Nvidia has presented at the SOLIDWORKS World event up to six new graphics cards, all with a professional destination, but the one that interests us most is the Quadro GP100. This unit presumes, among other things, to have 16GB of memory in HBM2 format. This is the most important model of this new batch, mainly for having a new GP100 chip that carries in its interior 3,840 CUDA cores and 240 cores in charge of working with textures. So, it has a monstrous power of 20 TFLOPs in 16 bits, half if the operations are in 32 bits. As the vice president of Professional Visualization at NVIDIA, Bob Pette said that “Professional workflows are now infused with artificial intelligence, virtual reality, and photorealism, creating new challenges for our most demanding users. Our new Quadro lineup provides the graphics and compute performance required to address these challenges. And, by unifying compute and design, the Quadro GP100 transforms the average desktop workstation with the power of a supercomputer”. Moreover, this card also has the latest superfast two-way NVLINK interconnect, which is an awesome, efficient feature that simply helps to get data in and out of the GPU much faster. Not only that even this NVLINK feature also allows for a pair of the GPUs to be combined for 32GB on tap in a single PC. So, let me remind you that we are talking about graphics cards which are actually designed to work at high performance in applications for video editing, artificial intelligence, or that require many ideal calculations for a graphic processing unit, such as virtual reality. So, basically, this is a tool for engineers and designers who work in Linux and Windows environments.
But, How much this monster is going to cost you?
As for now the pricing for the new Quadro cards has not yet been confirmed, but the availability is scheduled for March. So, if you want some more information regarding this cards then you can visit the NVIDIA’s official press release from here.
Δ
