Efficiency and Power Unleashed in Intel’s Latest Lunar Lake Microchip Innovations

We’re more than halfway through 2024 already, and that means we’re getting to know the hardware that will power our devices through the end of the year and into 2025. AMD has just showed off its Zen 5 chips, with both desktop and laptop releases. Intel wasn’t going to stay behind, though, and the company has just showed off its new Lunar Lake architecture set to power its upcoming mobile CPUs.

Intel has just introduced its Lunar Lake architecture at Computex 2024, promising lots of changes compared to its previous architecture. It will be featured in Intel’s upcoming mobile processors—while the company is yet to announce its range of chips, they will most certainly be branded something along the lines of Intel Core and Intel Core Ultra Series 2 (harking back to how Framework recently referred to the first batch of Core Ultra chips as “Series 1”). According to Intel, the new architecture will feature redesigned P-cores (performance) and E-cores (efficiency) that will deliver notable performance improvements while keeping processor power up to 40% lower compared to the previous generation.

Other improvements include a fourth-generation NPU to tackle local AI tasks (something that will become increasingly more important as the concept of Copilot+ PCs grows even bigger), which provides an improvement of up to four times in AI processing capabilities. It also has a new integrated GPU design codenamed Battlemage, which features new Xe2 cores for graphics and a new “Xe Matrix Extension” (XMX) component that’s also for AI processing tasks. The Xe2 cores improve gaming performance by up to 1.5x, while the new XMX component seems to act as an AI co-processor of sorts that will complement the NPU inside the processor.

Lunar Lake is set to be featured on more than 80 different laptops from 20 different manufacturers, but we’ll probably know those once the actual chips are announced. We’ll probably get these chips by the end of the year, as it’s usually the case, although Intel isn’t really giving us a precise timeframe here. By then, we will also be able to judge how big of a jump this new architecture actually is.

Source: Intel

Also read: