Google's Tensor G4 chip: The powerhouse behind the next-gen Pixel smartphones

Tensor G4 Chip Boosts Pixel: A Game-Changing Upgrade

Since its first introduction in 2021, Google Tensor has been a game-changer for the Pixel series, emphasizing specialized tasks to enhance user experience. The latest in the lineup, Google’s Tensor G4 chip, not only continues this innovative trajectory but takes it further by utilizing Samsung’s Exynos 2400 as a foundation, stepping away from Samsung’s conventional CPU design to feature a custom CPU blueprint.

Moreover, the manufacturing prowess of Samsung plays a critical role in bringing the udintogel Tensor G4 chip to life, cementing Google’s decision to continue with Samsung Foundry for its production. However, there’s buzz about a potential shift to TSMC for the production of the next iteration, the Tensor G5, indicating Google’s relentless pursuit to refine and enhance its chipset technologies.

Innovative Tensor G4 chip set to revolutionize Google Pixel user experience

Google’s Partnership with Samsung

  • Foundry and Fabrication Process
    • Google is set to continue its collaboration with Samsung for the production of the Tensor G4, utilizing the advanced 4nm LPP+ node for its fabrication.
    • The Tensor G4 and Samsung’s Exynos 2400 are expected to share the same 4LPP+ manufacturing process, illustrating the depth of Google’s partnership with Samsung in the development of its chips.
    • Despite initial plans to switch to TSMC for the production of the Tensor G4, geopolitical tensions between America and India prompted Google to revert to Samsung, highlighting the flexibility and strategic decision-making in Google’s manufacturing partnerships.
  • Technology and Performance Enhancements
    • The Tensor G4 chip may employ Samsung’s innovative ‘Fan-out Wafer Level Packaging’ (FOWLP) technology, which is known for its superior heat resistance and performance capabilities.
    • Improved process yield rates of Samsung’s 4nm process have been a convincing factor for Google to continue its partnership for the Tensor G4, indicating a mutual benefit in terms of technological advancements and production efficiency.
  • Strategic Decisions and Future Plans
    • Google’s decision to contract Samsung for the Tensor G4’s fabrication was influenced by the success of the Galaxy S24 and the positive reception of the Exynos 2400, showcasing the importance of past successes in future planning.
    • Although Google had considered significant changes for the Tensor G4 chip, delays in the “Redondo” chip project led to a continuation with Samsung, underscoring the adaptability in Google’s chip development strategy.
    • Google’s ongoing relationship with Samsung Foundry for the Tensor chips is expected to last until the production of the Tensor G5 in 2025, after which a switch to TSMC is anticipated, highlighting a long-term strategy in Google’s chip development and manufacturing partnerships.

The Evolution of Tensor G4 chip

The evolution of Google Tensor chips represents a significant journey in the realm of mobile computing, particularly for Google’s Pixel devices. The progression from the first-generation Tensor chip to the upcoming Tensor G4 and the anticipated Tensor G5 showcases Google’s commitment to enhancing the performance, efficiency, and capabilities of its devices. Below is a detailed overview of the key milestones and technological advancements in the Tensor series:

  • First-Generation Tensor Chip:
    • Debuted in the Pixel 6 and Pixel 6 Pro, later reused in the Pixel 6a.
    • Featured a unique 2+2+4 core configuration with two high-performance Cortex-X1 cores.
    • Included an in-house developed Tensor Processing Unit (TPU) for advanced machine learning tasks.
    • Received acclaim for enabling features like advanced speech recognition and real-time language translation.

Google Pixel phone showcasing Tensor G4 chip performance enhancements

  • Tensor G2 and G3:
    • Tensor G2, launched in 2022, was 60% faster and more power-efficient than its predecessor.
    • The Tensor G3 chip, used in the Pixel 8 and Pixel 8 Pro, marked a significant leap in performance and power efficiency.
    • Both chips continued to leverage Google’s AI and machine learning prowess to enhance user experiences.
  • Upcoming Tensor G4 and Future Tensor G5:
    • The Tensor G4, codenamed “Zuma Pro,” is expected to be a minor upgrade over the G3, featuring a custom CPU design for the first time.
    • It will be produced using Samsung’s advanced 4nm third-generation SF4P process, promising improved performance and efficiency.
    • The Tensor G5, rumored to be Google’s first fully self-made chip, is anticipated in 2025 with production shifting to TSMC for better power efficiency.

