NVIDIA Announces Pascal GPU Family

NVIDIA CEO Jen-Hsun Huang took the center stage at the company’s annual GPU Technology Conference in San Jose, California to announce the Pascal GPU family that will succeed the recent Maxwell family. The GPU roadmap will make the new generation faster, more energy efficient and smaller. We shall be taking a closer look at the features of the Pascal, starting with stacked DRAMS. By using chip-on-wafer 3D integration DRAM chips can be stacked on top of each other creating dense modules with wider interfaces inside the GPU itself. This helps the GPUs to transmit data efficiently and makes the GPU more compact. NVIDIA says with the help of 3D memory they can double the memory capacity and quadruple the energy efficiency.


The Pascal family of GPUs shall support unified memory that shall allow the CPU and GPU access the memory of each other for faster processing. Developers shall also be benefited by this move as they won’t have to spend extra time allocating resources between the GPU and CPU. NVIDIA has also noted the fact that one of the major obstacles that lie in the path of faster computing is the speed at which the data is transmitted between the CPU and the GPU. It has introduced NVLink in Pascal that allows data flow speeds to upto 80GB per second. This speed is five times more than the existing speed of 16GB per second found in the present generation of GPUs.

Before you rush out to upgrade your gaming rigs we would like to point out that you have plenty of time decide because NVIDIA will not be releasing these GPUs to the public before 2016.

Source: #-Link-Snipped-#