Why The Pixel 6's Tensor Chip Is Actually A Big Deal And Why Not?

Why The Pixel 6’s Tensor Chip Is Actually A Big Deal And Why Not?

It’s finally official. Google’s Pixel 6 will feature the company’s first custom SoC. While the company has already tackled the security add-ons Pixel Visual Core and Titan M with custom hardware, this is the first time Google has figured out the entire internal workings of the chip itself. (Although the company has licensed many of the building blocks for the SoC.) However, the Tensor Processing Unit (TPU) is completely in-house, and Google will be the heart of the Tensor SoC.

As expected, the Google Tensor processor is focused on enhanced imaging and machine learning (ML) capabilities rather than breakthrough performance. However, this gives rise to a lot of tension and some reservations.

Why Is The Google Tensor SoC So Important?

First and foremost, Tensor is a custom piece of silicon developed by Google to be efficient at the things the company wants to prioritize most. That means it should offer faster, more powerful image processing, language processing, and other machine learning-based features. At least it’s faster than the previous-generation Pixel 5.

With a powerful internal TPU at the core of the chip, Google talks about real-time on-device voice translation for subtitles, text-to-speech without an internet connection, dual keyboard and voice input methods, and great camera capabilities. We can imagine that Google Lens and other machine learning (ML) technologies will improve as well. While these are mostly improvements over what Google already did with existing hardware, hopefully, we’ll see some new features.

See also  Google has disabled Assistant reminders for G Suite users

Google Tensor is going to take what we loved about the Pixel 5 and make it even better.

AI and ML are at the heart of what Google does, and it probably does it better than anyone else – which is why it’s the core focus of the Google chip. As we have found in many newer SoC versions, performance is no longer the most important aspect of mobile SoCs. Heterogeneous compute and workload efficiencies are just as important, if not more important, to enable powerful new software features and product differentiation.

By leaving the Qualcomm ecosystem and choosing its own components, Google gains greater control over how and where valuable silicon space is spent on realizing its smartphone vision. Qualcomm has to do justice to a multitude of partner visions, while Google clearly has something much more specific in mind. It’s hard to doubt that Google thinks the Pixel 6 experience will benefit more from improved AI than Facebook opens 5% faster than last year. Much like Apple’s work on custom silicon, Google is turning to custom hardware to create custom experiences.

By moving to a custom or co-developed processor, Google may be able to deliver updates even faster and longer than ever before. Partners rely on Qualcomm’s support roadmap to provide long-term updates. Samsung offers three years of operating system and four years of security updates through Qualcomm, and Google promises the same for the Pixel 5 and earlier. It will be interesting to see if Google moves forward as it gets closer to the chip design process.

See also  Google Pixel XL Review, Specifications & Price

Why might it not be a big deal?

Anyone hoping for a performance that wouldn’t go against the grain will be disappointed. Google has not shared any benchmarks or details about the insides of the CPU, GPU, or other components. Without an architectural designer, however, Google certainly licenses ready-made arm parts like the Cortex-A78. We’re still in the dark about what 5G features the phone will have. Google won’t even say who made its chipset, despite rumors pointing to Samsung. Google hardware chief Rick Osterloh said Tensor will be “very competitive” in terms of CPU and GPU performance. Make it what you want.

Even Google isn’t doing anything groundbreaking with its image and machine learning pipeline. After all, Google’s development cycle does not stand alone. State-of-the-art hardware has evolved significantly from Google’s last premium mobile phone, the Pixel 4 series.

State-of-the-art hardware has evolved significantly compared to Google’s last premium mobile phone.

So far, Google’s demos show the application of its advanced image processing features to multi-camera and video scenarios. This is possible because Google’s machine learning chops are now integrated into the image processing pipeline (ISP) instead of sitting somewhere further away.

This isn’t a new idea, though, even for 2020 smartphones, let alone late 2021. In fact, the Qualcomm Snapdragon 855 powering the 2019 Google Pixel 4 has introduced computer vision elements into the ISP chain. Since then, the Snapdragon 865 and 888 have enhanced these capabilities, allowing partners to use data from multiple cameras simultaneously and apply effects such as HDR and real-time bokeh to 4K video at 60 frames per second. Google isn’t the first to come up with these ideas, but that doesn’t mean they can’t do them better.

See also  Google Tensor Chip Specs - Everything You Need to Know

Other SoC manufacturers also have their own low-power sensor chips for features such as always-on speech recognition, ambient display, and other sensor functions. Security enclaves like Titan M aren’t new either. In fact, they are essential in today’s biometric-obsessed devices. Similar features can be found in mobile SoCs from Apple, Huawei, Qualcomm, and Samsung. However, the exact functions differ.

Google’s Tensor SoC: A Departure From The Status Quo?

Google boss Sundar Pichai noted that the tensor chip took four years to manufacture, which is an interesting time frame. Google started this project when mobile AI and ML capabilities were relatively new. The company has always been a leader in the ML market and often seemed frustrated with the limitations of partner silicon, as demonstrated in the Pixel Visual Core and Neural Core experiments.

The Tensor SoC is Google with its own vision, not just for silicon for machine learning, but also for how hardware design affects product differentiation and software capabilities. It will be fascinating to see if all this comes together successfully in a Pixel 6 smartphone capable of some impressive industry firsts.

Qualcomm and others, however, haven’t sat on their hands for four years. Machine learning, computer imaging, and heterogeneous computing functions are at the heart of all major mobile SoC players, and not just in their premium products. It remains to be seen whether Google will simply reinvent the wheel for this or whether its TPU technology and the Tensor SoC will actually lead the way.


Similar Posts