Google Tensor SoC – Everything You Need To Know

Google Pixel 5 is near, and the rumours of Google using their own custom chipset in the new Pixel devices is finally confirmed. The custom chipset codenamed “Whitechapel” has been finally revealed to be the Google Tensor and it comes with integrated AI.

Google Tensor

The chip is called Tensor, and it’s the first system-on-chip (SoC) designed by Google. The company has been at this for about five years. Software expertise is something Google has come to be known for. The way Google changed photography by using AI with its Night mode and shocked the world. With a custom SoC, it will only become better. We can only imagine how well the integration could work out for Google as we have seen how well it works for Apple.

Let’s put all the hypothetic stuff to the side and focus on the actual features and specs of the Tensor SoC.

Google Tensor SoC

Infused & Better AI

Though Google hasn’t shared the full details about Tensor yet, leaks did explain that the SoC is an ARM chip designed around a Tensor Processing Unit. The mobile chip was co-designed with Google’s AI researchers and is strongly based on Artificial Intelligence & Machine Learning.

When Google Duplex was announced everyone was shocked at how good the AI was, it sounded too close to real humans and could learn to be better over time. This is now very easy for Google to implement through Tensor.

Hardware Improvements

Google has also redesigned the memory architecture to make it easier to access RAM, this too shall be around the whole AI and the phone will feel more natural over time. This is not the only new thing in Tensor. Google has also redesigned the image signal processor or ISP. Specifically, there are a few points in the ISP where they actually insert machine learning, which is new. There are also a few places where they’ve directly encoded their image processing algorithms into the hardware.

Google Tensor SoC

Those pieces of the SoC are what impact your day-to-day experience the most — how fast the phone feels, how long it lasts on battery, how well it connects to a cellular network, and so on. After that, there are generally some co-processors off to the side that handles discrete tasks like image processing or security.

Upgraded Camera

As we all know most of the camera improvements in Pixel phones come from software rather than hardware. And now as Google will have complete control over the whole CPU things will work more flawlessly. New features like the Pixel 6 can concurrently capture images from two sensors, with the main one recording at normal exposure and the wide-angle running at a much faster shutter speed.

The processor will run many tasks at once to recognize what’s going on in the scenery. It will intelligently set the filters, hue, saturation to fit the scene better, similarly, if you’re trying to capture a hyperactive puppy or toddler, you’ll be less likely to get a blurry shot.

Better Speech, Text and Voice

Don’t have much deeper information regarding this, but in short, the Speech, Text, Voice recognition will surely improve with better understanding and better suggestions for typing and replying, or even with Google Assistant.

The Look & Feel of the UI

Google Tensor

As you read earlier AI plays a huge role in the Google Tensor SoC, so naturally, the whole UI would feel a lot natural. Everything from the animations to the flow of work will be adapted according to your preference. But the most important change comes with Google YOU. “You” will automatically change the colours and contrast of your home screen and UI in accordance with the colour of your wallpaper or the choice of your widgets etc, in short, you can change the whole look of your phone with a single tap.

That is all we know regarding the Google Tensor SoC. All of it appears really good right now, we can only hope this doesn’t change like with Samsung & Exynos. So, what do you think about the Tensor SoC? Share your thoughts in the comments below.

Read Also, Google Pixel 6 and Pixel 6 Pro Announced – Sorta!

Recent Articles


Related Stories