Nvidia used CES 2025 to share its vision for the future of gaming laptop graphics: the GeForce RTX 50 series Blackwell graphics processors (GPUs). This new generation of mobile GPU rings in DLSS 4 super-scaling, which can reportedly predict and generate additional frames using machine learning.
Several of Nvidia’s partners, including Acer, Asus, Dell, Lenovo, MSI, and others, joined the firm in revealing their own mobile gaming PCs with Nvidia’s 50-series silicon inside.
Naturally, artificial intelligence (AI) plays a significant role in Nvidia’s plans for attacking the rapidly increasing demands of 3D gaming applications. Nvidia’s new Blackwell architecture, which is capable of driving thousands of AI TOPS (or trillions of operations per second), enables all of its new supporting technologies.
“Neural rendering is the future of computer graphics,” Nvidia CEO Jensen Huang said on stage during his CES 2025 keynote address.
These new GeForce RTX 50-series Blackwell laptops will be available in March for $1,299, but Nvidia has several more potent (and more expensive) options lined up.
Nvidia’s Current Stack of RTX Blackwell Laptops
For this March 2025 launch, Nvidia has four different tiers of RTX 50-series graphics silicon available: the GeForce RTX 5070, the RTX 5070 Ti, the RTX 5080, and the RTX 5090. Nvidia promises that the RTX 5070 drives RTX 4090 levels of mobile graphics performance at half the power draw, which is quite a claim.
(Credit: Nvidia)
According to Nvidia, RTX 5070 laptops will start at $1,299, with the base pricing increasing to $1,599 for RTX 5070 Ti, $2,199 for RTX 5080, and $2,899 for RTX 5090. So, clearly, the pricing can get out of hand rather quickly.
What RTX Blackwell Does For Your PC Gaming and Creation
Nvidia’s RTX Blackwell architecture packs fifth-generation machine-learning Tensor Cores and fourth-generation ray-tracing (RT) Cores as a foundation. These Tensor cores drive DLSS 4 frame generation technology to improve performance at higher resolutions by using AI algorithms crunched on by the Tensor cores to write additional frames, increasing frame rates. Now, DLSS 4 can generate up to three frames per rendered frame, Nvidia said in a press release.
DLSS 4 is a suite of AI-based graphics optimization technologies, including “the graphics industry’s first real-time application of the transformer model architecture,” according to Nvidia. This transformer technology enables DLSS Ray Reconstruction and Super Resolution models to use two times more parameters and four times more compute “to provide greater stability, reduced ghosting, higher details, and enhanced anti-aliasing in game scenes.”
Recommended by Our Editors
(Credit: Nvidia)
Blackwell brings even more AI-based enhancements to graphics processing, like RTX Neural Shaders. This feature introduces AI algorithms into programmable shaders, enabling film-grade assets and enhanced ray-traced lighting in real-time in 3D games using fewer resources.
Finally, the most relevant to gaming and creation is RTX Neural Faces. This feature uses generative AI to produce stable, high-fidelity 3D faces in real-time with just rasterized faces and 3D pose data as input. This reduces rendering errors and increases the consistency of 3D character models in motion within games and 3D applications.
We’re excited to test out this next generation of Nvidia mobile graphics, especially with supposed RTX 4090 performance for just $1,299 to start. Return to PCMag in March for the first fully tested Nvidia GeForce RTX 50-series gaming laptop reviews.
Get Our Best Stories!
This newsletter may contain advertising, deals, or affiliate links.
By clicking the button, you confirm you are 16+ and agree to our
Terms of Use and
Privacy Policy.
You may unsubscribe from the newsletters at any time.
About Joe Osborne
Deputy Managing Editor, Hardware
