CEO Richard Yu Unveils the Kirin 970, Huawei’s first mobile AI computing platform
BERLIN — September 2, 2017 – Today at IFA 2017, Huawei Consumer Business Group unveiled a new era in smartphone innovation. As part of CEO Richard Yu’s keynote address, he revealed Huawei’s vision for the future of artificial intelligence with the launch of the Kirin 970. By combining the power of the cloud with the speed and responsiveness of native AI processing, Huawei is bringing AI experiences to life and changing the way we interact with our devices.
“As we look to the future of smartphones, we’re at the threshold of an exciting new era,” said Richard Yu, CEO of Huawei Consumer Business Group. “Mobile AI = On-Device AI + Cloud AI. Huawei is committed to developing smart devices into intelligent devices by building end-to-end capabilities that support coordinated development of chips, devices, and the cloud. The ultimate goal is to provide a significantly better user experience. The Kirin 970 is the first in a series of new advances that will bring powerful AI features to our devices and take them beyond the competition.”
After years of development, Cloud AI has seen broad application, but user experience still has room for improvement, including latency, stability, and privacy. Cloud AI and On-Device AI can complement each other. On-Device AI offers strong sensing capabilities, which are the foundation of understanding and assisting people. Sensors produce a large amount of real-time, scenario-specific, and personalized data. Supported by strong chip processing capabilities, devices will become more cognitive of user needs, providing truly personalized and readily accessible services.
Kirin 970 is powered by an 8-core CPU and a new generation 12-core GPU. Built using a 10nm advanced process, the chipset packs 5.5 billion transistors into an area of only one cm². Huawei’s new flagship Kirin 970 is Huawei’s first mobile AI computing platform featuring a dedicated Neural Processing Unit (NPU). Compared to a quad-core Cortex-A73 CPU cluster, the Kirin 970’s new heterogeneous computing architecture delivers up to 25x the performance with 50x greater efficiency. Simply put, the Kirin 970 can perform the same AI computing tasks faster and with far less power. In a benchmark image recognition test, the Kirin 970 processed 2,000 images per minute, which was faster than other chips on the market.
New developments in AI require joint effort across the entire value chain, involving tens of millions of developers, and the experience and feedback of hundreds of millions of users. Huawei is positioning the Kirin 970 as an open platform for mobile AI, opening up the chipset to developers and partners who can find new and innovative uses for its processing capabilities.
About Huawei Consumer BG
Huawei’s products and services are available in more than 170 countries, and are used by a third of the world’s population, ranking third in the world in mobile phone shipments in 2015. Fifteen R&D centers have been set up in the United States, Germany, Sweden, Russia, India and China. Huawei Consumer BG is one of Huawei’s three business units and covers smartphones, mobile broadband devices, home devices and cloud services. Huawei’s global network is built on 20 years of expertise in the telecom industry and is dedicated to delivering the latest technological advances to consumers around the world.
For more information, visit Huawei Consumer BG online: http://consumer.huawei.com/en/
| IFA 2017 Key Messages: A NEW BRAIN IN YOUR MOBILE | |
| Huawei AI Strategy | Mobile AI = On-Device AI + Cloud AI |
| On-Device AI | On-Device AI Helps Huawei Deliver Breakthrough Device Experiences |
| Kirin 970 | Huawei’s first mobile AI computing platform |
| CBG Business Performance in 2017 H1 | Huawei is seeing solid growth and rapid expansion in high-end markets, combined with increasing brand recognition and influence. |
The evolution of AI is driving exciting new advancements in technology. Mobile AI = On-Device AI + Cloud AI: Huawei is committed to developing smart devices into intelligent devices by building end-to-end capabilities that support coordinated development of chips, devices, and the cloud. The ultimate goal is to provide a significantly better user experience. On-Device AI needs to fully leverage the capabilities and value of devices while making the most of the massive amount of information, services, and strong computing capabilities brought about by big data and cloud. By extensively integrating smart devices with the cloud and big data, On-Device AI will continue to advance the intelligent evolution of devices.
AI is a fundamental technology, and will continue to evolve based on the development of new applications. With breakthroughs in performance, on-device AI computing will result in a broad range of entirely new AI experiences. However, new developments in AI require joint effort across the entire value chain, involving tens of millions of developers, and the experience and feedback of hundreds of millions of users.
