NVIDIA launched RTX A2000 professional graphics card
After the Ampere architecture, NVIDIA began to play down the Quadro brand in professional card products and changed to RTX A series. The graphics cards previously launched were for high-end large-scale workstations, such as RTX A6000 and A5000. The performance and size are big enough, but after the big guys are out, NVIDIA suddenly came up with a little product: RTX A2000
In fact, NVIDIA’s first-level professional cards were all single-slot or half-height cards. After all, they are used in professional application tasks. GPU heat is far less than playing games, the heat sink requirements are also lower, and this time RTX A2000 looks very mini. The GPU has 3384 CUDA cores, and the computing power is 8TFLOPS. This seems to be closer to the RTX 3060 desktop version (3584 CUDA), and it is probably adjusted from the GA106 core. In addition, the video memory of RTX A2000 is only 6GB GDDR6, but it supports ECC error correction.
“The A2000 from NVIDIA has made our modeling flow faster and more efficient. No longer are we sitting and wasting valuable time for graphics to render, and panning around complex geometry has become smoother,” said Connor Reddington, mechanical engineer and certified SOLIDWORKS professional at Avid Product Development, a Lubrizol Company.
The NVIDIA RTX A2000 features the latest technologies in the NVIDIA Ampere architecture:
- Second-Generation RT Cores: Real-time ray tracing for all professional workflows. Up to 5x the rendering performance from the previous generation with RTX on.
- Third-Generation Tensor Cores: Available in the GPU architecture to enable AI-augmented tools and applications.
- CUDA Cores: Up to 2x the FP32 throughput of the previous generation for significant increases in graphics and compute workloads.
- Up to 6GB of GPU Memory: Supports ECC memory, the first time that NVIDIA has enabled ECC memory in its 2000 series GPUs, for error-free computing.
- PCIe Gen 4: Double the throughput with more than 40 percent bandwidth improvement from the previous generation for accelerating data paths in and out of the GPU.