The Tensor G5 doesn't have custom CPU cores like the Snapdragon 8 Elite, but the new GPU is said to offer ray tracing support and virtualization options

Just hours after it was announced that Google would be using TSMC's 3nm “N3E” and “N3P” processes for the Tensor G5 and Tensor G6, a new report talks about the specifications of the chipset that will be part of the Pixel 10 -Next year's innards will be. The disappointing aspect of the latest report is that the Tensor G5 will not have custom CPU cores like those adopted by Qualcomm and Apple, but the GPU will receive ray tracing and virtualization support, features already adopted by competing SoCs.

Google isn't even using ARM's latest Cortex-X925 as part of the Tensor G5's CPU cluster, but will be using faster Cortex-A725 cores

This year, chipsets like the Dimensity 9400 feature ARM's new Cortex-X925, which offers a huge boost in performance and efficiency compared to the Cortex-X4. Unfortunately, Android Authority's Kamila Wojciechowska reports that the Tensor G5 will fall well short of its competitors, as Google will stick with a single Cortex-X4 as part of the CPU cluster. Fortunately, the positive aspect is that the company will now use five Cortex-A725 cores instead of the four Cortex-A720 cores in this year's Tensor G4, as well as two low-power Cortex-A520 cores.

A key reason why Google has the freedom to increase the number of Cortex-A725 cores is the improved manufacturing process of the Tensor G5. 3nm lithography has allowed Qualcomm and MediaTek to integrate a pure performance core configuration into their chipsets, meaning Google could gain some advantages here. The Tensor G5's capabilities cannot be predicted based on this CPU cluster, but we should see a significant multi-core increase thanks to the increased Cortex-A725 cores.

As for the GPU, Google is said to be ditching ARM's designs in favor of Imagination Technologies and letting the new DXT-48-1536 part run at 1.10GHz. What sets this GPU apart from previous generations is that the Tensor G5 does not offer ray tracing and GPU virtualization support. In the past, Google has chosen to forego “gaming” features, but given that competitors support these technologies, the company has little choice but to take the same approach.

When it comes to the Tensor Processing Unit (TPU), the Tensor G5 is said to have made modest improvements over the Tensor G4, with Google internal documents showing it to be 14 percent faster. As far as die size is concerned, the Tensor G5 is said to measure 121 mm², making it larger than the A18 Pro. This should bring some benefits to the chipset, but we will discuss these improvements separately. For now, we feel like Google missed the opportunity to add custom CPU cores and GPU to the Tensor G5, but we can keep our fingers crossed for the Tensor G6.

News source: Android Authority