Editor’s Pick: Next-Gen Graphics Processing

The next generation of NVIDIA high-end graphics processing is the Ada generation. The flagship of this generation for professional workflows is the NVIDIA RTX 6000 Ada, now available from PNY. 

PNY says the NVIDIA RTX 6000 Ada is designed for high-end professional workflows in any environment where professionals need maximum performance and reliability. These fields include high-end design, real-time rendering, artificial intelligence research and delivery, and other high-performance compute workflows. 

The NVIDIA RTX 6000 Ada features 142 third-generation RT Cores, 568 fourth-generation Tensor Cores, and 18,176 CUDA cores. These specialized processors are matched with 48GB of error correction code (ECC) graphics memory. PNY says these top-of-the-line components deliver “the next generation of AI graphics and petaflop inferencing performance for unprecedented speed-up in rendering, Al, graphics, and compute workloads.”

PNY says this new RTX-class GPU is certified with a broad range of professional applications, tested by leading independent software vendors (ISVs) and workstation manufacturers. 

PNY says the 76 billion total transistors are built using NVIDIA’s 4nm custom process by TSMC. The graphics bus is the PCI Express Gen 4.0 x16. 

The GPU uses DisplayPort 1.4 technology to run up to four 5K monitors at 60Hz, or dual 8K displays at 60Hz. Each DisplayPort connector can drive ultra-high resolutions of 4096x2160 at 120Hz with 30-bit color. 

The Ada Lovelace generation of CUDA cores are rated to deliver more than 2x the single-precision floating point throughput (FP32), significant for many engineering workloads. The NVIDIA RTX 6000 Ada further extends the throughput by enabling two FP32 primary data paths, doubling the peak FP32 operations. 

PNY says the use of the latest APIs including NVIDIA OptiX, Microsoft DXR, and Vulkan ray tracing allows “truly interactive design workflows to provide immediate feedback.” This GPU is rated as offering 2x faster ray-triangle intersection throughput compared to the previous generation. 

The unit uses GDDR6 memory, which is rated as delivering 25% faster throughput than GDDR5. PNY says this is especially important for the largest datasets and models used in latency-sensitive applications. 

For video encoding and decoding, PNY says this GPU uses 5th Generation NVDEC and 8th Generation NVENC engines. The NVIDIA RTX 6000 Ada supports 8K video encoding. 

Other key specs: 

• Memory bandwidth: 960 GB/sec.
• Maximum power consumption: 300W
• Form factor: 4.4-in. height x 10.5-in. length dual slot 

A complete listing of features and specifications are available here. 

The NVIDIA RTX 6000 Ada is now available from PNY. 

Sources: Press materials received from the company and additional information gleaned from the company’s website.

More PNY Technologies Coverage

Building the GPU-Accelerated Data Center

Attend this webinar with Brian Corn, Vice President of Product Integration at Thinkmate, and learn more about GPU-accelerated technologies.