NVIDIA Unveils AI-Powered DLSS 5 Claiming 'Photorealism' Coming This Year
NVIDIA unveiled the next iteration of its DLSS (Deep Learning Super Sampling) technology with DLSS 5, claiming to achieve "photorealism" via a new real-time neural rendering model. What do you think of the results?
NVIDIA has unveiled DLSS 5, positioning the technology as a "new level of photoreal computer graphics" in its official statements.
The announcement builds on the company’s existing DLSS ecosystem by introducing new features in neural rendering, lighting reconstruction, and frame generation aimed at pushing visual fidelity beyond traditional rasterized pipelines. Just last week, more features were revealed for DLSS 4.5.
According to NVIDIA, DLSS 5 expands the role of AI in the rendering process even further, allowing GPUs to generate a significant portion of the final image through neural inference rather than conventional rendering. The company says this approach could enable more realistic lighting and higher frame rates simultaneously, particularly when paired with GeForce RTX 50-series GPUs and their latest Tensor Core hardware.
However, while NVIDIA claims DLSS 5 as a major breakthrough for graphics fidelity, early analysis from outlets like Digital Foundry suggests the technology represents a continuation of an ongoing shift toward AI-assisted rendering rather than pure native rendering.
According to Digital Foundry, the way DLSS 5 handles environments and materials is particularly notable:
"Handling of materials can be astonishing, with everything from metals, cloth and even the skin of fruits looking remarkably realistic. Particularly impressive is how DLSS 5 handles light and shadow around foliage - something that's very difficult for standard renderers to achieve, even with RT or path tracing. Right now, DLSS 5 is still a work-in-progress project - we did spot some screen-space errors, but Nvidia describes what we're seeing today as a "snapshot" of the technology as it stands, with further improvements and optimisations to come. It's set to be launched later in 2026 after three years of development at Nvidia."
- Richard Leadbetter, Founder, Digital Foundry
However, the results for character models (specifically faces) have much of the gaming community divided. In the examples shown in games like Resident Evil: Requiem, Grace and Leon's faces are altered significantly, and in games like EA Sports FC, it seems to alter the faces of actual athletes as well, quite dramatically.
It remains to be seen how much of this is automatically generated by DLSS 5 itself, or how much is guided by the game's developers in their integration.
The main concern from early reactions stems from the concern that the changes enacted by DLSS 5 are too dramatic, resulting in something that no longer resembles the intent of the game's creators. Rather than enhancing a character model to achieve a higher resolution or improved clarity, the art style, lighting, and textures of the game are now different.
80 Level has contacted NVIDIA and reached out to game developers to inquire about this very topic. According to NVIDIA in their announcement, buy-in and support are widespread:
“At CAPCOM, we strive to create experiences that feel cinematic, compelling and deeply believable — where every shadow, texture and ray of light is crafted with intention to enhance atmosphere and emotional impact. DLSS 5 represents another important step in pushing visual fidelity forward, helping players become even more immersed in the world of Resident Evil.”
- Jun Takeuchi, executive producer and executive corporate officer at CAPCOM
According to Digital Foundry, it's not quite the same as typical generative AI:
"The AI network powering DLSS is aware of the semantics of the scenes it processes. It "recognises" and processes elements like skin, hair, water and metal differently to apply photo-realistic lighting effects. While there could be some comparisons to generative AI, DLSS 5 is consistent and coherent in its rendering of the game world, the environments and the characters within it. It's capable of working with standard rasterised games, RT supported titles and path-traced experiences - the higher fidelity you give to the model, the better the end results you'll see from it in terms of material response, lighting and shading."
- Richard Leadbetter, Founder, Digital Foundry
DLSS began as an AI upscaling technique designed to render games at a lower resolution before reconstructing them using machine learning. Over time, the system expanded to include features like frame generation and ray reconstruction, allowing the GPU to synthesize new frames and improve ray-traced lighting quality.
With DLSS 5, NVIDIA appears to be taking that concept further. The company describes the system as using advanced neural models to help generate lighting information and frame data, enabling more complex scenes without the same traditional rendering cost.
Also, subscribe to our Newsletter, join our 80 Level Talent platform and our new Discord server, follow us on Instagram, Twitter, LinkedIn, Telegram, TikTok, and Threads, where we share breakdowns, the latest news, awesome artworks, and more.
