AMD and Nvidia upgrade with Polaris and Pascal for Computex 2016

After more than four years at 28 nanometers, AMD and Nvidia are ready to take the next big step to new manufacturing technologies. The shift means that the players will be able to double the number of transistors on a given surface and reduce the current leakage, two points that guarantee both higher performance and energy efficiency.

A new report from Digitimes states that AMD and Nvidia are getting ready for a battle for attention on this year’s edition of Computex, which starts on 31 May. During this year’s major electronics and component trade fair, the companies are expected to pay attention to their architectures Polaris (AMD) and Pascal (Nvidia).

AMD promises double energy efficiency with Polaris compared to their current range of graphics cards. This is attributed to architectural improvements and not least the transition to 14 nanometers, a manufacturing technology from Samsung that also has Globalfoundries as licensee.

For Nvidia’s part, the report claims that Pascal sees the light of day already during the GPU Technology Conference (GTC) in April, but whether it will be in the form of a finished product or something else is not clear. However, the company is said to announce the Geforce GTX 1080 and GTX 1070 in June, ie around the same period as AMD Polaris.

It is also interesting that for the first time in years, Nvidia and AMD will not use the same manufacturing technology. Nvidia remains faithful to its partner TSMC, whose 16-nanometer technology is a direct competitor to 14 nanometers from Samsung. However, if this is an advantage or a disadvantage, it is too early to say.

Assuming the information is correct, the next generation of graphics cards from AMD and Nvidia will arrive at about the same time. The sharp launch from both parties is said to take place during the third quarter of the year (July – September), while mid-range graphics cards from Nvidia will be delayed until the fourth quarter of this year.