A report from research firm Moor Insights and Strategy (via AndroidAuthority) says that the next Cortex-X CPU core could be the most powerful CPU core for smartphones to date. Considering that the current Cortex-X CPU core is the Cortex-X4, the thinking here is that the next iteration will be known as the Cortex-X5 CPU core and it is codenamed “Blackhawk.”
“Blackhawk,” according to Moor, will be found on phones shipping at the end of this year. These devices could be on shelves next year during CES (January 7th-January 10th, 2025) or MWC (the dates have yet to be announced). ARM CEO Rene Haas says that the company’s strategy is to “eliminate the performance gap between ARM-designed processors and custom ARM implementations.”
ARM has a new Cortex-X CPU core coming to smartphones in 2025
In other words, ARM wants chips like the upcoming MediaTek Dimensity 9400 and Samsung Exynos 2500, both of which will use the new ARM Cortex-X CPU core, to move closer performance-wise to Apple’s A18 Pro and Bionic SoCs which will feature a custom implementation of ARM’s technology. If you were thinking that the Snapdragon 8 Gen 4 would feature ARM’s new CPU core, forget about it. That application processor will debut Qualcomm’s own Oryon CPU cores.
Based on Geekbench 6, ARM says that the Blackhawk core delivers the “largest year-over-year IPC (instructions per cycle/clock) performance increase in 5 years.” Moor’s report also says that the Blackhawk will offer “great” LLM (Large Language Model) performance which means the new Cortex-X CPU core will deliver improved generative AI capabilities. Speaking of AI, ARM says that its Cortex CPU is the #1 target for developers as opposed to using the NPU or GPU.
The report states, “The NPU and GPU can be an efficient way to run AI, but a CPU is the easiest and most pervasive way, which is why developers target it. A higher-performing CPU obviously helps here, but as the world moves increasingly to smaller language models, ARM’s platform with higher-performing CPU and GPU combined with its tightly integrated ML (Machine Learning) libraries and frameworks will likely result in a more efficient experience on devices.”
