v0.5 🌳  

AMD Ryzen AI 7 Pro 160 Outperforms Previous-Gen Ryzen 9 on Geekbench

2024-07-05 15:27:39.486000

ASRock, a Taiwanese company specializing in PC components, made a significant announcement at the Computex 2024 computer expo in Taipei, Taiwan. The company unveiled new scale-up servers with the NVIDIA Blackwell architecture as part of its lineup. These servers include the Rack ORV3-NVIDIA GB200 NVL72, a liquid-cooled rack based on NVIDIA's GB200 NVL72 scale-up system, and the 6U8X-GNR2/DLC NVIDIA HGX B200, a direct-to-chip liquid-cooled GPU server [288bcc99].

In addition to ASRock's announcement, TYAN, a subsidiary of MiTAC Computing Technology Corporation, also showcased its latest AMD EPYC server platforms at COMPUTEX 2024. These platforms are optimized for data center compute performance and large-scale AI/HPC infrastructure. TYAN offers AMD EPYC 4004 CPU-based servers for smaller businesses and dedicated hosters, providing cost-effective solutions with enterprise-grade reliability, scalability, and security. The company presents various server models, including the Transport HX UT85A-B8267, Transport HX TN85-B8261, Transport HX FT65T-B8050, Transport CX TD76-B8058, Transport CX GC68C-B8056, Transport SX TS70-B8056, and Transport SX TS70A-B8056. TYAN's server platforms featuring AMD EPYC 8004 series processors are designed for cloud and edge computing tasks, delivering cost savings and energy efficiency. The company also offers the Tomcat HX S8040 and Tomcat CX S8047 server motherboards. TYAN's AMD EPYC 4004 CPU-based solutions reduce operating costs and meet the demands of hosted service providers. The company provides the Tomcat CX S8016, Transport CX HG68-B8016, and Transport CX GX40-B8016 server platforms [72cac9f5] [288bcc99].

ASRock's announcement at Computex 2024 demonstrates their commitment to innovation in the PC component industry. By partnering with NVIDIA and incorporating their Blackwell architecture, ASRock is able to offer powerful and efficient scale-up servers and GPU servers. The introduction of the Taichi & Phantom Gaming Series motherboards and the repositioning of the Taichi brand further solidify ASRock's position as a leading player in the market. TYAN's showcase of AMD EPYC server platforms optimized for data center compute performance and large-scale AI/HPC infrastructure highlights their commitment to providing cost-effective solutions with enterprise-grade reliability, scalability, and security. Together, ASRock and TYAN contribute to the advancement of server technology and cater to the diverse needs of businesses and organizations [72cac9f5] [288bcc99].

AMD also made an exciting announcement with the testing of their next-gen Zen 5-based Ryzen AI 9 365 'Strix Point' APU. The early testing confirmed AMD's claims of a 16% improvement in IPC (Instructions Per Cycle). The APU features 10 cores and 20 threads of Zen 5 power, with a boost clock of up to 5.0GHz. The Zen 5c cores operate at 3.3GHz. Testing showed a 22% relative improvement over Zen 4 and an estimated 9.7% IPC improvement. Geekbench 6 single-core performance showed a 15% IPC improvement, while Geekbench 5 saw an IPC improvement of 17.6%. This early testing validates AMD's promise of a 16% performance gain in IPC with Zen 5 [0563b182].

Furthermore, AMD's Strix Halo APUs were spotted on the FP11 (BGA) platform with a massive 128 GB memory configuration. The Strix Halo APUs are expected to have TDPs ranging from 55W to 130W and feature a 40 compute unit iGPU and 32 MB of MALL cache. These APUs combine next-gen Zen 5, RDNA 3.5, and XDNA 2 cores. The launch of the Strix Halo APUs is anticipated in the second half of 2024, further expanding AMD's APU lineup [7047f0b5] [0563b182].

According to insiders Hoang Anh Phu and Golden Pig Upgrade, AMD has reportedly moved the launch of its Ryzen AI 300 'Strix Point' APUs from July 15th to July 28th, just a day before the launch of the Ryzen 9000 CPUs. The review and sales embargoes for the Ryzen AI 300 APUs will also be lifted on July 28th. The Ryzen AI 300 lineup features a fully revised architecture with Zen 5 on the CPU, RDNA 3.5 on the GPU, and XDNA 2 on the NPU. The APUs will have up to 12 cores, 24 threads, and 16 compute units, with the fastest AI TOPS of all 'AI PC' chips. The launch of the Ryzen AI 300 APUs will be followed by the retail launch of the Ryzen 9000 Desktop CPUs on July 31st. The launch of both the APUs and CPUs will make for a busy week for reviewers and tech outlets. The Ryzen AI 300 APUs are highly anticipated and offer significant improvements in performance and features [e582714b].

In a recent development, ACER's Ryzen AI 300-powered Swift laptop has been spotted in the Geekbench database. The laptop features AMD's new Strix Point chips, including the Ryzen AI 9 HX 370. The Ryzen AI 9 HX 370 is a 12-core and 24-thread chip with boost clocks up to 5.1 GHz and integrated Radeon 890M iGPU. The Geekbench test shows single-core performance of 2,689 points and multi-core performance of 13,515 points. The ACER Swift 14 laptop will come with three platform choices: Qualcomm Snapdragon X Elite, Intel Lunar Lake, and AMD's Strix Point. The Ryzen AI 300 series laptops are expected to be released by 28 July and will be integrated by several manufacturers including ASUS, HP, Lenovo, MSI, and ACER [9dd51417].

AMD is set to release the Ryzen 9000 series, built on the Zen 5 architecture, promising a 16% improvement in instructions per clock (IPC) compared to Zen 4. The series will feature Ryzen AI, a dedicated AI engine, and support for DDR5 memory and PCIe 5.0. Accompanying the series are the X870E and X870 chipsets, offering support for USB 4.0 and AMD EXPO. The Ryzen 9000 series aims to reshape expectations and push the boundaries of desktop computing.

In the latest news, the Ryzen AI 7 Pro 160, featuring three Zen 5 cores and five Zen 5c cores, has been discovered on Geekbench. It is one of AMD's upcoming lower-end SKUs and is expected to be part of the Ryzen AI 300 series. The chip has a single-core score of 2,514 points and a multi-core score of 11,772 points, outperforming the previous generation Ryzen 9 8945HS. The L3 cache capacity is listed as 8MB, which is significantly lower than previous generation parts. The Ryzen AI 7 Pro 160 is expected to offer similar performance to the previous generation Ryzen 9 chips. It outperforms Intel's Core Ultra 9 185H but falls short of Intel's more performance-oriented mobile processors. AMD's Ryzen AI 300 series CPUs aim for a balance of power efficiency and performance for thin and light laptops [0383c16c].

Disclaimer: The story curated or synthesized by the AI agents may not always be accurate or complete. It is provided for informational purposes only and should not be relied upon as legal, financial, or professional advice. Please use your own discretion.