20nm is arguably one of the hotter (figure of speech!) processes out there so it was of considerable interest to me that I spotted one of the rumored 20nm APUs 'Nolan' in the AIDA64 driver update. Driver updates are usually a highly authentic and reliable method of predicting product existence and even arrival and its name popping up in the driver database means AMD might be making an announcement pretty soon. Please be advised that some of the information mentioned in the article is not confirmed.
AMD Nolan APU Spotted in Driver Update - Slated to be on the 20nm node
To those who might not know the Nolan and Amur platform is very special. It is not your average platform but AMD's attempt to break into the low power mobility sweet spot that Intel Cherry Trail currently targets. Using the 20nm process will give the chip a much needed advantage that a 28nm chip could not have provided, however, Intel's offerings are ofcourse at the 14nm node. AMD's APU will utilize HSA so theoretically, with the right developer ecosystem, the Nolan APU is an extremely strong contender.
Nolan is based on the Puma+ architecture and will be virtually identical to its ARM counterpart apart from the fact that it is x86 and not ARM. Both Nolan and Amur are slated to be produced on the 20nm node and have a release time stamp of Q3 2015. Now one of the more interesting questions that we can ask ourselves is whether AMD is going after Tegra? Tegra K1 caused a lot of hype with its recent Unreal Engine 4.0 demo and the Android Extension Pack and probably the only company that can well and truly match Kepler based mobile GPUs is AMD.
AMD Nolan is Beema’s replacement for 2015. Unlike Project Nolan, which is x86, AMD Amur will support on ARM based heterogeneous system architecture. It will be designed upon the Cortex A57 and will supposedly ship in third quarter of 2015. Amur will support both Android and Linux while Nolan will support both Android and Windows. Intel has been having major difficulties with players such as Qualcomm and Media Tek and now it seems another competitor is entering the fray. It doesnt help that the tablet market has been declining steadily since last year. The 20nm process while not good for high performance applications is a very good idea for low powe ASICs such as the Nolan APU. The real question that I am sure that will be on everybody's minds is whether we are looking at GloFo or TSMC. Personally I think the chances of it being GloFo are significantly more than it being TSMC. From the reports we have received TSMC just doesn't appear to be that interested in 20nm while Global Foundries is bustling with potential in the same