For the first time in years, Nvidia skipped unveiling new GeForce graphics card models at CES. CEO Jensen Huang’s characteristically sprawling and loosely rehearsed 90-minute keynote concentrated almost entirely on the company’s dominant AI business, while the company’s gaming-related news was pushed to a separate video released later that night.
Rather than new hardware, Nvidia emphasized software upgrades for its current products. The standout is DLSS 4.5, which brings several new additions to Nvidia’s collection of upscaling and frame-generation technologies.
DLSS upscaling is upgraded with a new “second-generation transformer model” that Nvidia says was “trained on an expanded data set” to better predict pixels when generating detail. According to Nvidia’s Bryan Catanzaro, this particularly improves image quality in Performance and Ultra Performance modes, where the upscaler must infer more from a lower-resolution source.
DLSS Multi-Frame Generation is also enhanced, increasing AI-generated frames per rendered frame from three to five. This new 6x mode for DLSS MFG is paired with Dynamic Multi-Frame Generation, which adjusts the number of AI-generated frames dynamically—raising them in “demanding scenes” and lowering them in simpler scenes “so it only computes what’s needed.”
The usual caveats for Multi-Frame Generation still apply: it requires an RTX 50-series GPU (40-series cards can still only generate one AI frame per rendered frame, and older cards can’t generate extra frames at all), and the game needs to run at a reasonably high base frame rate to minimize latency and visual artifacts. It’s a useful way to make already-fast games run faster, but it won’t convert an unplayable frame rate into a playable one.
