,

DLSS

Listen to this glossary

DLSS or Deep Learning Super Sampling is a rendering technology developed by NVIDIA that uses artificial intelligence to improve the graphic performance of games and applications, while reducing the amount of computational resources required. It applies machine learning to produce sharper images by examining multiple frames of a game and then using this information to predict what following frames will look like. This technique leads to higher frames per second (fps) and better image quality in games, while keeping the graphics processing unit’s (GPU's) workload at bay.

In the scope of gaming, frame rate and image quality are vital for the best gaming experience. However, increasing the image resolution and graphical settings to improve the quality often requires a significant computational load which most GPUs cannot handle, lowering the frame rates drastically. DLSS mitigates this issue by processing games at a lower resolution internally and then up-scaling the final image to a higher resolution using deep learning, which can efficiently handle this task. Thus, it delivers smooth gaming performance even with setting the graphic details to maximum.

DLSS utilizes a pre-trained convolutional auto-encoder network from millions of high-resolution images in supercomputers at NVIDIA's HQ. This extensive training prepares it to predict high-resolution images, which are similar to the low-resolution ones, without having any prior information about the game's environment. The final output gives out a synthetic image of superior quality than the traditional rendering solutions like Temporal Anti-Aliasing (TAA) while sacrificing less performance.

DLSS comes in several versions, each developed to address previous edition’s weaknesses and improve overall performance. The most recent version, DLSS 2.0, claims improved image quality, better graphics, and more efficient resource usage than its predecessor. This version offers three image quality modes: Quality, Balanced, or Performance mode, giving users the option to prioritize either image quality or performance according to their preference.

In conclusion, DLSS is a groundbreaking technology that uses Deep Learning and AI to enhance the gaming experiences by increasing the frame rates while improving the image quality simultaneously. And as Deep Learning continues to evolve, it is likely that this technology will continue to refine, offering even higher-quality visual experiences and more efficient hardware usage in the gaming industry in the future.