To support these collaborative efforts, Huawei will release the Kirin 970 as an open Mobile AI computing platform, providing comprehensive support for multi-app modes and different machine learning frameworks. This will enable developers to take full advantage of the Kirin 970’s AI computing power using the frameworks and models they prefer, while providing a platform for new and imaginative applications. Together with developers, Huawei hopes to forge a healthy and robust Mobile AI ecosystem.
In the age of artificial intelligence, intelligent devices are becoming an extension of who we are. They provide us with efficient, useful, and readily accessible information and services, and offer us more compelling experiences. AI is set to propel the intelligent development of smart devices to the next level. Instead of passively responding to user needs, AI will sense our mood and environment, and proactively offer new ways for us to interact with the world around us through targeted services.
Huawei believes that AI is still in its early stages of development. The technology is set to experience rapid development, and will soon create value across many domains. AI presents new requirements for chip architecture. While AI today is mainly about heterogeneous computing, brain-inspired computing will soon emerge to drive new advances in AI. AI also raises new requirements for software frameworks. In response, open source frameworks have become a basic approach to ecosystem development. Algorithm innovation and optimization as well as basic services are becoming the focus of competition.
AI development has the potential to profoundly influence society, gradually change the way we live and work, and improve productivity. However, AI also poses real challenges that can impact lives, including its impact on employment and applications related to security, ethics, and privacy. These can lead to legal, ethical, political, and religious implications that we must take seriously. We must develop the relevant technical standards, laws, and regulations in advance to deal with these challenges.
- On-Device AI Helps Huawei Deliver Breakthrough Device Experiences
- Mobile AI = On-Device AI + Cloud AI
According to Huawei, Mobile AI = On-Device AI + Cloud AI. To deliver a fully intelligent experience, we need to optimize the synergy between devices and the cloud. After years of development, Cloud AI has seen broad application, but user experience still has room for improvement, including latency, stability, and privacy. Cloud AI and On-Device AI can complement each other. On-Device AI offers strong sensing capabilities, which are the foundation of understanding and assisting people. Sensors produce a large amount of real-time, scenario-specific, and personalized data. Supported by strong chip processing capabilities, devices will become more cognitive of user needs, providing truly personalized and readily accessible services.
Perception: A device’s ability to sense is the foundation of AI and the starting point for human-machine interaction. What a device can sense and how accurate that sensing is will have a direct impact on the level of intelligence, and determine whether the interaction is natural and convenient.
Cognition: While sensing is the foundation for AI functionality, cognition is what will truly maximize the potential of intelligent AI. With accurate cognition enabled by powerful sensing, intelligent devices will truly “know you and understand you” and offer personalized services.
Security: Greater quantities and types of data lead to greater intelligence. However, all this data can lead to new security and privacy challenges. As the general level of device intelligence increases, it is also necessary for us to develop a robust system to ensure security and protect privacy. Data anonymization is critical to ensuring information security in the intelligent device era, and must be performed to ensure On-Device AI and Cloud AI operate and function independently.
Power: Intelligent devices will have to process tasks in more complex scenarios, so they must have a powerful chip that can optimize the use of battery power and support dynamic task processing.
- For a wider range of AI applications, the industry needs a new Mobile AI computing architecture.
As AI technology sees wider application, demand for an intelligent mobile experience will continue to increase. People want a device that can understand and predict their needs in an innately human way, providing them with relevant information and services based on that understanding. At its core, AI technology relies on the ability to process massive amounts of data. However, traditional computing architecture based on CPUs, GPUs, and DSPs are no longer sufficient to meet overwhelming demand for computing performance in the age of artificial intelligence. The entire industry is actively exploring new computing architectures to overcome this challenge.
On-device AI processing capabilities are of critical importance, because mobile devices need to process data in real time, any time, while providing optimal protection of user data. The Kirin 970 helps address the challenges specific to on-device AI with an efficient heterogeneous architecture.
Smartphones are different from servers in many respects, including size, power supply, power consumption, and heat management – all of which place restrictions on smartphone chipset design. Mobile SoCs have to deliver the best possible performance, while ensuring the highest possible density of core functions with optimal use of energy.
