Nvidia RTX A6000 and Nvidia A40: Specifications and Features
The new GPUs bring heavy-duty computing power that the company says will accelerate rendering, AI, VR/AR and compute workloads for artists, designers and engineers. Built on the NVIDIA Ampere architecture, they feature new RT Cores, Tensor Cores and CUDA cores. Image Courtesy: Nvidia According to Nvidia, the second-gen RT cores offer up to 2x the throughput of the previous generation, plus concurrent ray tracing, shading and compute. The third-gen Tensor cores, meanwhile, provide up to 5x the throughput of the previous generation, up to 10X with sparsity, with support for new TF32 and BF16 data formats. As for the new CUDA cores, they deliver up to 2x the FP32 throughput of the previous generation for significant increases in graphics and compute. In terms of hard numbers, both cards come with Nvidia’s Ampere architecture and pack 10,572 CUDA cores alongside 48GB of VRAM. Other specs are relatively similar as well, including 336 tensor cores, 384-bit memory bus width and 300W TDP. However, while the former is meant for standard workstations, the latter is intended for use in commercial data-centers and server farms. Also, while both come with GDDR6 RAM, the A600 is clocked at 16Gbps, while the A40 comes with a memory clock of 14.5Gbps. The cards will only use DisplayPort for video output and will connect to computers through PCIe Gen 4.
Nvidia RTX A6000 and Nvidia A40: Price and Availability
Nvidia is yet to announce either an exact release date or prices for its new GPUs, but says that the RTX A6000 will be available from a number of channel partners, including PNY, Leadtek, Ingram Micro, Ryoyo and on nvidia.com starting mid-December. They will also be available from leading systems manufacturers, like Dell, HP, Lenovo and Boxx, from next year. As for the A40, it will also be available from OEM workstation and server vendors, including Cisco, Dell, Fujitsu, Hewlett Packard Enterprise and Lenovo, starting early next year. Featured Image Courtesy: Nvidia
