- The fastest GPU of Nvidia is, as expected, expensive and hungry for power, as well as its predecessor
- It is also extremely fast and leaves the RTX 4090 and the fastest GPU of AMD, well back
- The first reference points show that there is much more potential once drivers and updates are implemented
Nvidia’s last flagship GPU, the GEFORCE RTX 5090, was announced in CES 2025 and just went on sale, although Nvidia warned that she hopes to run out quickly. Built in the Blackwell architecture, the RTX 5090 is a successor of the RTX 4090, and presents 32 GB of VRM, greater CUDA nucleus count and a greater memory bandwidth.
Puget Systems, which previously tested the GEFORCE RTX 4090, has just compared the RTX 5090 in several creative applications, including Adobe Premiere Pro, After Effects and Davinci Resolve and, as expected, the results were impressive.
In Premiere Pro, Puget Systems found that RTX 5090 was slightly faster than RTX 4090 in approximately 9% in terms of general performance. In the Davinci Resolve studio of Blackmagic, the GPU had a 17% advantage over 4090 and 35% over 3090 TI. However, Puget pointed out that the “5090 was executed in a slightly different version of resolving that the rest of the cards: a review version designed to be fully compatible with the 50 series card and that we hope to join the application in the application in the Application in the application in the application in the application of the application in the application. “
Erasing the AMD
Adobe After Effects also benefits from the increase in the power of RTX 5090.
According to Puget Systems, the GPU recorded a “mass improvement of 35% generation generation in RTX 4090” in 3D representation tasks, so it is a solid option for movement graphics professionals. On the other hand, the reference points of a Unreal engine suggest that the RTX 5090 leads the RTX 4090 by 17% in general. In representation applications such as Blender and V-Ray, “the RTX 5090 is 38% massive faster than 4090 and three times faster than 3090 TI.”
The RTX 5090 surpassed the fastest AMD consumption GPU, RADEON RX 7900 XTX, in multiple tests, although AMD’s card faced some challenges. In Adobe After Effects, Puget Systems observed that the 7900 XTX “currently fights with the advanced ‘3D’ rendering, with the 7900 XTX half that even 2080 TI”. It was a similar story in Unreal Engine where “AMD struggles with the layout of rays, seeing a much greater performance drop than Nvidia enabling the function.”
Despite the solid reference points, there are some early software compatibility problems. Puget Systems said that “at present, the RTX 5090 is not compatible with red displacement (Cinebench) or the Octanebench” and has known performance problems in the CUDA representation of V-ray. Nvidia is expected to solve these problems with future controller updates.
With a launch price of $ 2,000, the RTX 5090 is obviously positioned as a premium option for users who need the highest level of performance. As Puget Systems concludes, “if you need the most powerful consumption GPU ever done, this is all.”
