Google is about to debut its first in-house smartphone chipset, the Tensor SoC, in its upcoming Pixel 6 and Pixel 6 Professional telephones. And if the latest report from XDA is correct, the Tensor’s rumored CPU setup might be very, very bizarre — even by Google requirements.

To this point, Google has largely been hyping up the Tensor SoC’s AI efficiency; however it hasn’t revealed any data on the essential CPU and GPU specs of the chip. Google’s Rick Osterloh would solely inform The Verge that “the standard stuff people look at will be very competitive and the AI stuff will be totally differentiated.”

Some points of the Tensor’s parts have already come to mild. An earlier XDA report notes that the Pixel 6 will probably use an off-the-shelf Arm Mali-A78 GPU design (which Samsung makes use of on its flagship Exynos 2100), whereas Reuters reports that Google can be sourcing its 5G modem from Samsung.

However the CPU nonetheless remained a thriller, till as we speak, when XDA printed a report primarily based on each a Geekbench rating and a supply who claims to have an precise Pixel 6 Professional. The report claims that the CPU setup on the Tensor will encompass two Cortex-X1 efficiency cores clocked at 2.802GHz, two Cortex-A76 efficiency cores clocked at 2.253GHz, and 4 Cortex-A55 effectivity cores.

Google’s new Pixel 6 phones, regular and Pro

Picture: Google

When you’ve been maintaining with main flagship smartphone chips, that’s a really bizarre listing to see, one which mixes collectively highly effective new cores with weaker outdated ones. Let’s take a step again to clarify why:

After we’re taking a look at most smartphone SoCs, there are usually two most important components: efficiency CPU cores and effectivity CPU cores. Arm-based designs have a tendency to combine these collectively in massive.LITTLE configurations, to permit for units that may ramp up efficiency through the use of the extra highly effective “big” cores for intensive issues like gaming, whereas operating much less demanding duties (like checking your e-mail) on the “little” effectivity cores to delay battery life.

A typical Arm-based design would possibly embody 4 efficiency cores (just like the Cortex-A78) and 4 effectivity cores (just like the Cortex-A55). However final yr, Arm added a brand new, much more highly effective efficiency choice for chip makers to make use of: the Cortex-X1.

So, the highest smartphones of 2021 have a tendency to supply a triple-cluster design: the Snapdragon 888 makes use of partially personalized variations of a single Cortex-X1, three Cortex-A78, and 4 Cortex-A55 cores, whereas Samsung’s Exynos 2100 makes use of an analogous configuration. Tensor, alternatively, is alleged to supply two Cortex-X1 cores, two Cortex-A76 cores, and the same old 4 Cortex-55 cores.

Which makes for a really unusual model of a triple cluster design. By together with not one, however two Cortex-X1 efficiency cores, Tensor might theoretically enable it to outclass even the very best chips from Qualcomm and Samsung, on paper, if not for the second half of the rumor, which is that Google can be utilizing two older Cortex-A76 cores… which, merely put, doesn’t make any sense.

As XDA factors out, the Cortex-A76 was launched in 2018 and is a full two generations behind the Cortex-A78 design utilized in 2021’s flagship chips. There’s no instantly logical cause why Google would use the older design, both; the A78 is each quicker and extra environment friendly than its older counterpart, making it an extraordinarily unusual alternative to incorporate as a part of the Tensor CPU cluster, particularly if Google is already going all out with two X1 cores.

There’s the possibility that Google is just obfuscating its CPU design within the Geekbench rating, though the report does observe that it will be unlikely.

For now, although, the thriller of the Pixel 6’s Tensor chip has gotten even weirder. And it’ll probably keep that method, too, till Google reveals extra data when the telephones arrive later this fall.

LEAVE A REPLY

Please enter your comment!
Please enter your name here