Nvidia announces a new lineup of Pascal-based Quadro GPUs
Nvidia announces a new lineup of Pascal-based Quadro GPUs
 Â
Â
Please note that Nvidia’s GP100 supports NV-link, using a new SLI-finger like solution at the top of the GPU. The GP100 will support two-way NV-link support using two of these new connections, allowing workstations to use two GP100 GPUs for a combined 32GB of HBM2 memory.Â
Â
-
Unify simulation, HPC, rendering and design – The GP100 combines unprecedented double precision performance(2) with 16GB of high-bandwidth memory (HBM2) so users can conduct simulations during the design process and gather realistic multiphysics simulations faster than ever before. Customers can combine two GP100 GPUs with NVLink technology and scale to 32GB of HBM2 to create a massive visual computing solution on a single workstation.
Â
Nvidia’s new flagship Quadro GPU is the GP100, which is based on the GP100 GPU core and uses 16GB of HBM2 memory. When looking at specifications Nvidia’s GP100 and P6000 offer very similar specifications, with the GP100 offering fewer CUDA cores but much faster HBM2 memory.  On the other hand, the P6000 also offers more theoretical compute performance and more VRAM with a total of 24GB of GDDR5X memory.  Â
Below are the specifications of Nvidia’s new Pascal Quadro lineup,Â
Â
Nvidia Quadro Pascal GPUs | ||||||||
 | P400 | P600 | P1000 | P2000 | P4000 | P5000 | P6000 | GP100 |
GPU Core | 14nm GP107 | 14nm GP107 | 14nm GP107 | 16nm GP106 | 16nm GP104 | 16nm GP104 | 16nm GP102 | 16nm GP100 |
CUDA Cores | 256 | 384 | 640 | 1024 | 1792 | 2560 | 3840 | 3584 |
Boost Clock | 1170MHz | 1430MHz | 1400MHz | 1470MHz | 1480MHz | 1733MHz | 1530MHz | >1395MHz |
FP32 Compute | 0.6 TFLOPs | 1.1 TFLOPs | 1.8 TFLOPs | 3.0 TFLOPs | 5.3 TFLOPs | 9.0 TFLOPs | 12.0 TFLOPs | 10 TFLOPs |
Memory | 2GB GDDR5 | 2GB GDDR5 | 4GB GDDR5 | 5GB GDDR5 | 8GB GDDR5 | 16GB GDDR5X | 24GB GDDR5X | 16GB HBM2 |
Memory Bus | 128-bit | 128-bit | 128-bit | 160-bit | 256-bit | 256-bit | 384-bit | 4096-bit |
Power | None | None | None | None | 1x 6-pin | 1x 8-pin | 1x 8-pin | 1x 8-pin |
TDP | 30W | 40W | 47W | 75W | 105W | 180W | 250W | 235W |
Launch Date | March 2017 | October 2016 | March 2017 |
Â
 Â
 You can join the discussion on Nvidia’s Pascal-based Quadro GPUs on the OC3D Forums.Â
Â