Earlier this month, Google launched the much-talked and most-anticipated smartphone in the face of the Pixel 7 / Pro. Though there were few improvements, the series came with a couple of selling points, including the company’s in-house-made Tensor G2 chip. As the manufacturer proves, due to the Tensor G2 processor, the Pixel 7 and Pixel 7 Pro bring more useful personalization features for photos, videos, security, and voice recognition.
However, the actual performance test of the chip is not what we have been waiting for. On GeekBench 5, the processor scores 1068 and 3149 in single-core and multi-core tests, respectively. Thus, this is only a slight improvement over the previous generation Tensor chip. Moreover, it has nothing in common with the flagship chips of this year.
In this week’s Made by Google podcast, Google responded to the performance of the Tensor G2 chip, saying that they value the chip’s AI capabilities more than the hardware performance.
Monika Gupta, senior director of product management for Google’s chip team, said they designed the chip to focus on machine learning needs five years from now.
“I’m not making decisions based on where machine learning is today, and I can say that because I work at Google. Same with the software that our software team is doing. I know where the software team wants to take the user experiences five years from now. That’s the benefit of not being a merchant silicon supplier, but an in-house silicon supplier. So those trade-off decisions are very tough, but I think they get a little easier when you’re vertically integrated.”
Google said they are “very satisfied” with the Tensor chip, and it didn’t win the benchmark because of its focus on prioritizing end-user experience.
By the way, Google’s Tensor G2 chip uses Samsung’s 5nm process. The Arm Cortex-X1 ultra-large core frequency in Tensor G2 is 2.85GHz, which is only 50MHz higher than the original Tensor’s 2.80GHz.