NVIDIA launches RTX A4500 20GB and A2000 12GB workstation graphics cards

Both new card models are targeted at a specific set of workloads, and feature the latest Ampere architecture to power content creation, AI, and other intensive capacitors more efficiently.

As such, NVIDIA's RTX Ampere Workstation dedicated workstation GPU lineup currently includes 6 options across different price segments and workflow capabilities. The two new models will be positioned in the high-end and affordable segments. Specific information is as follows.

NVIDIA RTX A4500 20GB Specifications

Get started with the NVIDIA RTX A4500. This is a high-end touch GPU model with 7168 CUDA cores out of a total of 56 SMs enabled on the GA102 GPU. Other specs include 224 Tensor units, 56 RT cores, and core clocks up to 1.63GHz. The card owns 20GB of GDDR6 memory operating on a 320-bit bus interface, with speeds up to 16Gbps, and the ability to provide bandwidth up to 640GB/s for the GPU. The TDP of the RTX A4500 stays steady at 200W.

In terms of performance, the NVIDIA RTX A4500 offers up to 23.7TFLOPs FP32, 46.2TFLOP RT and 182.2TFLOPs of Tensor performance. The card has four DisplayPort 1.4 output ports and supports dual-slot configurations with active cooling. Power is provided by a single 8-pin connector. NVIDIA also equips an NVLINK connection on the A4500, allowing up to 2 cards to run in parallel with the combined memory for a total capacity of 40GB.

NVIDIA RTX A2000 12GB Specifications

The 12GB A2000 is basically an upgraded variant of the earlier RTX A2000 graphics card model, with double the memory capacity. While the rest of the specifications are generally similar to the original version.

Specifically, NVIDIA RTX A2000 12GB also owns Ampere GPU architecture. In terms of specifications, the card using the GA106 GPU is equipped with 3328 CUDA cores, 104 Tensor cores, and 26 RT cores, all of which deliver better performance than previous generation products. GPU performance is rated at 8TFLOPs FP32, 15.6TFLOPs RT and 63.9TFLOPs Tensor. The RTX A2000's 12GB memory features a 192-bit bus design with chips operating at 12Mbps, for a total effective bandwidth of 288GB/s.

Appearance is also a highlight of the RTX A2000 12GB with its small frame design, for more flexible compatibility with many different hardware settings, especially in compact low-cost PCs. Since the TDP level only stops at 70W, there will be no power connector included. The RTX A2000 is a simple plug-and-use card that delivers great performance in a compact design. There are 4 Mini DisplayPorts (1.4) near the rear panel IO shroud, which comes with a small vent to let hot air out.

The official selling price of the two products has not yet been revealed by NVIDIA.

Isabella Humphrey

Update 25 November 2021