Pixel 6: with its Tensor chip, Google does nothing like everyone else

Google unveiled its Pixel 6 yesterday, also an opportunity to tell us a little more about the Tensor, its first in-house SoC. And the least we can say is that the manufacturer wanted to stand out from the competition. By looking at the architecture of the chip, we realize that the choice of cores is far from common.

Credit: Google

Last night, Google unveiled its Pixel 6 after months of teasing. The conference was an opportunity to discover in detail the new flagship and the novelties it brings. Among them, there is one that makes a lot of noise. We are of course talking about the Tensor chip, which marks the firm’s entry into the world of SoC builders. After Apple and Samsung, Google has therefore decided to equip its smartphone with its own processor..

And this choice is not trivial. First, because it allows it to set new standards in the industry by breaking away from Qualcomm, such as supporting security updates for 5 years. But above all, because it allows pushing the limits of performance Pixel smartphones. This he owes to its unique architecture.

The Pixel 6’s Tensor chip is one of a kind

To understand what makes the Tensor an extraordinary SoC, you have to look at the competition. Usually high-end chips like the Snapdragon 888 or the Exynos 2100 have one high power core, three medium power cores and four low power cores. In the case of the processors mentioned, these are respectively the Cortex X1, Cortex A78 and Cortex A55. However, Google has decided to integrate two Cortex X1 at 2.8 GHZ, two Cortex A76 (less powerful than the A78) and four Cortex A55.

Read Also:  Apple is considering a partnership with Google to bring AI to the iPhone

The choice to use two Cortex X1s may seem obvious, since the latter makes the Pixel 6 more efficient during resource-intensive tasks. Indeed, the comparisons with the Pixel 5 are clear: the CPU is 80% faster while the GPU is entitled to a 370% jump in power. But the presence of two Cortex A76s is strange. Here’s how Phil Carmack, vice president of Google Silicon, explains it:

“We’ve focused a lot of our design effort on how the workload is distributed, how energy is distributed across the chip, and how processors come into play at different times. When a heavy workload comes in, Android tends to hit it hard, and that’s how we get responsiveness. “

On the same subject: Getting started Pixel 6 and 6 Pro – the design of the first phones running Android 12 will be debated

He thus underlines the crucial importance of the two Cortex X1 which, when “The processor has a lighter load but which remains significant”, will start to obtain more satisfactory performance. Therefore, Google wants to transfer the average tasks to its most powerful cores, so that when“A workload that you would normally have done with two A76s, at most, is hardly felt with two X1s”.

Source: ArsTechnica




Recent Articles

Related News

Leave A Reply

Please enter your comment!
Please enter your name here