,,Tensor is our first custom-built SoC specifically for Pixel phones, and it will power the Pixel 6 and Pixel 6 Pro later this fall"
Google in the Pixel series presented solutions such as HDR +, Night Sight (together with AI), speech recognition capable of decoding, recognizing, and searching for e.g. music songs. Unfortunately, innovative solutions require many calculations of algorithms, and the last smartphones are not fast enough. A new solution from Google is to help in this, namely, the giant is working on its own processor that will be used only in Pixel smartphones, something like A14 Bionic processors from Apple.
Google, taking advantage of the possibility of equipping its Pixel series phones with innovative solutions, encountered a serious problem, which is the computational limitation of the processors used by it.
Is the Tensor processor a modification of the Exynos?
Google says it has been developing its SoC, Tensor, for four years. Recently, the company announced that it is resigning from Qualcomm Snapdragon processors in its upcoming Pixel 6 smartphones. The latest reports, however, indicate that Google Tensor is a chip that we already know, only under a different name.
Leaks say Google had some help from Samsung in creating its SoC. Samsung’s commitment to Tensor could be much more than just helping, according to GalaxyClub. Apparently, the Tensor is also called “Exynos 9855” on an operative basis. It is to be structurally close to the system called Exynos 9840 … which is the internal name of the Exynos 2100 system, i.e. processors known from the Galaxy S21.
If reports of a very close affinity between Tensor and Samsung systems prove to be true, we should expect the Pixel 6 and Pixel 6 Pro to be equipped with an efficient ARM Cortex-X1 core, three Cortex-A78 cores and four energy-efficient Cortex-A55 cores. The frequencies are unknown, but you can assume they will be very close to what we saw in the Exynos 2100.
What does TENSOR mean at all?
Tensor Processing Units (TPUs) are Google’s custom-developed application-specific integrated circuits (ASICs) used to accelerate machine learning workloads. TPUs are designed from the ground up with the benefit of Google’s deep experience and leadership in machine learning.
In mathematics, a TENSOR is an algebraic object that describes a multilinear relationship between sets of algebraic objects related to a vector space.
