Nvidia’s supercomputer is constantly improving DLSS
See Full Size
The only purpose of the supercomputer is DLSS
VP of applied deep learning research at Nvidia Bryan CatanzaroIn his statement at CES, he stated that the development of DLSS technology was a six-year learning process. Catanzaro is a massive supercomputer equipped with Nvidia’s latest and most powerful GPUs. works 24/7, 365 days a year and this computer Its sole purpose is to improve DLSS is he emphasized.
The process in developing DLSS relies heavily on analyzing failures. When the DLSS model fails, this usually manifests itself in games such as blurring, juddering or ghosting. When such errors are detected, Nvidia engineers try to understand the cause of the error and examine the factors that caused the model to make the wrong choice. It is possible to improve training data by analyzing errors. The model is then retrained with this new data and tested on hundreds of games. This cycle continues constantly.
See Full Size
According to performance tests shared by Nvidia, the RTX 5090 is approximately 30 percent faster than the 4090 without using DLSS. Additionally, the RTX 5080 performs 15 percent faster than the 4080, and the RTX 5070 performs 20 percent faster than the 4070. However, since these data are based on Nvidia’s own test results, more objective results will be obtained through real-world testing.
This news our mobile application Download using
You can read it whenever you want (even offline):
