Intel’s Core Ultra 200V “Lunar Lake” processors have allegedly been delayed to the last quarter of 2024. Forming the heart of Team Blue’s “AI PC” marketing campaign, these chips are going to offer over 100 TOPS of inferencing throughput, with a 45 TOP NPU unit. The report comes from DigiTimes, claiming that the Lunar Lake lineup has been delayed by a quarter, with shipments commencing in September, instead of June as originally planned.
This delay (if true) would place Intel at a disadvantage against AMD and Qualcomm whose next-gen notebook designs with advanced NPU units have already started shipping. AMD announced its Ryzen AI 300 “Strix Point” processors at Computex earlier this month with similar NPU performance.
The Ryzen AI 300 series led by the Ryzen AI 9 HX 370 and the Ryzen AI 9 365 represent AMD’s first (true) hybrid core processors with a mix of 4x Zen 5 and 8x Zen 5c cores. The accompanying Radeon 890M iGPU features 16 CUs or 1,024 shader cores, offering nearly RTX 3050-level performance. The NPU unit, powered by the XDNA 2 architecture, offers 50 TOPS of AI inferencing throughput.
Lunar Lake will have a whole node advantage over Strix Point as it leverages TSMC’s 3nm (N3B) process, while Strix Point is based on 4nm (N4P). The RDNA 3.5 integrated graphics, a faster NPU unit, and a common ISA across all cores will be AMD’s primary advantages over Intel’s Core Ultra 200 offerings.
