There is a growing demand from all sectors for low-power chipsets that can provide faster processing and deep-learning capabilities
After five years of research Qualcomm has released its next generation of digital chipsets for wearable and IoT devices. The Snapdragon Wear 3100 replaces its predecessor, the 2100.
The Snapdragon Wear 3100 platform is a designed to operate on a low-power system architecture, utilising the chip giant’s low-power co-processor QCC1110 which measures only 21mm2. The co-processor has an integrated deep-learning engine that enables custom workloads and keyword detection.
Included in the power management system in the Snapdragon Wear 3100 platform is the sub-system PMW3100 which helps extend the battery life and improve performance. Qualcomm claims it will provide 4 – 12 hours longer life than its predecessor the W2100 platform.
The CPU in the Wear 3100 platform is Qualcomm’s Quad ARM Cortex A7 which allows CPU clock speeds up to 1.2 GHz, this is complimented with their Adreno 304 GPU which facilitates a display of 640×480 running at 60fps.
Qualcomm are working in collaboration with teams at Alphabet who are working on the Wear operating system designed by Google’s software platform teams, integrating it into a smart watch.
Anthony Murray, GM of Wearables at Qualcomm Technologies International commented in a release that: “With the Snapdragon Wear 3100 Platform, we envisioned a new ultra-low power system architecture and in collaboration with the latest from the Wear OS by Google team, to help deliver a rich interactive mode, bring in new personalized experiences and support extended battery life for tomorrow’s smartwatches.
At a launch event in California Qualcomm and Google announced the first customers who would be integrating the Snapdragon Wear 3100 platform into their wearable devices. Watchmakers Fossil Group, Montblanc and Louis Vuitton will use the Wear 3100 in their next generation of smart wearable devices.
While this technology is starting in the fashionable wearable device sectors, there is a growing demand from industry for small powerful chipsets that can operate out on the edge of networks. Accenture’s 2018 Technology Vision report notes that: “Current predictions suggest that by 2020, smart sensors and other Internet of Things devices will generate at least 507.5 zettabytes of data.”
“Trying to do all of the computational heavy lifting offsite ultimately will become a limiting factor. To fully enable real-time intelligence, businesses must shift event driven analysis and decision processing closer to points of interaction and data generation.”
This need for real-time systems that can operate at the edge is driving the call for ultra low-power co-processors and device architecture. This comes as more and more enterprises see the need for smart surveillance cameras and wearable devices for their employees.
While Qualcomm’s Snapdragon Wear 3100 has been designed with fashionable wearable devices in mind it is a clear indication of the growing trend towards low-power processors with integrated deep-learning engines and high power CPU’s.