How chipset companies are turning up the heat at Computex4 min read . Updated: 28 May 2019, 01:13 PM IST
- For PC users, Intel has got two new high-end products lined-up
- Qualcomm and Lenovo have got a new 5G enabled line-up of laptops
New Delhi: Chipset companies have taken the Computex 2019 (28 May-1 June) in Taiwan by storm announcing several new products that will not only make your PCs, smartphones, cloud and Edge computing solutions faster but will also improve the deployment of AI solutions.
PC and gaming
For PC users, Intel has got two new high-end products lined-up- a 10nm mobile processor codenamed Ice lake and a new Core i9-9900KS processor. Equipped with gen11 integrated graphics which is a lot more faster than the current gen 9 integrated graphics, the Ice Lake is the first processor designed to run AI (artificial intelligence) solutions on the PC. The Core i9-9900KS is based on the same architecture as the i9-900K, but has been tweaked to achieve 5.0 GHz of over-clocking on all 8 cores simultaneously, making it one of the most powerful desktop processor for gamers. Both chipsets will start shipping sometime year. Intel’s rival AMD took the stage to unveil their new line-up of 7nm based 3rd Gen Ryzen desktop processors. At the top of the table is Ryzen 9 3900X with 12 cores and 24 threads and 4.6GHz turbo speed. It is expected to consume lot less power than Intel’s new processors and is likely to arrive in July at $499.
The other newbies in the Ryzen family will include Ryzen 7 with 8 cores and 16 threads and the Ryzen 5 with 6 cores and 12 threads. The former starts at $199 and the latter at $329.
To make AMD powered machines more appealing to gaming community, AMD has developed a new RDNA (Radeon DNA) gaming architecture, which can muster 1.25x better performance and 1.5x improvement in terms of power efficiency compared to previous generation GCN (Graphics Core Next) architecture. It will provide the foundation for the upcoming 7nm Radeon RC5700 series graphics card, and put it in direct competition with Nvidia’S RTX 2070 graphics. Rumours are that the PlayStation 5 might be using this new GPU along with AMD Zen 2.0 processor.
More details on the new graphics card are expected at the upcoming E3 2019 (Electronic Entertainment Expo) gaming event in Los Angeles in June.
For buyers eagerly waiting for laptops with 5G chipsets, Qualcomm and Lenovo have got a new 5G enabled line-up of laptops. Showcased at the Computex, and dubbed as the Project Limitless, the laptop runs on 7nm 8cx compute platform and uses Snapdragon X55 5G modem that can eke out download speeds of up to 2.5Gbps.
The leading suppliers of mobile CPUs and GPUs, ARM Holdings announced two new chip architectures and GPUs at the Comutex. The new Cortex A77 promises 20% faster IPC (instruction per cycle) performance over its predecessor the Cortex A76, resulting in 35 times faster machine learning performance on the device. ARM also unveiled a new mobile GPU, called Mali-G77 which assures 40% faster graphics performance and 60% boosts in machine learning speeds compared to predecessor the G76. Ina nutshell, this means, the next crop of flagship smartphones will not only be faster will also take lesser time to process AI related tasks on the device.
AMD is also upping the ante in the data centre business with a 2nd gen EPYC based processor powering the EPYC server platforms. In its tests, AMD processors showed a 2x better performances over Intel’s Xeon processors used in workstation and data centres. AMD also announced a breakthrough performance in computational fluid dynamics (CMD) using Microsoft’s Azure HB cloud platform and AMD’s system powered by 1st gen EPYC processor.
A branch of fluid mechanics that uses numerical analysis and data structures to analyse and solve problems, CFD simulations require significant compute time along with specialised hardware.
Intel on its part also showcased its 2nd gen Xeon scalable processors with built in deep learning boost capability, which will improve the speed of AI inference based workloads like image-recognition, object-detection and image-segmentation by up to 14 times compared to its predecessors. This will allow developers to optimise performance in AI based IOT workloads.
Expanding its footprint into Edge computing, Nvidia announced the EGX platforms, to hasten use of AI in Edge applications. According to Nvidia, by 2025, 150 billion machine sensors and IoT devices will stream continuous data that will need to be processed and that is where Edge servers like the ones used in Nvidia EGX platform will come in handy and process data in real time, reducing the amount of data that must be sent to the cloud. At the heart of the EGX platform is
Jetson Nano, a tiny kit which can carry out one-half trillion operations per second of processing for tasks such as image recognition and other real time AI tasks. Nvidia has reportedly roped in Dell EMC, Cisco, Fujitsu and Lenovo as EGX server partners, while in IOT space they have tied up with Microsoft and AWS.
“Hospitals are increasingly using AI to predict adverse patient events, support clinical decisions making and operate more efficiently. However, these applications rely on patient data. Nvidia’ EGX AI Edge computing platform provides hospitals easy AI infrastructure to deliver real time AI and scale to thousands of AI applications needed to improve patient care and reduce cost of care delivery," said Keith Dreyer, chief data science officer at Partners Healthcare at Harvard Medical School, in a press statement shared by Nvidia.