Home / Component / CPU / Intel Lunar Lake to offer almost 50% more multi-thread performance than Meteor Lake

Intel Lunar Lake to offer almost 50% more multi-thread performance than Meteor Lake

It seems Intel Lunar Lake won't need hyper-threading to outperform its predecessors. According to a new report, the upcoming client platform is expected to deliver a significant 50% performance increase over Meteor Lake-U, at least based on the reported Cinebench R23 and Geekbench 5.4.5 multi-core results.

Based on previous reports, the Lunar Lake-MX platform is expected to feature an octa-core CPU configuration with four P-Cores based on the Lion Cove architecture and four E-Cores based on Skymont. Arrow Lake will use the same microarchitecture as Lunar Lake and is planned to debut alongside it, but the former will focus on mid- and high-end devices. While Intel has not yet released all the details of Lunar Lake, they have clarified that the architecture is aimed at thin and low-power laptops. Leaked presentations from last year positioned Lunar Lake as the successor to Meteor Lake-U.

As per Bionic_Squash's tweet, compared to Meteor Lake SKUs (15W) with up to 12 cores (2P+8E+2LP), Lunar Lake (17W) should offer a 1.5x multithreaded boost despite the absence of hyperthreading. It's worth noting that Lunar Lake SKUs are expected to be released with up to 30W TDPs and as low as 8W TDPs.

In addition to improved CPU performance, Lunar Lake is expected to feature a faster NPU with a threefold speed boost over its predecessor. Moreover, it will also have the new Xe2-LPG graphics architecture.

Discuss on our Facebook page, HERE.

KitGuru says: Would you be impressed if Lunar Lake could deliver these gains with a small TDP bump? Considering Arrow Lake will feature cores similar to those on Lunar Lake, do you think we will see a similar performance bump in the mid- and high-end segments?

Become a Patron!

Check Also

AMD Krackan Point APU appears in the Geekbench database

AMD's next-gen Ryzen AI 300 series APUs are on the horizon, and a fresh Geekbench …