According to the latest leak NVIDIA is already working on two new Pascal-based GP100 GPUs, both Graphics cards will be manufactured by TSMC at 16nm process node accompanied by memory HBM2.
NVIDIA Is Woring On Two Pascal-based GPUs: Manufactured on 16nm and Powered By HBM2
The first one has a 4-Hi stack memory while the second model is featuring 8-Hi. Both reach a memory interface of 4096 bits at a frequency of 1.00 GHz (twice as fast as the HBM memory) and it is speculated that the simplest model is the GeForce GTX 1080, while the second model would be aimed at the professional market (a new Quadro or Tesla).
The HBM1 memory has a physical limitation of up to 4 GB of capacity and a bandwidth of 512 GB / s, memory HBM2 allows the use of up to 16/32 GB of VRAM that can exceed 1024 GB / s.
Previously, Jen-Hsun Huang, NVIDIA co-founder, confirmed that Pascal will use the second generation of memory known as HBM HBM2, which will make their graphics cards offer a bandwidth that starts from 300 GB / s to around amazing 1.2 TB / s (1200Gb / s). The next generation of AMD will use both the HBM memory as HBM2, hoping that this last memory is used in top-end models such as the Radeon Radeon R9 490 and R9 490x.
“We have found over the years to be able to focus on just one thing, which is visual computing, and be able to leverage that one thing across PC, cloud, and mobile, and be able to address four very large markets with that one thing: gaming, enterprise, cloud, and automotive,” said Mr. Huang. “We can do this one thing and now be able to enjoy all and deliver the capabilities to the market in all three major computing platforms, and gain four vertical markets that are quite frankly very exciting.” …“I cannot wait to tell you all about it, you are just going to have to wait just a little longer,” – CEO, Nvidia
Pascal could very well be poised to offer a lineup of one of the first 4K capable single-Gpu cards at the 4k@60 fps standard.
Remember that NVIDIA Pascal will not arrive until mid-2016.