At Computex Taipei 2024, AMD unveiled the next generation of "Strix Point" or "Ryzen AI 300" series APUs, which are poised to be one of the most revolutionary processor designs in the mobile space, catering to the craze for AI-powered PCs.
The new Strix Point APU features the latest Zen 5 architecture and the RDNA 3 update, which the company calls RDNA 3.5. It also includes the improved XDNA 2 AI engine, which delivers the highest TOPS of any NPU released this year or coming soon, boasting 50 dedicated AI compute functions. These advancements enable Strix Point APUs to deliver more efficient and powerful AI computation, enhancing the overall user experience.
AMD plans to introduce the Ryzen AI PRO series designed for professional users, including Ryzen 9 PRO AI HX and Ryzen 7 PRO AI. These PRO variants feature enhanced AI computing capabilities, offering higher performance and functionality compared to the base models.
Surprisingly, even the Ryzen 7 PRO features 12 cores, whereas the standard Ryzen AI 300 variant only contains 10 cores. The high-end Ryzen AI 9 HX SKU also comes with 12 cores. Currently, there are four models in the Strix Point series: the Ryzen AI 9 HX 370 and the AMD Ryzen AI 9 365, which have already been launched and will feature in new laptop designs debuting in July.
Additionally, there has been recent news about RDNA 4. AMD has just introduced a new set of patches for the forthcoming Linux 6.11 release, focusing on the upcoming RDNA 4 "Radeon RX 8000" GPU. These patches further enhance support for the Linux platform, providing increased performance and new features.
AMD has accelerated support for RDNA 4 on Linux platforms. The patches include multiple tweaks and introduce support for Incremental Color Compression (DCC), a technology that improves rendering efficiency by leveraging redundancy in color data and reducing memory bandwidth usage. This improvement is crucial for enhancing the performance of RDNA 4 GPUs on Linux platforms.
AMD's "IP Block" approach plays a significant role in this update. This approach allows AMD to roll out pre-release support more quickly, enabling developers and users to test and optimize their applications before the official release. The patch also includes performance optimizations addressing issues related to SR-IOV and power management, further improving system stability and efficiency.
Note: SR-IOV (Single Root I/O Virtualization) is a technology that enables a single computer hardware device, such as a network card or storage controller, to be used by multiple virtual machines simultaneously. Each virtual machine has its own operating system and applications. With SR-IOV technology, a physical device can be divided into multiple virtual devices, each resembling a separate piece of hardware that can be directly assigned for use by different virtual machines. This allows VMs to share the same physical device while maintaining efficient and independent performance.
SR-IOV is primarily used in cloud computing and virtualization environments, such as data centers and cloud computing platforms. This technology improves the network and storage I/O performance of virtual machines, delivering more efficient services to users. In data centers, it enables better management and allocation of hardware resources, enhancing overall system performance and efficiency. In summary, SR-IOV technology significantly improves performance and flexibility in virtualized environments through efficient resource allocation and management mechanisms, making it one of the key technologies in modern data centers and cloud computing platforms.
With the launch of the RDNA 4 GPU patch and the release of the Ryzen AI 300 series APUs, AMD is further solidifying its position in the Linux community. These new products and technologies not only enhance AMD's capabilities in graphics and AI computing but also provide users with a more efficient and powerful experience. We can expect to see more key updates and improvements in the future that will deliver even better performance and user experiences.
