Qualcomm has expanded its mobile roadmap in the 7 and 6 series to include the latest Qualcomm Snapdragon 730, 730G and 665 Mobile Platforms.
Commercial devices based on Snapdragon 730, 730G and 665 are expected to be available in mid-2019.
The company is also bringing its artificial intelligence (AI) expertise to the cloud with the Qualcomm Cloud AI 100.
The Snapdragon 730 brings on-device AI to mobile experiences with image capture, gaming and optimized performance.
Packing 2x the power of its predecessor, Qualcomm Technologies’ 4th generation multi-core Qualcomm AI Engine accelerates on-device interactions for camera, gaming, voice and security. The Qualcomm Hexagon 688 Processor inside Snapdragon 730 supports improved base scalar and Hexagon Vector eXtensions (HVX) performance, as well as the new Hexagon Tensor Accelerator—now adding dedicated AI processing into the Hexagon Processor.
For the first time in the 7 series, the Snapdragon 730 features the Qualcomm Spectra 350, featuring a dedicated Computer Vision (CV) ISP, to provide up to 4x overall power savings in Computer Vision compared to the previous generation. The lower power and faster CV can capture 4K HDR videos in Portrait Mode (Bokeh). The CV-ISP is also capable of high resolution depth sensing and the ability to support triple cameras that feature ultra-wide, portrait and telephoto lenses. It also captures photos and videos in the HEIF format so users can document life from multiple angles and store it all at half the file size to the previous generation.
The Snapdragon 730 features the Adreno 618 GPU and is the first 7 series platform to integrate Vulkan 1.1, a next-generation graphics library that integrates enhanced developer tools and runs off 20% less power than Open GL ES, resulting in better graphics and longer battery life. The surround sound audio experience is enhanced by both Qualcomm aptX Adaptive audio and Qualcomm Aqstic audio technology.
Based on an 8nm process node, the Qualcomm Kryo 470 CPU and the Adreno 618 GPU are engineered with new architectures, enabling up to 35% performance uplift for each core. Featured in the Snapdragon 730, the Snapdragon X15 LTE Modem offers cellular download speeds up to 800 Mbps and one of Qualcomm's first platforms that comes Wi-Fi 6-ready.
The Snapdragon 730G has been designed for the elite gamer. It offers up to 15% faster graphics rendering over the Snapdragon 730, featuring an enhanced Adreno 618 GPU.
The 730G supports HDR gaming and according to Qualcomm, it helps reducing janks by up to 90% in games running at 30fps with the Jank Reducer feature. In addition, anti-Cheating Extensions help keep gamers on the same level playing field.
Wi-Fi latency manager optimizes network settings and Wi-Fi configurations to support efficient online gaming and fast gameplay. Additionally, as part of the Snapdragon 730G offering, there are also improvements in the camera pipeline, adding support for Cinemagraph, low power voice activation and HD super slow-mo at 960fps.
The Snapdragon 665 Mobile Platform supports the 3rd generation Qualcomm AI Engine, featuring the Hexagon 686 DSP, with the addition of HVX. It is engineered to deliver up to 2x faster AI on-device processing compared to the previous generation.
The Spectra 165 ISP in the Snapdragon 665 is reinforced with AI, providing scene recognition and auto adjustments like HDR. The Snapdragon 665 supports camera features including Hybrid Autofocus, Optical Zoom, Zero Shutter Lag Video, Video Style Transfer, and Triple Camera with support for Telephoto, Wide and Ultra-Wide, and can take super resolution snapshots at up to 48 megapixels.
For gaming enthusiasts, Snapdragon 665 is designed to allow users to play latest games with surround-sound audio and realistic graphics. Snapdragon 665 features the Adreno 610 GPU, which supports Vulkan 1.1. Snapdragon 665 also supports aptX Adaptive Audio and Qualcomm Aqstic.
Utilizing the Kryo 260 CPU, this energy-efficient platform disperses tasks between four performance and four efficiency cores. Snapdragon 665 also features the Snapdragon X12 LTE Modem offers LTE download speeds up to 600 Mbps and integrated 802.11ac Wi-Fi.
|Snapdragon 665||Snapdragon 675
|CPU||4x Kryo 260 (CA73)
4x Kryo 260 (CA53)
|2x Kryo 460 (CA76)
6x Kryo 460 (CA55)
|2x Kryo 470 (CA76)
6x Kryo 470 (CA55)
|GPU||Adreno 610||Adreno 612||Adreno 618|
|DSP||Hexagon 686||Hexagon 688|
25MP single / 16MP dual
25MP single / 16MP dual
36MP single / 22MP dual
|Memory||2x 16-bit @ 1866MHz
|2x 16-bit @ 1866MHz
1MB system cache
|Integrated Modem||Snapdragon X12 LTE
DL = 600Mbps
3x20MHz CA, 256-QAM
UL = 150Mbps
2x20MHz CA, 64-QAM
|Snapdragon X15 LTE
DL = 800Mbps
3x20MHz CA, 256-QAM
UL = 150Mbps
2x20MHz CA, 64-QAM
H.264 & H.265
|Manufacturing process||11nm LPP||11nm LPP||8nm LPP|
Artificial Intelligence Inference Processing to the Cloud
Qualcomm is also bringing the company’s artificial intelligence (AI) expertise to the cloud with the Qualcomm Cloud AI 100.
“Our all new Qualcomm Cloud AI 100 accelerator will significantly raise the bar for the AI inference processing relative to any combination of CPUs, GPUs, and/or FPGAs used in today’s data centers,” said Keith Kressin, senior vice president, product management, Qualcomm Technologies, Inc. “Furthermore, Qualcomm Technologies is now well positioned to support complete cloud-to-edge AI solutions all connected with high-speed and low-latency 5G connectivity.”
In addition, Qualcomm Technologies is supporting developers with a full stack of tools and frameworks for each of our cloud-to-edge AI solutions.
According to Qualcomm, the Cloud AI 100 offers:
- More than 10x performance per watt over the industry’s most advanced AI inference solutions deployed today
- A highly efficient chip specifically designed for processing AI inference workloads
- 7nm process node
- Available support for software stacks including PyTorch, Glow, TensorFlow, Keras, and ONNX
Qualcomm said it would provide more details of its new product later this year. The chip is aimed at inferencing work – making decisions based on analysis of data streams such as digitized voice or images. Qualcomm claims the chip is more than 50 times as powerful in processing AI workloads as the company’s flagship mobile chip. The new offering will be in production in 2020.
With the new offering, Qualcomm is joining a crowded race to provide the fundamental building blocks of new forms of computing, in a market worth $17 billion by 2025.
The business of recognizing images, speech and making data-based decisions by computers is expanding rapidly and stretching the capabilities of the type of chips that have dominated since the birth of the personal computer in the early 1980s.
Cloud-computing providers are increasingly moving toward designing their own components.
Intel , whose processors are pervasive in data centers, has bought smaller companies that make different types of chips with attributes helpful in handling work for AI accelerators. It’s also adding capabilities to its processors to try to make them more efficient and better at making sense of the increasing flood of data created by phones and machines. Qualcomm is also playing catch up with Nvidia which has built a multibillion-dollar annual business adapting its graphics chips for data-center use.