Tensor G3 Is Not Fabricated On Samsung’s 4nm LPP+ Process, May Be Inferior To Upcoming Exynos 2400 In Power Efficiency

wccftech.com:

Google did not touch on the specifications of the Tensor G3 during its official announcement, though it spoke incessantly about its computational imaging, video capabilities, and leveraging AI to enhance the final results. Even on the Pixel 8 and Pixel 8 Pro comparison page, there are no additional details provided on the latest SoC, but we did stumble across information that mentions that Samsung’s older 4nm manufacturing process is being used to mass produce the Tensor G3, not the latest one.

Past leaks have already spoken about the CPU configuration of the Tensor G3, with the chipset featuring a 9-core CPU and a Mali-G715 GPU. While the CPU cluster is a decent one, the manufacturing process is Samsung’s older 4nm LPP (Low Power Plus) instead of the 4nm LPP+, according to Notebookcheck. The newer manufacturing process may be dedicated to the Exynos 2400, and next year’s Tensor G4 is also fabricated on it, though an earlier report states that the latter will not provide a meaningful upgrade over the Tensor G3.

The Tensor G3 is mass produced on an inferior node as it no doubt helps Google save immensely on chipset shipments. Where the Snapdragon 8 Gen 2 reportedly costs $160 to Qualcomm’s smartphone partners and the Snapdragon 8 Gen 3 is said to be costlier than its predecessor, Google can instead invest those funds by focusing more on the features side of the Tensor G3, giving it capabilities in the software department. Where it falters in raw CPU performance, as evidenced in an earlier single-core and multi-core Geekbench 6 leak, Google has managed to increase Pixel 8 and Pixel 8 Pro capabilities in both imaging and video.

Unfortunately, the drawback of using Samsung’s inferior 4nm LPP process is the lack of power