The Kirin 970’s development team introduced a new innovative HiAI mobile computing architecture with a dedicated Neural Network Processing Unit (NPU), which delivers an AI performance that far surpasses any CPU/GPU build. Compared to a quad-core Cortex-A73 CPU cluster, the Kirin 970’s new heterogeneous computing architecture delivers up to 25x the performance with 50x greater efficiency. Simply put, the Kirin 970 chipset can perform the same AI computing tasks faster and with far less power. In a benchmark image recognition test, the Kirin 970 processed 2,000 images per minute, which was faster than other chips on the market.
The Kirin 970 breaks through existing bottlenecks in mobile hardware, effectively redefining the collaborative relationship between devices and the cloud. Based on big data processing in the cloud, cloud-based AI systems can provide mobile devices with general or collective intelligence. Meanwhile, through on-device training and applications, the Kirin 970’s on-device AI systems provides users with great individual intelligence, delivering a rich intelligent experience. Devices with on-device AI aren’t limited to tasks like object recognition. They can more thoroughly understand their users and satisfy demand for a personalized and real-time AI experience, while providing thorough protection of their private data.
- Redefining mobile internet experience: 1.2Gbps peak download, a brand new GPU, and an advanced dual ISP
The Kirin 970 supports LTE Cat.18 DL and Cat.13 UL connections. Able to reach peak download rates of 1.2Gbps, the Kirin 970 supports the highest possible download speeds on all carrier networks, worldwide.
The Kirin 970 introduces the new generation of Mali G72 12-core GPU and makes it the first to market. Its graphics processing performance is 20% better than the previous generation (Kirin 960 with the Mali G71MP8), with a 50% improvement in energy efficiency. Users are in for much longer battery life and a smoother 3D gaming experience.
The Kirin 970’s proprietary ISP has also undergone a complete upgrade. The new dual ISP, enabled by Kirin 970’s Mobile AI computing capability, supports scene recognition and intelligent photography, which automatically chooses the settings that are best suited to a given object in a specific environment. This will help users take more natural and professional pictures. The chipset also supports an advanced autofocus function for clearer portraits, and a motion capture feature that enables perfectly focused images of objects in motion. Enhanced noise reduction across multiple bands of spectrum makes for more clear and natural photographs in low-light conditions.
- Business Performance in 2017 H1
- Huawei is seeing solid growth and rapid expansion in high-end markets, combined with increasing brand recognition and influence
- Overall business performance is growing steadily
Earnings in H1 2017 amounted to CNY 105.4 billion, presenting a year-over-year growth of 36.2 percent. In addition, Huawei smartphone shipments reached 73.01 million units, an increase of 20.6 percent year-on-year.
According to data from IDC, in Q1 2017 (Q2 data is yet to be released), Huawei’s share of the global smartphone market climbed to 9.8 percent.
The MBB and home product line continued to perform well. The tablet business registered remarkable growth against a sluggish market, with shipments growing almost 60 percent year-on-year.
- Continuous breakthroughs in high end markets
In major overseas markets (21 countries), shipments of high-end smartphones (P series & Mate series) increased by over 100 percent year-on-year.
Key models sales overview：
Shipment of Mate 9 & Mate 9 Pro: 8.5 million+.
Shipment of P10 & P10 Plus: 6 million+.
Shipment of nova 2 & nova 2 Plus: 1 million+ within one month since being launched.
Sales volume of Honor 9: 1 million+ within 28 days since being launched.
Huawei’s market share of devices in the‘500 USD and above’ bracket expanded by a remarkable 8.2 percent to 14%.
- Regional market share highlights
Huawei and Honor together constantly become the leading smartphone market share in Greater China. According to GfK data, as of May 2017, Huawei claimed 22.1 percent of market share in Greater China.
Steady and robust growth in Europe
Huawei experienced major breakthroughs in Asia-Pacific emerging markets: Thailand, Malaysia, Japan, South Korea and Russia.
Stable growth in the Middle East, Africa and Latin America.
- Global brand influence is increasing
- Continued improvement in brand awareness in key countries. Global brand awareness reached 81 percent in 2016, an improvement of five percent over the previous year
Sundus Anwar believes in the power of clear, straightforward writing. Her blog posts tackle everyday topics with relatable insights and easy-to-follow advice. With a conversational style, she makes complex subjects feel understandable. She’s dedicated to sharing knowledge and empowering readers to take action. Find her latest posts on trending in social.
