What is DLSS that Nvidia Brags about?
Artificial Intelligence has revolutionized computer games with improved graphics and performance. We have seen a change from the rectangular aliased to realistic graphics with greater accuracy. DLSS is one of those technologies which have made this a reality. DLSS stands for Deep Learning Super Sampling which is an NVIDIA RTX technology. DLSS uses the power of AI to boost frame rates under heavy Graphics workload.
How is it better?
Before the introduction of AI into the gaming world, problems related to anti-aliasing, frame rates, resolution were to be handled individually perhaps needed better coding for individual issues. DLSS makes life easier as it uses AI to fix the aforementioned issue. DLSS makes use of aliased/rectangular frames and for each one, a matching “perfect frame” is generated using either super-sampling or accumulation rendering. This is how the DLSS model is trained by NVIDIA’s supercomputer. The greater the data to be trained on, the better is the model which would eventually result in a better performance. This process is repeated to generate additional pixels for better resolution.
If your game is already running with a high frame rate, DLSS neural network would not be enabled or available as this would not boost your frame rate. However, if your GPU is working on full capacity with a lower frame rate than 60 FPS, DLSS can come to its rescue and boost the frame rates. Since these were trained mostly on high-resolution data especially 4K, there are around 3.5-5 million pixels from which it can generate the final frame. These are powered by dedicated AI processors on NVIDIA RTX GPUs called Tensor Cores. The latest version is DLSS 2.0 which has proven itself in revolutionizing gaming from in-game and animation simulation to real-time rendering and AI-broadcasting features.