000 processors with NPUs. This would make it faster than the 45 TOPS offered by the Qualcomm Snapdragon . They all exceed Microsoft’s Copilot+ requirement of 40 TOPS, enabling some Windows 11 features that aren’t typically available on typical PCs. Copilot+ PCs can perform more AI processing locally on the device rather than relying on the cloud, potentially improving performance and giving users more privacy.

If you don’t particularly care about generative AI, running locally or otherwise, the Ryzen AI 300 processors also come with an updated CPU based on the same Zen 5 architecture as the desktop chips, plus an integrated “RDNA 3.5” GPU for boost the games. performance for thin and light systems that can’t accommodate a dedicated graphics processor.

AMD announces two chips today, both from the high-end Ryzen 9 series. The Ryzen AI 9 HX 370 includes 12 CPU cores and 16 GPU cores, up from a maximum of eight CPU cores and 12 GPU cores for the Ryzen 8040 series. The Ryzen AI 9 365 drops to 10 CPU cores and 12 GPU cores. Both have the same NPU on board.

Although an increase in the number of CPU cores suggests large improvements in multithreaded performance, note that on both chips the majority of CPU cores (8 in the 370, 6 in the 365) actually use the “Zen 5c” architecture, a variant of Zen 5 that supports exactly the same instructions and functions, but is optimized for small sizes instead of high clock speeds. The result is essentially AMD’s version of one of Intel’s E-cores, albeit without the truly heterogeneous CPU architecture that has caused incompatibility issues with some apps and games.

This isn’t the first time we’ve seen a combination of large and small CPU cores from AMD, but it is the first time we’ve seen it at the high end. Zen 4c cores only appeared in lower-end, lower-power CPU designs in the Ryzen 3 and 5 and Ryzen Z1 families.

Perhaps tellingly, AMD hasn’t offered direct comparisons between the CPU performance of the Ryzen AI 300 chips and the Ryzen 8040 series, opting instead to compare with offerings from Intel, Qualcomm, and Apple. This certainly doesn’t mean that performance has regressed generation after generation, but it’s generally code for “this isn’t the kind of improvement we want to draw attention to.”

AMD also did not offer performance comparisons between the new Radeon 890M and 880M and the old Radeon 780M. The company said the 890M was an average of 36 percent faster in a small selection of games compared to the integrated Intel Arc GPU in the Meteor Lake Core Ultra chips and 60 percent faster than the Snapdragon 3DMark Night Raid benchmark (this was part of a slide specifically highlighting the performance impact of translating x86 code onto Arm chips, although at the moment it’s true that the vast majority of games running on Snapdragon PCs will have to deal with the overhead of code translation).

AMD says Ryzen AI chips are planned to appear in “more than 100 OEM platforms” starting in July 2024, about a month after the first wave of Snapdragon X-equipped Copilot+ PCs from Microsoft and Qualcomm. Ryzen AI will also compete with Intel’s next-generation Lunar Lake chips, which are also due out later this year.

AMD listing image

Leave a Comment