Nvidia recently revealed that it has maintained a dedicated supercomputer for six years, designed specifically to enhance the quality of its Deep Learning Super Sampling (DLSS) technology. This acknowledgment came during the CES 2025 event held in Las Vegas.

At RTX Blackwell Editor’s Day, Brian Catanzaro, NVIDIA’s VP of Applied Deep Learning Research, disclosed how this supercomputer utilizes thousands of cutting-edge GPUs, operating continually to refine the DLSS algorithms.

“How is it that we’ve been able to make progress with DLSS over the years? It’s been a six-year, continuous learning process for us,” Catanzaro explained.

He emphasized the importance of this supercomputer in augmenting training data sets for the DLSS improvement process, ensuring ongoing advancements to address graphical glitches like ghosting and blurriness.

In summary, Nvidia’s commitment to evolving its DLSS technology through substantial resource investment over the years highlights its dedication to quality in gaming graphics.