This progression illustrates Google’s strategic shift towards more in-house design and reduced reliance on external partners. By developing its chipsets, Google aims to tailor the hardware specifically to the needs of its Pixel devices, enhancing core areas like speech, language, imaging, and video. The move towards a fully custom chip with the Tensor G5, expected to be built on a more efficient 3nm process, represents Google’s ambition to have full control over the development of its Tensor chips, potentially setting new benchmarks for performance and efficiency in the mobile computing space.

Blueprint of Success: Exynos 2400

In the development of the Google Tensor G4 chip, the Exynos 2400 by Samsung stands as a pivotal blueprint, albeit with significant custom modifications to cater to Google’s specific requirements. The comparative analysis below elucidates the key distinctions and similarities between these two groundbreaking chips:

  • CPU Configuration:
    • Exynos 2400: Features a robust 10-core CPU cluster, designed to deliver high performance and efficiency across a range of applications.
    • Tensor G4: Adopts a slightly pared-down 8-core CPU cluster, a strategic choice by Google to tailor the chip’s performance to the specific needs of Pixel devices.
  • Graphics Processing Unit (GPU):
    • Exynos 2400: Equipped with the Xclipse 940 graphics processor, which is based on AMD’s cutting-edge RDNA3 architecture, offering superior graphics performance.
    • Tensor G4: Utilizes ARM’s Immortalis-G715 GPU, indicating Google’s preference for a different approach to graphics processing tailored to its ecosystem.
  • Manufacturing Process and Packaging Technology:
    • Both chips are expected to be built on Samsung’s advanced 4nm process, showcasing the technical prowess of Samsung’s manufacturing capabilities.
    • The adoption of FOWLP (Fan-out Wafer Level Packaging) technology in both chips signifies a shared commitment to enhancing performance through superior heat resistance and packaging efficiency.

These distinctions underscore Google’s strategy of leveraging the success and technological foundation of the Exynos 2400 while implementing custom modifications to create a chip that aligns with the unique performance and efficiency goals of the Tensor G4. The success of the Exynos 2400, coupled with its role as a foundational blueprint for the Tensor G4, is anticipated to bolster Samsung’s standing in the semiconductor industry, potentially attracting orders from other major players in the tech sphere.

Cutting-edge Google Tensor G4 chip designed for superior efficiency in Pixel devices

Custom CPU Design in Tensor G4 chip

The Google Tensor G4 chip represents a significant leap in Google’s custom CPU design, aiming to enhance performance and efficiency for its Pixel devices. This section delves into the specifics of the Tensor G4’s CPU architecture, comparing its components with its predecessor and highlighting Google’s innovative approach to its development.

CPU Architecture:

  • Core Configuration:
    • The Tensor G4 is an octa-core unit, featuring a diverse array of cores: one high-performance Cortex X4 core at 3.1GHz, three Cortex A720 cores at 2.6GHz for balanced performance, and four Cortex A520 cores at 1.95GHz for efficiency.
    • This setup is a custom solution from Google, indicating a departure from conventional designs to tailor performance specifically for Pixel devices.

Custom Components:

  • CPU, GPU, and Backend System:
    • Google has taken a hands-on approach in designing the Tensor G4, modifying the Exynos 2400 to better suit its needs. This includes customizations across the CPU, GPU, and backend system, showcasing Google’s ambition to create a highly integrated and optimized chipset for its flagship devices.
    • The chip features Arm’s Cortex-X4, A720, and A520 CPU cores, alongside the Immortalis-G715 GPU, which together are expected to deliver a modest improvement over the Tensor G3 in both performance and power efficiency.

Manufacturing and Packaging Technologies:

  • Advanced Processes:
    • Utilizing Samsung’s 4LPP+ fabrication process and ‘Fan-out Wafer Level Packaging’ (FOWLP), the Tensor G4 is poised to benefit from cutting-edge manufacturing and packaging technologies. These choices reflect Google’s commitment to leveraging Samsung’s expertise while maintaining a degree of customization and control over the chipset’s design and performance.

This detailed overview of the Tensor G4’s CPU design illustrates Google’s strategic shift towards more in-house development and customization. By tailoring the chipset’s architecture to the specific needs of Pixel devices, Google aims to achieve a balance between high performance, efficiency, and the advanced capabilities required to power the next generation of mobile computing.

Manufacturing Process: Samsung’s 4nm SF4P

The Google Tensor G4 chip is poised to leverage Samsung’s advanced semiconductor manufacturing capabilities, specifically utilizing the 4nm SF4P process. This section delves into the technical aspects of this process, highlighting its significance in the evolution of Google’s chipset technology.

