Nvidia has announced a new workstation-grade graphics card to replace the previous-generation RTX A2000 12GB. The company promises up to 50% performance uplift in a low-power GPU made for compact workstations for creative professionals, AI acceleration, edge applications, and more.
The RTX 2000 delivers breakthroughs in speed, efficiency, and power for everyday workflows, enabling creators, designers, and engineers to achieve new levels of productivity and innovation from the desktop.
As you can tell by the graphics card"s name, the RTX 2000 Ada Generation is powered by the same architecture as the latest consumer-grade GPUs in the RTX 4000 Series. It features third-generation RT cores, fourth-generation Tensor cores, and 16GB of GDDR6 video memory. Interestingly, the card does not feature the latest DisplayPort versions. Instead, there are four mini DisplayPort connectors for display output (four 2160p at 120Hz, four 2880p at 60Hz, or two 4320p at 60Hz).
According to information published by Nvidia, the RTX 2000 Ada Generation outperforms its predecessor by 30% in graphics, 50% in rendering, and 50% in generative AI tasks.
Here is the complete spec sheet:
RTX 2000 Ada Generation | |
---|---|
Architecture | Ada Lovelace |
Form-factor | Dual-slot GPU PCIe 4.0 x8 with active cooling |
Memory | 16GB GDDR6, EEC with 128-bit bus operating at 224GB/s |
CUDA Cores | 2,816 cores |
Tensor Cores | 88 cores, 191.9 TFLOPS |
RT Cores | 22 cores, 27.7 TFLOPS |
Single-precision Performance | 12 TFLOPS |
Total Board Power | 70W |
Ports | 4x mini DisplayPort 1.4a |
Display Configurations | 4x 4096x2160, 120Hz 4x 5120x2880, 60Hz 2x 7680x4320, 60Hz |
Media Engine | AV1 encode and decode support |
The Nvidia RTX 2000 Ada Generation graphics card is available now at Nvidia"s global distribution partners, such as Arrow Electronics, Ingram Micro, Leadtek, PNY, Ryoyo Electro, and TD SYNNEX. It will also be available from Dell, HP, and Lenovo starting April 2024. The MSRP for the GPU is $625.