NVIDIA QUADRO GP100 GPU

Description

Powered by the NVIDIA Pascal architecture, the Quadro GP100 provides a wide range of features including NVIDIA NVLink. With NVLink, users can connect two GP100 cards to scale performance to 32GB HBM2 of combined memory. Whether you’re designing innovative products or sharing vibrant visual stories, Nvidia Quadro gives users the ability to do it efficiently.

Memory

The GP100 is outfitted with 16GB of second-generation high bandwidth memory (HBM2) with Error Correcting Code (ECC) for maximum accuracy and reliability. It’s powered by a 4096-bit memory interface running at 715MHz. Users can tap into the power of NVLink to double the amount of memory capacity to 32GB of HBM2 memory, while reducing latency. The dual-slot GPU operates at a frequency of 1304MHz, which can be increased up to 1442MHz. This GPU is a great choice for designers, analysts, and any professional needing crucial graphics processing.

Performance

NVIDIA’s Quadro GP100 graphics card has excellent performance to render photorealistic design concepts, create detailed 3D models, and run immersive VR. The 15,300 million transistors support 3584 NVIDIA CUDA cores, 224 texture mapping units, and 96 ROPs. It also provides 20.67 TFLOPS of FP16 (half) performance, 10.34 TFLOPS of FP32 (float) performance, and 5.17 TFLOPS of FP64 (double) performance. These features make it an ideal workstation solution for professionals working on the latest deep learning and artificial intelligence algorithms.

Features

At the heart of the GP100 GPU lies a graphics processor, which supports DirectX 12. The card also introduces a 2-way NVLink connector for PCI-Express-based cards. This allows double the memory access and low latency data transfer benefits. Like the rest of the Quadro lineup, this GPU has a traditional, fan and shroud active cooler. Additionally, the Quadro card is virtual reality ready and can also run simulations while designing and getting realistic with multi-physics.

Summary

The NVIDIA Quadro GP100 GPU allows professionals to build large-scale visual computing solutions on a single workstation. The 16GB HBM2 memory and 3584 CUDA cores deliver significant power to create, build, and solve the impossible. This card reinvents systems to meet the demands of interactive designs, detailed 3D models, high-end computer-aided design (CAD), computer-aided engineering (CAE), and immersive VR.