/
The new creative powerhouse could cost as much as a new car
/cdn.vox-cdn.com/uploads/chorus_asset/file/24040901/nvidia_rtx_6000.jpg&w=2400&q=75)
The Nvidia RTX 6000 ADA professional GPU isn’t your run-of-the-mill gaming card.
Image: Nvidia
Nvidia has announced the RTX 6000 48GB graphics card, the latest model to join its family of workstation-focused GPUs designed for content creators and enterprise-grade graphics. Nvidia describes the RTX 6000 as the perfect tool for creating content for the metaverse alongside Nvidia Omniverse Enterprise, thanks to its Ada Lovelace generation AI, massively increased raytracing and CUDA cores, and programmable shader technology.
While the recently announced GeForce RTX 4090 is the top-end Ada Lovelace generation card for most consumers, the Lovelace-powered RTX 6000 dominates it in regards to raw power. It has 18,176 CUDA cores compared to the RTX 4090’s 16,384, and will ship with 48GB of GDDR6 ECC memory — double that of the RTX 4090’s 24GB GDDR6X. The RTX 4090 does have a higher TDP rating of up to 450W against the 300W used by the RTX 6000 48GB, however, this is likely intentional as Nvidia has a tendency to lower the power consumption of its workstation-grade graphics cards to extend their lifespan.
The RTX 6000 GPU can provide up to 2X – 4X the performance of the previous-generation RTX A6000 workstation card, with 568 Tensor cores and 142 RT cores, up from the 336 Tensor cores and 84 RT cores used by the RTX A6000. You can also expect 3x the video encoding performance of the previous generation, and virtualization to support Nvidia virtual GPU (vGPU) software, enabling multiple remote users to share resources and drive high-end design, AI and high-performance compute workloads.
You certainly won’t want to add this to your basket if you’re in the market for a new gaming GPU however as all that power comes at an eye-watering cost. We don’t have the official pricing for the Nvidia RTX 6000 at the time of writing, but previous workstation graphics cards such as the Ampere-powered RTX A6000 still retail for almost $5,000. Given the price increases across both the RTX 4090 and RTX 4080 compared to previous generation releases, it’s safe to assume the Nvidia RTX 6000 will also follow suit.
So, who uses Nvidia’s workstation GPUs? Broadcasters are one example, in need of Nvidia’s latest RTX 6000 to help power TV stations. “The new workstation GPUs are truly game changing, providing us with over 300 percent performance increases — allowing us to improve the quality of video and the value of our products,” says Andrew Cross, CEO of Grass Valley, maker of television and broadcasting equipment.
Other industries stand to gain from the specialized drivers and longevity within workstation cards. The RTX 6000 will likely be used within environments such as scientific computation workloads and creating CGI for the movie industry, and Nvidia’s mention of pairing the card with Omniverse (its own real-time graphics collaboration platform) is no coincidence as we steadily march towards Web 3.0.
“The NVIDIA RTX 6000 is ready to power this new era for engineers, designers and scientists to meet the need for demanding content-creation, rendering, AI and simulation workloads that are required to build worlds in the metaverse” said Bob Pette, vice president of professional visualization at NVIDIA.
The Nvidia RTX 6000 workstation GPU will be available from December via global distribution partners and manufacturers.