Featured Articles

TSMC: Volume production of 16nm FinFET in 2H 2015

TSMC: Volume production of 16nm FinFET in 2H 2015

TSMC has announced that it will begin volume production of 16nm FinFET products in the second half of 2015, in late…

More...
AMD misses earnings targets, announces layoffs

AMD misses earnings targets, announces layoffs

AMD has missed earnings targets and is planning a substantial job cuts. The company reported quarterly earnings yesterday and the street is…

More...
Did Google botch the Nexus 6 and Nexus 9?

Did Google botch the Nexus 6 and Nexus 9?

As expected, Google has finally released the eagerly awaited Nexus 6 phablet and its first 64-bit device, the Nexus 9 tablet.

More...
Gainward GTX 970 Phantom previewed

Gainward GTX 970 Phantom previewed

Nvidia has released two new graphics cards based on its latest Maxwell GPU architecture. The Geforce GTX 970 and Geforce GTX…

More...
EVGA GTX 970 SC ACX 2.0 reviewed

EVGA GTX 970 SC ACX 2.0 reviewed

Nvidia has released two new graphics cards based on its latest Maxwell GPU architecture. The Geforce GTX 970 and Geforce GTX…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Tuesday, 25 March 2014 18:43

Nvidia announces Pascal, its 2016 chip

Written by Fuad Abazovic



3D stacked chip

Nvidia just announced a new 2016 chip and it's called Pascal. This is the successor to the Maxwell DirectX 12 architecture and it will bring a few unique features to the market.

The first interesting feature is NVLink, a bandwidth improving inter-chip communication solution that should solve the bandwidth problem faced by multi GPU platforms. It is based on PCIe 3.0 but it can make things significantly faster, 5 to 12 times faster according to Nvidia. This is a huge gain and it should bring communication between the CPU and GPU to a whole new level.

Pascal also packs 3D stacked memory that is in the same chip packaging and it is about a third of the size of a PCIe card. It will launch in 2016, it will follow Moore's Law and deliver 20 Teraflops per second. Just for comparison, the original Tesla launched back in in 2008 was a 1TFLOPS unit, while Maxwell maxes out at 12TFLOPS.

Nvidia CEO Jen-Hsun Huang showed off the Pascal test module on stage, but we still have to see the demo. 

We will ask around and try to find out what happened to Volta, that is either renamed to Pascal, or it got scrapped, but Nvidia's roadmaps placed Volta right after Maxwell.

nvidia GTC2014Pascal-1

nvidia GTC2014Pascal-2

nvidia GTC2014Pascal-3 

 

Last modified on Tuesday, 25 March 2014 18:59
blog comments powered by Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments