Menu Close

Google Tensor 2 chip should bring improved modem, better battery life to the Pixel 7 line

Google made a point of letting the world know that the Pixel 6 series was powered by the first Google-designed chipset, the Tensor. As Google put it, this would allow the company to design new features for the Pixel 6 and Pixel 6 Pro that would be made possible thanks to the Tensor. These are features that wouldn’t be possible had Google resorted to its previous habit of buying Snapdragon chips off the shelf.
The problem with the first Tensor chip was the Samsung-produced Shannon A5123 5G modem which led to several connectivity problems with the Pixel 6 and Pixel 6 Pro. An Apple executive has called Qualcomm’s Snapdragon modems the best in the business while testifying under oath against Qualcomm. The connectivity issues, which prevented Pixel 6 series users from making or taking phone calls, led some to switch to the Samsung Galaxy S22 series which used the Qualcomm Snapdragon X65 5G modem.
According to SamMobile, the new Google Tensor 2 chip, produced using Samsung Foundry’s 4nm LPE (compared to last year’s 5nm chip), features Samsung’s Exynos 5300 5G modem that is faster and more energy-efficient compared to the Shannon A5123 which drained the batteries on the Pixel 6 and Pixel 6 Pro.

The new Tensor 2 chip has two ARM Cortex-X1 CPU cores with a clock speed of 2.85GHz, two ARM Cortex-A78 CPU cores clocked at 2.35GHz, and four ARM Cortex-A55 CPU cores running at a clock speed of 1.8GHz. The chip comes with ARM’s Mali-G710 GPU with seven cores, Google’s Titan M2 security chip, and the second-generation EdgeTPU for AI demands.

The chip also works with LPDDR5 RAM and thanks to an improved image processor, it can support 4K video from all cameras at 60 frames per second. The chip also can support cameras up to 108MP (with zero shutter lag). The chip also allows the Pixel 7 series to unblur photos, even those that are years old. 

Those upgrading from the Pixel 6 line to a Pixel 7 series phone should notice some differences thanks to the improvements made in the Google Tensor 2.

Source link

Leave a Reply

Your email address will not be published.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