Google Tensor First Chipset for Pixel 6 & 6 Pro?

Yes, the Google Tensor chip is the first custom system-on-a-chip (SoC) designed by Google for its Pixel smartphones, debuting in the Pixel 6 and Pixel 6 Pro in October 2021. It replaced the Qualcomm Snapdragon processors used in all previous Pixel models, marking a major shift in Google’s hardware strategy.

Background: Why Google Made Its Own Chip

For years, Google’s Pixel phones relied on Qualcomm’s Snapdragon processors. While capable, these chips weren’t tailored to Google’s specific needs—especially its heavy focus on AI, machine learning, and computational photography. By designing its own chip, Google could integrate its software algorithms directly into the hardware, enabling features that weren’t possible with off-the-shelf processors. The Tensor chip’s name is a nod to TensorFlow, Google’s machine learning platform, signaling its AI-first design philosophy.

How the Tensor Chip Works

The Tensor chip is a system-on-a-chip that combines several components into one package:

• CPU: An octa-core processor with two high-performance Cortex-X1 cores, two mid-range Cortex-A76 cores, and four power-efficient Cortex-A55 cores.
• GPU: A 20-core Mali-G78 graphics processor for gaming and visual tasks.
• TPU: A dedicated Tensor Processing Unit for machine learning tasks, handling things like live translation, voice recognition, and photo processing.
• ISP: An image signal processor that works with the TPU to enhance photos and videos, enabling features like Magic Eraser and Face Unblur.

This integration allows the chip to run complex AI models on-device, reducing reliance on cloud servers and improving privacy and speed.

Key Features Enabled by Tensor

The Tensor chip powers several signature Pixel 6 features that rely on on-device machine learning:

• Magic Eraser: Removes unwanted objects or people from photos with a single tap.
• Face Unblur: Automatically sharpens faces in motion shots by combining multiple frames.
• Live Translate: Provides real-time translation of text, speech, and conversations in multiple languages.
• Assistant Voice Typing: Offers fast, accurate voice dictation with on-device processing.

These features demonstrate how the chip’s AI capabilities directly improve the user experience, something a generic Snapdragon chip couldn’t deliver as effectively.

How Tensor Compares to Other Chips

In benchmark tests, the Tensor chip’s raw CPU and GPU performance was competitive with flagship chips from 2021, like the Qualcomm Snapdragon 888 and Samsung Exynos 2100, though it often trailed in sustained performance and efficiency. However, Tensor’s real strength lies in its AI and ML capabilities, where it outperformed competitors in tasks like image recognition and natural language processing. For everyday use, the Pixel 6 felt smooth and responsive, but heavy gaming could cause more heat and battery drain than some rivals.

Common Problems and Misconceptions

One common misconception is that Google Tensor is a fully in-house design. In reality, Google designed the architecture and worked with Samsung’s foundry to manufacture it, using some of Samsung’s Exynos IP blocks. Another issue reported by users was the modem—the Tensor chip used Samsung’s Exynos 5123 modem, which had connectivity and battery drain issues for some. Additionally, the chip’s thermal management was not as robust as some competitors, leading to throttling during intense tasks. These are important trade-offs to consider when evaluating the Pixel 6’s performance.

How to Choose: Pixel 6 vs. Pixel 6 Pro

Both phones use the same Tensor chip, so performance is identical. The choice comes down to other factors:

• Display: The Pixel 6 has a 6.4-inch 90Hz screen; the Pro has a 6.7-inch 120Hz LTPO display.
• Camera: The Pro adds a 4x telephoto lens and a periscope zoom, while the base model has only wide and ultrawide lenses.
• RAM and storage: The Pro offers up to 12GB RAM and 512GB storage, versus 8GB and 256GB on the base model.
• Battery: The Pro has a larger 5,003mAh battery vs. 4,614mAh, but the base model often lasts longer due to the lower refresh rate.

If you want the best camera and display, choose the Pro. If you prefer a smaller, more affordable phone with the same core performance, the Pixel 6 is a great choice.

Cost and Value

At launch, the Pixel 6 started at $599 and the Pixel 6 Pro at $899, significantly undercutting competitors like the Galaxy S21 and iPhone 13. The Tensor chip allowed Google to offer premium features at a lower price because it didn’t have to pay Qualcomm’s licensing fees. This value proposition made the Pixel 6 series a popular choice for those wanting flagship performance without the flagship price.

Safety and Longevity

Google committed to five years of security updates for the Pixel 6 series, a major advantage over many Android phones. The Tensor chip’s on-device processing also enhances privacy by keeping sensitive data like voice recordings and photo analysis on the phone rather than in the cloud. However, users should be aware that the chip’s modem and thermal issues could affect long-term reliability, and battery degradation is normal over time.

Frequently Asked Questions

Is Google Tensor faster than Snapdragon?

In raw CPU and GPU benchmarks, the Tensor chip was comparable to the Snapdragon 888 but often slower in sustained performance. However, in AI and machine learning tasks, Tensor outperformed Snapdragon, making it faster for features like live translation and photo processing.

Does the Tensor chip support 5G?

Yes, the Tensor chip includes an integrated 5G modem, supporting both sub-6GHz and mmWave bands (though mmWave was limited to certain models and carriers). Some users reported connectivity issues, particularly with the modem’s power efficiency.

Can I use a Pixel 6 without a Google account?

Yes, you can set up a Pixel 6 without a Google account, but you’ll miss out on many Tensor-powered features like Google Assistant, Live Translate, and backup services. The phone is designed to work best with a Google account.

Why is it called Tensor?

The name comes from TensorFlow, Google’s machine learning framework. Tensors are mathematical objects used in machine learning, and the name reflects the chip’s focus on AI and ML processing.

Is the Tensor chip used in newer Pixel phones?

Yes, Google has continued to develop Tensor chips for subsequent Pixel models, including the Tensor G2 in the Pixel 7 series and Tensor G3 in the Pixel 8 series. Each iteration improves performance and AI capabilities.

Related

• Can’t Hear Anyone in Valorant? Fixes
• AI Sea Ice Forecasts: Evidence & Limits
• Gigabyte X870 Aorus DDR5 Specs: What They Promise