  • Heat Management and Power Efficiency:
    • The updated Samsung 4nm process is a cornerstone of the Tensor G4’s design, aiming to significantly enhance heat management and power efficiency. This is achieved through an innovative packaging method developed by Samsung, which is crucial for maintaining optimal performance in the compact form factor of smartphones.
  • Performance Enhancements:
    • Samsung’s SF4P process, which underpins the latest Exynos 2400 SoC, delivers substantial performance improvements. Specifically, it facilitates a 1.7x increase in CPU performance and a staggering 14.7x boost in AI performance. These enhancements are critical for the Tensor G4, as they directly contribute to the chip’s ability to handle complex computational tasks and advanced AI algorithms with greater efficiency.
  • Technological Advancements:
    • The Tensor G4’s adoption of Samsung’s 4nm process technology allows for a denser packing of transistors on the silicon area. This architectural improvement is expected to yield significant gains in both performance and power efficiency. The chip is also reportedly based on Samsung’s third-generation 4nm process and utilizes fan-out wafer-level packaging technology, further underscoring Google’s commitment to leveraging cutting-edge semiconductor manufacturing techniques for its Tensor chipset lineup.

The utilization of Samsung’s 4nm third-generation SF4P process represents an updated version of the process used for the current Tensor G3 chip. This progression underscores Google’s ongoing efforts to refine and enhance the capabilities of its Tensor chips, ensuring that each new iteration delivers tangible improvements in performance, efficiency, and overall user experience. The expected inclusion of the Tensor G4 in the 2024 line of Pixel phones, created with the SF4P process, marks another step forward in Google’s chipset innovation journey.

Tensor G4 chip comparison with previous generation for Google Pixel series

Performance Benchmarks and Implications

  • Benchmark Scores:
    • Geekbench 5 Results:
      • Single-Core: The Tensor G4 chipset achieved scores of 1082 and 1024 in single-core tests .
      • Multi-Core: In multi-core tests, scores reached 3,121 and 2,788, indicating a moderate performance.
    • Comparison with Tensor G3:
      • The Tensor G4 scored lower than its predecessor, the Tensor G3, in a Geekbench 6 leak, raising concerns about its performance improvements.
  • Performance Implications:
    • Impact on Pixel Devices:
      • The Tensor G4 is expected to power the Pixel 9 and Pixel 9 Pro, suggesting its performance will directly influence the user experience of these upcoming devices.
    • Potential Consumer Response:
      • Given the underwhelming performance in benchmark tests, there’s a risk that Google’s reputation could be negatively impacted, potentially driving consumers towards alternatives.
    • AI Capabilities:
      • Despite the Tensor G3’s challenges with handling Google’s latest AI algorithms on-device, the Tensor G4 is anticipated to deliver an 8% improvement in multi-core performance, which may enhance its AI processing capabilities.
  • Comparative Performance Analysis:
    • Against Qualcomm Snapdragon and Exynos:
      • The Tensor G4’s performance situates it between mid-range and high-end chips, not as fast as the Snapdragon 8 Gen 3 or the Exynos 2400 but expected to be a decent improvement over the G3.
    • AI Processing Capabilities:
      • Google’s assertion about the unimportance of ‘traditional performance metrics’ is challenged by the Snapdragon 8 Gen 3’s superior on-device AI processing capabilities, underscoring the need for the Tensor G4 to significantly bolster its AI performance.

This analysis highlights the critical reception and performance benchmarks of the Google Tensor G4 chipset, underscoring the implications for Google’s reputation and the potential impact on consumer choice. While the Tensor G4 marks a slight improvement over its predecessor and is expected to enhance CPU, GPU performance, and power efficiency, its moderate benchmark scores and comparison with competitive chipsets suggest that Google may face challenges in convincing consumers of its advantages.

