Recent benchmark tests for the Pixel 10 Pro Fold have raised some eyebrows regarding the performance of the Tensor G5 application processor (AP). Set to be revealed on August 20, the Pixel 10 series—including the Pixel 10, Pixel 10 Pro, Pixel 10 Pro XL, and Pixel 10 Pro Fold—will feature the Tensor G5 AP, marking the first time Google has designed its chip from scratch. This new AP will be manufactured by TSMC rather than Samsung Foundry, potentially allowing for enhanced Pixel-specific features.
During the Geekbench benchmark tests, concerning details about the Tensor G5 AP emerged. The tests confirm that the Tensor G5 has an octa-core configuration consisting of a Cortex-X4 Prime core at 3.78 GHz, five Cortex-A725 cores at 3.05 GHz, and two Cortex-A520 cores at 2.25 GHz. Its GPU is the PowerVR D-Series DXT-48-1536.
Calculating the Instructions Per Cycle (IPC) score by dividing the single-core Geekbench score by the CPU’s clock speed yields worrisome results for the Tensor G5 when compared to other recent processors. The IPC scores show that the Tensor G5 is less efficient than its predecessor, the Tensor G4, and also lags behind competitors like the Snapdragon 8 Gen 3 and Dimensity 9300. This disparity indicates that the Tensor G5’s efficiency in executing instructions is concerningly low.
Additionally, comparisons highlight that although the Snapdragon 8 Gen 3 shares a similar core configuration, it achieves a significantly higher multi-core score. There’s speculation that Google may have throttled the AP design for this iteration, given its status as a foldable device. Some believe the benchmark scores could improve as the launch date approaches.
While there remains time for optimization, the results thus far present challenges for the Tensor G5 and could impact its reception upon release.