Nvidia RTX A6000 (Ampere) professional graphics card introduced

0
290

It does not have additional PCIe power connectors

Nvidia has unveiled the RTX A6000 professional graphics card, the first product in the professional rendering segment based on the Ampere graphics architecture. At the same time, the company moved away from using the Quadro brand in relation to the video card as a whole, leaving it only in the names of some software functions.

Nvidia RTX A6000
Nvidia RTX A6000

Nvidia RTX A6000 (Ampere) professional graphics card introduced

The card is based on the same 8nm GA102 die as the GeForce RTX 3080 but configured differently. The memory subsystem is also different: there is support for ECC, and the amount of GDDR6 memory is 48 GB. The memory is connected to the GPU by a 384-bit bus. The manufacturer did not disclose the number of CUDA cores. The typical power consumption of the card is 300W. It is noteworthy that the board does not have additional PCIe power connectors. An EPS connector (4 + 4 pins) is used instead. It can supply up to 336 W (4 x 7 A at 12 V).

NVLink support allows up to two A6000 cards to be combined. There is also support for GPU virtualization, including Nvidia GRID, Nvidia Quadro Virtual Data Center Workstation, and Nvidia Virtual Compute Server.

The accelerator A40 is also presented. It is similar in performance to the RTX A6000 but lacks video outputs.

Also Read:  NVIDIA Introduces DLSS 2.1: Supports 8K, VR And Dynamic Resolution