Potential Impact on Pixel Devices

  • Performance Enhancements and User Experience:
    • The introduction of the Tensor G4 SoC in the Google Pixel 9 series is poised to significantly elevate device performance, particularly in handling on-device AI tasks, a core focus of Google’s Tensor initiative. This enhancement is critical for devices like the Pixel Fold 2, which is expected to feature 16GB of LPDDR5 RAM and 256GB of UFS 4.0 storage, thereby ensuring smooth multitasking and ample storage for applications and dat.
    • Additionally, the adoption of Fan-Out Wafer Level Packaging (FOWLP) technology in the Tensor G4 could revolutionize heat management by mitigating thermal bottlenecks. This improvement is anticipated to enable higher clock speeds and extend battery life, directly benefiting the user experience by ensuring the device remains cool during intensive tasks and prolonging usage between charges.
  • Implications for Future Pixel Devices:
    • The Pixel Fold 2, currently in the Engineering Validation Test (EVT) phase, highlights the ongoing development and refinement process of Google’s hardware. The EVT stage is crucial for assessing the device’s performance and reliability, indicating that Google is several iterations away from mass production. This meticulous approach to hardware development underscores Google’s commitment to quality and innovation, setting the stage for future Pixel devices.
    • Furthermore, the Tensor G4’s expected role in enhancing battery performance in the Pixel 9 lineup could address one of the most common consumer concerns regarding smartphone usage, thereby potentially increasing consumer satisfaction and loyalty towards Pixel devices.
  • Strategic Considerations for Google:
    • Google’s decision to continue its collaboration with Samsung for the production of the Tensor G4, despite earlier considerations to switch to TSMC, reflects a strategic choice influenced by the success of previous Pixel devices and the technological benefits of Samsung’s 4nm process. This partnership is indicative of Google’s long-term planning, with the fully custom chip, expected to be produced by TSMC, now delayed to 2025. This delay allows Google to further refine its chipset technology, ensuring that future Pixel devices, starting with the 2024 line, will benefit from even more advanced and efficient chipsets.
    • The cessation of major updates for the last Snapdragon-powered Pixels, including the Pixel 4a, 4a 5G, 5, and 5a, signifies a shift in Google’s strategy towards fully embracing its Tensor chips for future devices. This move not only highlights Google’s focus on leveraging Tensor’s AI capabilities but also indicates a strategic pivot towards enhancing and differentiating Pixel devices in a competitive market.

Close-up view of the Google Tensor G4 chip highlighting its advanced technology

The Future Beyond Tensor G4 chip

  • Transition from “Redondo” to “Zuma Pro” and the Introduction of “Laguna Beach”:
    • The Google Tensor G4, initially codenamed “Redondo,” was slated for the 2024 Pixel series but missed its deadline, leading to the introduction of a new chip called “Zuma Pro” for the Pixel 9 instead.
    • This pivot underscores Google’s agile approach to chipset development, adapting its roadmap to ensure the timely release of new Pixel devices with the most advanced technology available.
  • The Path to a Fully Custom Chip:
    • Google is firmly on the path to developing its next-generation fully-custom Tensor chip, known as “Laguna Beach” or simply “Laguna,” targeting a 2025 release with the Pixel 10 series.
    • This represents a significant milestone for Google, marking its transition to a fully custom System on Chip (SoC), which is expected to bring larger improvements in performance and efficiency.
  • Fabrication and Future Considerations:
    • The Tensor G5 (“Laguna Beach”) is set to be fabricated by TSMC, utilizing the advanced N3E and N3P nodes, which should be fully operational by 2025. However, Google’s access to these nodes remains uncertain, posing potential challenges in the chipset’s development.
    • If TSMC’s capacity is not available, Google may consider using Intel’s 18A node as an alternative for the fabrication of the Tensor G5, indicating Google’s flexibility and strategic planning in navigating the complex semiconductor manufacturing landscape.

This roadmap not only illustrates Google’s ambition and strategic foresight in the development of its Tensor chips but also highlights the challenges and considerations involved in the semiconductor manufacturing process. The transition towards fully custom chips with the Tensor G5 and beyond represents a pivotal moment for Google, potentially setting new benchmarks for performance, efficiency, and integration within the Pixel device ecosystem.

Conclusion

Throughout the detailed exploration of Google’s advancement with its Tensor G4 chip, we have encapsulated the transformative journey from its inception to the anticipated performance enhancements it vows to bring to Pixel devices. The critical partnership with Samsung, the strategic foresight in maintaining a flexible yet innovative approach towards chip manufacturing, and the significant shift towards custom CPU designs underline Google’s unwavering commitment to refining the core of mobile computing. Equally important is the anticipation built around the impending fully custom Tensor G5 chip, marking a promising future for Google’s chipset evolution and its implications for the Pixel series.

As we stand at the cusp of witnessing the real-world impact of the Tensor G4 on user experience and performance benchmarks, it becomes imperative to consider the broader ramifications of Google’s strategic decisions in the semiconductor industry. The potential shift towards TSMC for the future Tensor G5 throws light on Google’s adaptive strategy amidst rapidly evolving technological and geopolitical landscapes. This journey not only underscores Google’s ambition to lead in the realm of mobile technology but also poses intriguing possibilities for future research and innovation in custom chip development, setting a pioneering path that others may follow.

It’s clear that Google’s commitment to enhancing user experience and performance through hardware innovation continues to set new benchmarks. For readers intrigued by the strategic moves of influential global leaders and their impact on technology and beyond, our article on Kim Jong-un offers insightful analysis into one of the world’s most enigmatic figures.

Author

Leave a Reply