Meteor Lake vs. Lunar Lake: Intel’s Shifting Manufacturing Strategy

At COMPUTEX 2024, Intel unveiled technical details of its Lunar Lake architecture, a processor designed for next-generation AI PCs, part of the Core Ultra 200 series, slated for release in the third quarter of 2024. Lunar Lake represents a significant leap in graphics and AI processing capabilities, focusing on enhancing the high-efficiency computing performance of ultra-thin laptops. Compared to its predecessors, Lunar Lake will reduce SoC power consumption by up to 40% and deliver over three times the AI computational power.

Anandtech reported that Lunar Lake’s entire computing module, including P-Core and E-Core, is manufactured using TSMC’s N3B process, while the platform control module uses TSMC’s N6 process. Intel’s Foveros packaging technology is also utilized, without any in-house fabrication. It’s worth noting that even the currently marketed Meteor Lake, part of the Core Ultra 100 series, still uses Intel’s own Intel 4 process for its computing modules.

The development of Lunar Lake can be seen as a collaborative effort between Intel’s architecture design team and TSMC’s process nodes, featuring the new Lion Cove architecture P-Core and Skymont architecture E-Core, significantly enhancing performance and efficiency. This combination might be what many enthusiasts have been eagerly awaiting. In recent years, AMD has gained a certain advantage in chip manufacturing processes through TSMC. As Intel adopts TSMC’s most advanced processes, its architectural design level may be better reflected in performance comparisons. By the end of the year, Intel will also introduce Arrow Lake, and it remains to be seen whether it will follow a similar approach as Lunar Lake.

There won’t be a long wait, as terminal devices equipped with Intel and AMD’s next-generation mobile processors will soon hit the market, and everyone is looking forward to their performance.