TSMC itself promised to deliver industry leading mobile SoCs to the market and open the 16nm FinFET node for business completely. There have been idle speculation that the Red and Green GPUs are actually going to be be based on the 16nm FinFET process instead of the 20nm so that they catch up with lost mileage of Moores Law, but to be honest, at the current state of affairs, it would take at-least a year before the node becomes mature enough for any kind of high performance ASIC work. Since we have at least a year before that theoretically happens, unless the giants are willing to wait it out even longer, we are probably going to get 20nm GPUs.
ARM and TSMC Reveal the First 16nm FinFET Heterogeneous Architecture Processor
TSMC recently entered Risk Production of 16nm FinFET and has since produced the first commercial circuit on the same; an ARM core with Heterogeneous Architecture. The Multi-Core processor is based on the big.LITTLE ARM core and stems from the Cortex A57. The resulting CPU was clocked in at speeds of 2.3Ghz.
ARM's First 16nm FinFET Processor Tested, Supports Heterogeneous Architecture and Clocked at 2.3 Ghz
Many companies around the world have collaborated to optimize the ARMv8 64 Bit architecture and the 16FF+ iteration is no exception. The forecasts indicate that ARM expects the performance of the A57 to increase by 11%, and,the power envelope of the A53 to reduce by 35% when it gets the same treatment. The initial result, such as the chip created today, are quite encouraging. Pete Hutton, executive VP and president of ARM product groups, said the joint effort will eventually transform end-user experiences across the next generation of consumer devices and enterprise hardware: "This silicon proof point with ARM Cortex-A57 and Cortex-A53 processors demonstrates the additional benefits in performance and power efficiency that 16nm FinFET technology delivers to big.LITTLE implementations," said Hutton.