Search

Intel claims that the Lunar Lake laptops will be AI powerhouses capable of outperforming the Snapdragon X Elite CPU.

Analysis: Next-gen AI performance will be a close-fought battle

Intel claims that the upcoming Lunar Lake laptops are set to be AI powerhouses that will outperform competitors such as the Snapdragon X Elite CPU. The Lunar Lake CPUs are not only focused on power-efficiency but also on delivering superior AI performance for the best ultrabooks in the market.

At the recent Vision 2024 event, Intel revealed that the Lunar Lake chips will have exceptional AI capabilities, specifically boasting 45 TOPS for AI processing, tripling the capabilities of the current-gen Meteor Lake NPU that offers 10 TOPS. When taking into account the AI power of both the CPU and integrated GPU, the Lunar Lake is expected to deliver over 100 TOPS, crossing the threshold for being classified as an ‘AI PC’ based on Microsoft’s standards.

Despite the lack of detailed breakdown on how the additional 55 TOPS from the CPU and GPU are allocated, it is speculated that the graphics solution, likely Battlemage, will be the primary contributor to this processing power advancement.

Comparing the AI performance predictions of the Lunar Lake with other upcoming notebook silicon, Intel seems to have an edge over its competitors. While the Snapdragon X Elite is expected to offer similar NPU performance with 45 TOPS, the overall AI performance, including the CPU and GPU, is significantly lower at 75 TOPS compared to Lunar Lake’s 100+ TOPS.

AMD’s next-gen Strix Point chips are also anticipated to compete in this space, with an expected NPU performance of around 48 TOPS, slightly edging out Intel and Qualcomm. However, details on the additional CPU and GPU contributions to overall AI performance from AMD are still awaited. The race for AI superiority promises to be a close one, with general performance under Windows 11 also playing a significant role.

You might also like

Share it

🤞 Don’t miss these tips!

Solverwp- WordPress Theme and Plugin