Pascal microarchitecture.
DirectX 12 feature level 12_1 or higher.
Successor to the GM200 GPU found in the GTX Titan X and GTX 980 Ti.
Built on the 16FF+ manufacturing process from TSMC.
Allegedly has a total of 17 billion transistors, more than twice that of GM200.
Taped out in June 2015.
Will feature four 4-Hi HBM2 stacks, for a total of 16GB of VRAM for the consumer variant and 32GB for the professional variant.
Features a 4096bit memory interface.
Features NVLink and support for Mixed Precision FP16 compute tasks at twice the rate of FP32 and full FP64 support.
Read more:
http://wccftech.com/nvidia-pascal-nvlink-200-gbs-interconnect-hbm2-stacked-memory-1-tbs-bandwidth-powering-hpc-2016/#ixzz3nhrPXZRz
10X Faster - Sure No : it's only 10x faster on hyperspecific metrics. But not on general real-world, gaming, or compute performance.
I expect Pascal-vs-Maxwell to be similar (in terms of real performance comparisons) to Maxwell-vs-Kepler. Meaning that most of the engineering refinements are about power efficiency rather than raw performance gains. At least initially, until the "entry-level" Pascal GPUs stage upwards through better silicon yields, tweaked/debugged ASIC revisions, and effective driver support.
HBM 1.0 vs GDDR5 , 980Ti is faster than Fury X with GDDR5 , but remember Pascal raw performance will be on its high end cards and I suspect Nvidia will pullout there GDDR5 variants , it will be a similar situation to what AMD did. For Now Until June 2016 , around 8 Months - Still the 980Ti is a great upgrade I recommend to do.