Google Acknowledges the Importance of Optimizing the GPU in Tensor G5

Google Acknowledges the Importance of Optimizing the GPU in Tensor G5

Google’s latest System on Chip (SoC), the Tensor G5, integrates Imagination’s IMG DXT-48-1536 GPU. However, recent observations over the past weeks reveal persistent overheating and throttling issues, which many attribute to the performance shortcomings of this particular GPU.

In response to these challenges, Google has seemingly acknowledged the situation by committing to enhance “driver quality”through its regular system updates, both monthly and quarterly.

Google’s Commitment to Enhancements

AnTuTu 10 benchmark screen for Google Tensor G5 showing a total score of 1508196.
Google Tensor G5 AnTuTu 10 Benchmark Score
AnTuTu 10 benchmark screen displays MediaTek Dimensity 9500 with a total score of 3533222.
MediaTek Dimensity 9500 AnTuTu 10 Benchmark Score

The performance of Imagination’s IMG DXT-48-1536 GPU within the Tensor G5 is notably underwhelming. For context, its AnTuTu 10 GPU score represents only 40% of the performance achieved by MediaTek’s Dimensity 9500, which features the ARM Mali-G1 Ultra MC12 GPU. This stark comparison underlines the critical performance gap that exists (as reported by NanoReview).

In an update to Android Authority, Google reaffirmed its commitment to optimize GPU performance, noting:

“We are continuing to improve driver quality in our monthly and quarterly system updates. For example, the most recent September and October patch releases included driver improvements. In future releases, we are planning further GPU driver updates.”

The cautious wording in Google’s response is significant. Although Google collaborates closely with Imagination in the development of the IMG DXT-48-1536 GPU, it is important to note that Imagination maintains exclusive control over drivers for the DXT series. This means that while Google can implement certain optimizations—particularly concerning AI tasks and power management—it remains dependent on Imagination for fundamental driver updates and hardware-specific programming.

It is hoped that the forthcoming driver updates will effectively address the persistent throttling issues associated with the Tensor G5 that many Pixel 10 users are currently experiencing.

Source & Images

Leave a Reply

Your email address will not be published. Required fields are marked *