Google Tensor G3 will bring amazing features to Pixel 8

We've discovered bits and pieces of information about the Tensor G3, dubbed Zuma, the next-generation Google-branded SoC. Kamila Wojciechowska, a well-known leaker, has already revealed nearly everything we need to…

By Nivedita Bangari

June 4, 2023

2 min read

Comments

We’ve discovered bits and pieces of information about the Tensor G3, dubbed Zuma, the next-generation Google-branded SoC. Kamila Wojciechowska, a well-known leaker, has already revealed nearly everything we need to know about the chipset that will power the Pixel 8 and 8 Pro later this year. The new device is said to have nine CPU cores (4+4+1), including four Cortex-A510 cores and four Cortex-A715, as well as a single high-performance Cortex-X3 core clocked at 3.0 GHz.

While the inclusion of the second-generation Armv9 CPU will undoubtedly improve the Tensor G3’s performance, it may still lag behind the Armv9.2 version revealed recently.

Another significant difference with the Tensor G3 is the inclusion of 64-bit programmes out of the box, whereas Tensor G2 phones such as the Pixel 7 retained 32-bit libraries.

The Tensor G3 will apparently use the Arm Mali-G715 (Immortalis) GPU, which is an improvement above the seven-core Mali-G710 graphics engine. Although Wojciechowska’s source was unable to confirm the amount of cores available on this new GPU, she speculates it could be set up in a 10-core configuration, enabling ray-tracing capabilities, which are now present on flagships such as the Samsung Galaxy S23 Ultra.

The AV1 encoder will be one of the more eye-catching features of the Tensor G3, making the Pixel 8 one of the first to include it out of the box. This is because to the current “BigWave” block, which is an improved version of the previous Tensor chip’s “BigOcean” hardware video decoder block.

The Tensor G3 also includes a brand-new TPU (Tensor Processing Unit), codenamed “Rio.” It is said to have a clock speed of 1.1 GHz, but little more is known about it. However, we anticipate significant advancements in this area, particularly in Natural Language Processing (NLP) and AI/ML-related jobs.

Also Read:

Source