DLSS 4 FAQ

Q: What is DLSS?

A: NVIDIA DLSS is a suite of neural rendering technologies that multiplies frame rates up to 8x over traditional rendering by using AI to create entirely new frames, display higher resolution through image reconstruction, and improve the image quality of intensive ray-traced content — all while delivering best-in-class image quality and responsiveness.

Q: How do I get access to DLSS 4 for my custom engine?

A: DLSS Multi Frame Generation, Ray Reconstruction and Super Resolution are available to all game developers through the latest NVIDIA Streamline SDK.

Q: How do I get access to DLSS in Unreal Engine?

A: You can access the binary Unreal Engine DLSS plugin. You can also get source code via GitHub. Follow the instructions on the UE/GitHub page to link your Epic Games and GitHub accounts to gain access.

Q: What graphics cards are supported by DLSS?

A: DLSS Multi Frame Generation is available for GeForce RTX 50 Series GPUs. DLSS Frame Generation is available for GeForce RTX 40 Series and RTX 50 Series GPUs. DLSS Ray Reconstruction, DLSS Super Resolution, and DLAA are available for all GeForce RTX GPUs.

Q: Does NVIDIA offer an upscaling solution for non-RTX GPUs?

A: Yes, NVIDIA Image Scaling (SDK) is an open-source, best-in-class, spatial upscaler and sharpening algorithm that works cross-platform on all GPUs. NVIDIA Image Scaling SDK is a great compliment to NVIDIA DLSS for developers looking for a solution to support non-RTX GPUs.

Q: What is DLSS Transformer model?

A: Previously, DLSS Ray Reconstruction, Super Resolution, and DLAA used a Convolutional Neural Network model, also known as a CNN. A CNN processes pixel information through local operations, spatially around a small number of neighboring pixels, and temporally across multiple frames. The DLSS Transformer-based architecture improves image quality by using self attention to more easily identify the longer range patterns across a larger pixel window.

1 Like