At the GPU Technology Conference 2022 (GTC), Nvidia also revamped its workstation GPU portfolio alongside a lot of tech prospects and a monstrous AI accelerator card (Hopper / H100). Slightly below the top model RTX A6000, there is now an RTX A5500, which is roughly equivalent to an RTX 3080 Ti. The name Quadro for workstation models is no longer available - instead, a simple A now marks the GPUs from this series.
Compared to the RTX 3080 Ti, the RTX A5500 has twice as much memory (24 instead of 12 GB), which even works with error correction ECC, but is clocked slower in return. Thus, the A5500 "only" achieves 768 GB/s memory throughput, while the RTX 3080 Ti manages 912 GB/s. Nvidia has not yet announced a price for the A5500, but the larger A6000 (with 48 GB ECC RAM) costs around 5,000 Euros.
The new Nvidia RTX A5500 comes with 24 GB ECC memory.
At the same time, Nvidia has also announced numerous mobile workstation GPUs that should be found in various laptops soon. These are in detail the RTX A500, RTX A1000, RTX A2000 8GB, RTX A3000 12GB, RTX A4500 and the RTX A5500. These GPUs are also very similar to their "gaming" siblings.
They all feature 2nd generation RT cores, 3rd generation Tensor cores, and Ampere architecture streaming multiprocessors, and are comparable to NVIDIA's Max-Q models. The RTX A2000 with 8 GB, the RTX A3000 with 12 GB and the RTX A4500 now have double the memory capacity compared to their predecessors. The following table from Nvidia provides a complete overview of the current workstation lineup:
Current Nvidia Workstation GPUs for Desktop and Mobile
However, a new GPU generation is already expected from Nvidia in the fall (RTX 40xx), which is why the refresh of the current workstation series is a bit more opulent than expected. Furthermore, the RTX gaming models have the better price/performance ratio for video editing, especially since the memory usually works a bit faster here. And special studio drivers are also available for the RTX Gaming series.
;)
;)