If the GeForce RTX 40 experienced a rather cautious initial reception from consumers, it is largely because of pricing considered by many to be excessive. The reason for this is the increase in production costs, but also the need to sell off the significant remainder from the previous generation, due to the explosion of the bubble linked to the mining of cryptocurrencies and the end of periods of confinement. However, the differential in performance and efficiency was very significant, thanks to significant progress in the manufacturing process, going from the Samsung N8 (10 nm engraving node) to the TSMC N4 (5 nm engraving node) and the Introduction to image generation. The RTX 50 reuse exactly the same manufacturing process for their GPU and must therefore rely on other factors to differentiate themselves. Chief among them, DLSS 4 and its Multi Frame Generationillustration of what NVIDIA calls Neural Rendering. The first card to see the embargo on its services lifted is the GeForce RTX 5090, the flagship of the range. What performance can we expect? What is DLSS 4 worth? Answers to these questions and many others within this file.
2025 © 
