- NVIDIA RTX PRO 4000 SFF offers the performance of the AI of its predecessor more than twice
- Nvidia says that RTX Pro 2000 exceeds RTX A2000 in 3D modeling performance
- The two new GPUs maintain a limit of 70 watts, allowing deployment in the limited systems in space
NVIDIA launched two new RTX Pro SFF (Small Form Factor) and GPUS of Workstation built on its Blackwell architecture that uses a single fan, blowing and low profile.
The new RTX Pro 4000 SFF and RTX Pro 2000 slit in the high -end RTX Pro 6000 high -end, but are designed to maintain solid performance for professional workloads in smaller and lower workstations.
The two GPUs aim to move more processor processor to the GPU, even allowing the small factor work stations of small form to manage the inference, rendering and simulation of AI faster than before.
Compact equipment with capacity at the workstation
The RTX Pro 4000 SFF is built with 24 GB of ECC GDDR7 memory and provides up to 770 AI peaks.
NVIDIA says that it offers more than double the performance of the IA of the SFF RTX A4000 of previous generation, and also has an improved tracing of shelves, 50% more bandwidth of memory in addition, and has a thermal limit of 70 W, which makes it suitable for workstations with a limited cooling capacity.
The RTX PRO 2000, also limited to 70W, bears 16 GB of ECC GDDR7 memory and 545 AI of flow.
Nvidia says it is about 1.5 times faster than RTX A2000 in 3D modeling, CAD and rendering.
The company also underlines its usefulness in the generation of images and texts fueled by AI, potentially accelerating the design and workflows of content in small studios or engineering teams.
These workstation GPUs share the advantages of Blackwell architecture found in the RTX Pro larger range of NVIDIA, including the FP4 precision support through fifth generation tensor nuclei and the second generation transformer engine.
This is intended to improve the performance of inference while maintaining energy consumption.
Nvidia has not yet disclosed specific prices for these models, but said that it will make them available by partners such as PNY, TD Synnex, Dell, HP and Lenovo later this year.
Given their position below the RTX Pro 6000, they should cost considerably cheaper while providing a significant increase in performance compared to the GPUs of workstation in a previous small format.
Although it is marketed as an upgrade path for professionals, the real advantage on the fastest workstation configurations of the CPU will depend on the type of workload and the optimization of software for the acceleration of the GPU.
For tasks that already benefit from the calculation of the GPU, such as IA inference, large -scale rendering and simulation, the new RTX Pro 4000 SFF and RTX Pro 2000 could offer strong gains without requiring a larger and more expensive workstation.
For processor -oriented applications, improvements can be less dramatic, but the added GPU capacity always widens the performance envelope for small limited power systems.
