“Big Pascal” as we’d like to call it has been spotted in Zauba’s shipping data base going from Taiwan to Bangalore India. The data indicates that at least 9 engineering samples have been shipped so far, each one costs nearly 10 times as much as a GM200-310 GPU found in the GTX 980 Ti. So if you were wondering what 16FF+ yields look like right now, that should give you a rough idea.
from WCCFtech
TSMC’s 16FF+ (FinFET Plus) technology can provide above 65 percent higher speed, around 2 times the density, or 70 percent less power than its 28HPM technology. Comparing with 20SoC technology, 16FF+ provides extra 40% higher speed and 60% power saving. By leveraging the experience of 20SoC technology, TSMC 16FF+ shares the same metal backend process in order to quickly improve yield and demonstrate process maturity for time-to-market value.
Nvidia has confirmed that its “Volta” GPU architecture will be the successor to Pascal and is set for a 2018 introduction in the consumer market. While the first Volta powered supercomputers are expected to be operational by 2017. Volta represents Nvidia’s sixth generation of General Purpose GPU architectures since the introduction of the company’s first unified shader graphics architecture code named Tesla. Which debuted with the company’s highly successful GeForce 8 – 8000 series – back in 2006.
from WCCFtech
上記ロードマップにも書かれていますが,Pascal後継とされるVoltaに関してもスーパーコンピュータ向けに2017年から提供が始まり,その後コンシューマー向けとして2018年頃を目処に流通を開始するようです.本来Maxwellの後継として検討されていたVoltaですが,Hybrid Memory Cubeの開発が難航したこと及びHBMの登場から,Pascalを間に挟むことで対応しようとしているそうです.