Last week there was an announcement by Global Foundries that they were licensing Samsung’s 14nm FinFET process for their own use. This immediately started the internet going with talk about how AMD would be using this new process despite there not being any indication that AMD was ready for 14nm for APUs, CPUs, or GPUs for that matter. Now, in true internet fashion the rumor mill has shifted from AMD to NVIDIA. The claim is that NVIDIA will soon utilize Samsung’s 14nm FinFET tech too.
During NVIDIA’s recent GTC announcements the world was shown the new Titian X with 12GB of GDDR5. This impressive monster of a card has shown that it has a large amount of power to push your games and other graphical information. While the Titian X received adoration and several very positive reviews from the technical press there was another story that was also very important. This was the conversation about NVIDIA’s next GPU, Pascal.
It seems that the little “error” that was made when listing the specifications of NVIDIA’s GTX970 it getting ugly. A Law firm in Alabama has taken a single complaint and request for monetary compensation and pulled it into a full blown class action law suit against both NVIDIA and Gigabyte. For those of you that have been living under a rock for the last couple of months, where is a summary of the issue.
CES 2015 – Las Vegas NV
At the Consumer Electronics Show, nVidia showed off something that was interesting, but also confusing: The Tegra X1. This is a Maxwell based version of the Tegra that sports 256 Maxwell GPU cores, an 8-core 64-Bit CPU and is capable of pushing 4k 60Hz 10-Bit video in either H.265 or VP9. This new member of the Tegra family was hailed by nVidia CEO Jen-sun Huang as the first mobile superchip for its expected performance. Of course nVidia has always like to use the term “super” when talking about new products. I can remember them trying to coin the term Super Phone when Tegra 2 hit the market.
The US patent system is broken. There is not really much more to say about that it is and has been broken for many years and probably will continue to be broken for many years to come. Now why am I bringing this up? Well in looking over the way patents are filed, processed, and approved you will find that there are a large number of patents that are essentially the same thing just with a little twist. This type of patent allows for someone to put a legal hold on a very broad concept and then comeback and sue someone at a later date. Over the last few years we have watched a number of companies work the system with this type of patent and worse.
When I first started covering the computer world the most common resolution was 640x680 with the hard core gamers getting 800x600. The dream of the day, which some called the golden age of gaming, was 1600x1200 with around 30Fps. Now the dream is “photo realistic” resolutions without the need for heavy anti-aliasing and texture filtering. Even in the mobile world this is becoming a bigger issue with retina displays on the Apple side of the world and 3 and 4k screens on the PC side.
It is the dream of many computer enthusiasts to have a mobile system that is at least as powerful (if not as flexible) as their desktops. I know that I personally have spent a small fortune on finding mobile devices that can run all of the rendering and editing software I have without tearing my shoulders off my body. For their part both Intel and AMD have cut the gap between desktop and mobile performance in the CPU world, but mobile GPUs have traditionally lagged way behind their desktop counterparts.
On Friday the big news was that NVIDIA had launched two separate suits against Samsung and Qualcomm for patent infringement (International Trade Commission (ITC) and the US District Court in Delaware). In the suits NVIDIA claims that the two companies have willfully infringed on their patented GPU technology and (of course) they would like to start banning devices. So far the list includes the yet to be released Note 4 and Note Edge. The patents in question appear to be Patent Nos. 6198488, 6992667, 7038685, 7015913, 6697063, 7209140 and 6690372.
According to recent rumors it seems that TSMC and Samsung will be able to push out 14nm full node and 16nm half node FinFET products earlier than anticipated. This is certainly going to be good news for many customers of the two foundry companies including Apple and nVidia.
Not all that long ago we reported that there was a chance that nVidia might skip over the 20nm half-node and move to the next full node in line due to issues that TSMC has been having with their 20nm process. 20nm has been the dream of GPU manufacturers for some time and despite the money that has been thrown at it, neither TSMC nor GlobalFoundries can seem to get it right. One reason for this could be due to leakage while another is just that as the process gets smaller building large and complex devices using a half-node just does not work.