During the Consumer Electronics Show 2017, chipmaker Qualcomm elaborated on its impending Snapdragon 835, the next-gen mobile processor expected to power upcoming Android smartphones like the LG G6 and Galaxy S8
On Jan. 10, Qualcomm declared that it would be introducing more support for Google's TensorFlow (an open source software library) for enhanced performance.
The latest processor from Qualcomm — the Snapdragon 835 — will deploy a 10 nm FinFet process node, ensuring improved performance. The chipset will also be compatible with TensorFlow. Powered with this technology, the processor is expected to make AI apps run faster.
"TensorFlow is Google's machine learning software library, and it's open source. This means developers are free to use TensorFlow to create their own models that add artificial intelligence to their mobile and desktop apps or cloud services," shared Qualcomm.
Consumers look for maximum performance from the batteries and want it to last long. The ability of a processor to handle new apps and games is also important. Consumers not only look for processor speed on a mobile device but also its ability to run seamlessly without any snags or delays.
"With an advanced 10-nanometer design, the Qualcomm Snapdragon 835 processor can support phenomenal mobile performance. It is 35% smaller and uses 25% less power than previous designs, and is engineered to deliver exceptionally long battery life, lifelike VR and AR experiences, cutting-edge camera capabilities and Gigabit Class download speeds," notes the company.
The chipset also has a Snapdragon X16 LTE modem packed in to ensure high download speeds of up to 1 gigabit per second. In addition to this, the chip also offers support for Bluetooth 5.0.
What Are The Uses Of TensorFlow?
TensorFlow has already been used by Google in its own Android apps such as Google Cloud Speech and Google Photos.
Qualcomm asserts that it has been developed to run on a processor's processing units. Moreover, TensorFlow is also compatible with Snapdragon 835 as it houses a Kyro 280 CPU, Hexagon 682 DSP and the Adreno 540 GPU.
For apps that deploy TensorFlow, the Snapdragon 835 will peruse the Hexagon 682 DSP so that the applications do not require to task the processor.
To show the difference in performance, as an example, Google developed two AI apps, both of which were effectively able to identify real-world items when placed in front of the handset's camera.
However, the handset which made use of the TensorFlow framework and ran on the Hexagon DSP, was easily able to identify a higher number of objects, at a quicker speed when compared to the same app which deployed the CPU to carry out the task.
If one is thinking of purchasing a new smartphone, then waiting till the Snapdragon 835 processor make its way inside new devices makes sense.