When Google initially teased the Pixel 6 at the beginning of August, it focused heavily on its first custom-built System on a Chip (SoC). However, Tensor was unveiled today at Google’s Pixel Fall Event, and it was dubbed the “largest mobile hardware breakthrough in the company’s history.” According to Google, Tensor’s declared purpose is to push the boundaries of what is possible on smartphones. In addition, the business intends to bring “AI advances directly to Pixel” and advance its Ambient Computing vision of always-available technology.
The former stems from Google’s hardware division’s belief that AI-powered intelligent features are the best way to set Pixel apart from the competition. Google mentioned Ambient Computing again at the Pixel Launch Event. The last time this happened in a big way was when the Pixel 4 was released in 2019. Rick Osterloh told The Verge that work on the project began in 2017 after Google realized it couldn’t increase AI models in a piecemeal manner, such as constructing a single co-processor, such as the Pixel Visual/Neural Core. Instead, a chip that is tuned for specific activities is required. The Tensor chip was created to bring Google’s latest AI developments to mobile devices.
Google went into Tensor mode at the Pixel 6 Launch Event, emphasizing the existence of two high-performance ARM Cortex-X1 cores running at 2.8 GHz. They’re joined by two “mid” 2.25 GHz A76 CPU cores, which are based on a 5nm process rather than the 7nm process used in flagship phone chips last year, according to Ars Technica’s Google Silicon interview. The CPU is completed by four tiny A55 cores with high efficiency. As a result, Google can use the dual-X1 technique to provide medium-intensity jobs with extra processing power. For example, the central cores of a regular CPU would do tasks like Google Lens visual analysis, but they would be “maxed out.”
In that case, Google claims that employing two X1 cores would be more efficient, which is precisely what Tensor is designed for. It’s 80 percent faster than the Snapdragon 765G in the Pixel 5. It’s 370 percent faster than the Adreno 620 GPU in the Pixel 5. Meanwhile, the Tensor security core is a CPU-based subsystem that runs critical operations and controls and is segregated from the application processor. Meanwhile, the Tensor security core is a CPU-based subsystem that runs critical operations and controls and is segregated from the application processor.