Nvidia unveils the RTX A6000 and A40

Nvidia has unveiled the RTX A6000 and A40 , new professional graphics cards for workstations and servers based on the Ampere architecture. Both solutions feature a GA102 GPU with 48GB of GDDR6 . These are professional solutions designed for engineers, designers, and scientists—offerings that, until the last generation of Nvidia cards, would have been called Quadro, but Nvidia has dropped that branding.

Technical specifications

The new Nvidia RTX A6000 will be available from partners PNY, Leadtek, Ingram Micro, Ryoyo, and on nvidia.com around mid-December, and will then arrive on workstations from BOXX, Dell, HP, and Lenovo. In terms of technical specifications, the A6000 replaces the Turing Quadro RTX 8000/6000 solutions, as it combines a GA102 GPU (the same as the GeForce RTX 3090 and 3080 graphics cards) with 10,752 CUDA cores, 336 Tensor cores, and 84 RT cores. It comes with 48 GB of GDDR6 memory at 16 Gbps on a 384-bit bus .

Nvidia RTX A6000 Specifications

According to Nvidia, the new RTX A6000 offers roughly double the performance of the Quadro RTX 8000 in certain situations, particularly in operations utilizing more CUDA and RT cores, as well as higher processing throughput. Currently, the American company has not disclosed the clock speeds and processing power (in FPS) of the new card, but it is rumored to have a TGP of 300W (50W less than the RTX 3090). The GPU is therefore expected to operate at lower frequencies.

The new RTX A6000 naturally benefits from all the new features of the Ampere architecture, such as a greater number of data types supported by the Tensor Cores (including BFloat16), as well as the new decoder for the AV1 codec and the PCI Express 4.0 interface . There is also an NVLink connector for connecting two A6000s in parallel, and partial ECC support via software. The RTX A6000 utilizes 3D Vision Pro and is equipped with four DisplayPort 1.4 ports.

Nvidia A40 Specifications

As for the Nvidia A40 , it's a very similar solution to the RTX A6000, except that it's passive. The target customer is different; it's designed for high-density servers. The other difference compared to the A6000 concerns the memory, which is set at 14.5 Gbps , while the TGP remains at 300W. It features video outputs: a new feature that allows the card to drive a monitor.

Nvidia justified the inclusion of video ports on these cards, which were generally absent on previous versions, by citing requests from users in the multimedia and broadcasting industries. The Nvidia A40 will be available starting next year on servers from Cisco, Dell, Fujitsu, Hewlett Packard Enterprise, and Lenovo. These cards connect via an 8-pin EPS connector, not a standard PCIe connector.