Late last night Intel joined the Copilot + PC conversation with the announcement of its Lunar Lake chip that brings an integrated CPU, GPU, and NPU on a single System-on-a-chip (SoC) that could offer up to four times the AI performance than its previous Meteor Lake processor.
During its time on stage at Computex in Taipei, Taiwan, Intel unveiled its second generation of disaggregated SoC architecture called Lunar Lake. Lunar Lake will supplant Meteor Lake PCs in the Core Ultra mobile processor segment for Intel starting in Q3 2024, just in time for those holiday laptop purchases.
As for the details of Luna Lake as a processor, Intel will use a similar efficiency core (E-core) paired with a new P-core codenamed Lion Cove. Like the way Meteor Lake processors dynamically allocate task between the E and P cores, Lunar Lake will also lean heavy on Intel’s Thread Director technology and its new advance low-power island alongside some Microsoft magic with the re-architected Windows 11 OS scheduler to position the chip for best performance structures based on real-time workloads.
Intel is also introducing its fourth generation NPU architecture that comes with support of up to forty-eight tera-operations per second (TOPS) that the company claims offers up to four times the AI performance delivered by Meteor Lake.
Microsoft has put a lot of emphasis on NPUs capable of forty plus TOPS being Copilot + PC approved NPUs, as it heralded Qualcomm’s Snapdragon X Elite chip’s ability to reach a 45 TOPS threshold. Intel’s new Luna Lake will match that of both AMD and Qualcomm’s TOPS performance while also running a new integrated GPU. Intel’s new integrated GPU is being called Battlemage and includes Xe 2 GPU cores alongside Xe Matrix Extension (XMX) arrays intended to tackle even more AI tasks with its own dedicated 67 TOPS for AI content chores.
Intel introduced its Xeon 6 processors as well as its Intel Guadi 2 and 3 AI accelerator kits during its Computex presentation.
- Launches Intel® Xeon® 6 processors with Efficient-cores (E-cores), delivering performance and power efficiency for high-density, scale-out workloads in the data center. Enables 3:1 rack consolidation, rack-level performance gains of up to 4.2x and performance per watt gains of up to 2.6x1.
- Announces pricing for Intel® Gaudi® 2 and Intel® Gaudi® 3 AI accelerator kits, delivering high performance with up to one-third lower cost compared to competitive platforms2. The combination of Xeon processors with Gaudi AI accelerators in a system offers a powerful solution for making AI faster, cheaper and more accessible.
Despite its seemingly late entry into the computer purchasing cycle with a Q3 2024 launch date for Lunar Lake powered PCs, Intel seems bullish on the AI PC market as whole and expects that by 2028, PCs powered by chips like its own Lunar Lake will make up 80 percent of the market.
Quickly building on these unmatched advantages, the company today revealed the architectural details of Lunar Lake – the flagship processor for the next generation of AI PCs. With a massive leap in graphics and AI processing power, and a focus on power-efficient compute performance for the thin-and-light segment, Lunar Lake will deliver up to 40% lower SoC power and more than 3 times the AI compute. It’s expected to ship in the third quarter of 2024, in time for the holiday buying season.
Intel Newsroom
