2023-06-27 03:20:27
© Archyde.com. Nvidia (NVDA.US) releases new gaming graphics card DLSS, which is expected to improve RTX 4060 game performance. Zhitong Finance APP learned that Nvidia (NVDA.US) released the GeForce RTX 4060 graphics card this week. How an investment in intelligence paid off for its core business, video games. Nvidia says the new line of processors will bring huge performance boosts to games like Cyberpunk 2077 and Microsoft’s Flight Simulator, thanks to an artificial intelligence called Deep Learning Super Sampling (DLSS) Function. With all the hype regarding how Nvidia’s chips are powering ChatGPT and other consumer-grade AI applications, it’s easy to forget that gaming is still a big part of the company’s sales, generating billions of dollars in revenue each quarter. According to CEO Jensen Huang, DLSS will revolutionize the way computer graphics are handled. “DLSS is essentially predicting missing pixels,” he compared the technique to how AI language models choose the next word in a sentence: “We should be able to predict the next pixel.” It’s a complex concept, But DLSS basically relies on artificial intelligence to generate frames in video games rather than the traditional way, which is more processor-intensive. When DLSS was first launched in 2018, it was widely criticized. In the first-person shooter Battlefield V, reviewers found blurry textures and negligible performance improvements. Part of the problem is that engineers must develop custom AI models, a complex and expensive endeavor. “We can’t have specific models like, ‘Call of Duty is like this, Doom is like that,’” says Bryan Catanzaro, Nvidia’s vice president of applied deep learning research. “It takes thousands of people to collect the training data, And model each game individually.” The team realized that Nvidia needed to develop a single common model that would work across a variety of games while allowing each game to retain its unique aesthetic. They spent two years writing the second generation of DLSS. While the first version used a neural network to learn which pixels to display next, DLSS 2 more efficiently guesses which pixels to keep from the previous frame. While it’s still a long way from perfect, and critics have complained regarding poor visuals under heavy rendering workloads, many big games have adopted the latest version of DLSS. Hopefully it will give a major frame rate boost to low-end processors like the GeForce RTX 4060. If optimized properly, this approach can be vastly more efficient, Catanzaro says, “and seven out of every eight pixels are generated by these models.”
1687853348
#Nvidia #NVDA.US #released #game #graphics #card #DLSS #expected #improve #RTX #game #performance #author #Zhitong #Finance
