Nvidia officially revealed their newest lineup of consumer graphics cards yesterday at CES 2025. Built on their new Blackwell architecture, the GeForce RTX 50 Series of GPUs start at just $549 for the RTX 5070.
In addition to the RTX 5070 at $549, Nvidia announced the RTX 5070 Ti at $749, RTX 5080 at $999, and the RTX 5090 at $1,999. Both the 5080 and 5090 will be out starting January 30th. The 5070 and 5070 Ti will be made available in February. Founders Editions for all of these cards will be available directly from nvidia.com and select retailers. Nvidia board partners at ASUS, Gigabyte, MSI, Zotac, and more will release their own 50 series cards at the same time. Desktop system builders such as Maingear, Origin PC, and more will begin to offer GeForce 50 Series configurations when the cards are made available.
Laptops equipped with an RTX 5090, RTX 5080, and RTX 5070 Ti will arrive in late March. RTX 5070 equipped laptops will be out starting April 2025. These laptops will arrive from Nvidia partners such as Acer, ASUS, Dell, Gigabyte, HP, Lenovo, Mechrevo, MSI, and Razer.
A quick comparison if the specs for each 50 Series card can be found below, as shared by Nvidia.
| GeForce RTX 5090 | GeForce RTX 5080 | GeForce RTX 5070 Ti | GeForce RTX 5070 | |
| GPU Engine Specs: | ||||
| NVIDIA CUDA® Cores | 21760 | 10752 | 8960 | 6144 |
| Shader Cores | Blackwell | Blackwell | Blackwell | Blackwell |
| Tensor Cores (AI) | 5th Generation 3352 AI TOPS |
5th Generation 1801 AI TOPS |
5th Generation 1406 AI TOPS |
5th Generation 988 AI TOPS |
| Ray Tracing Cores | 4th Generation 318 TFLOPS |
4th Generation 171 TFLOPS |
4th Generation 133 TFLOPS |
4th Generation 94 TFLOPS |
| Boost Clock (GHz) | 2.41 | 2.62 | 2.45 | 2.51 |
| Base Clock (GHz) | 2.01 | 2.30 | 2.30 | 2.16 |
| Memory Specs: | ||||
| Standard Memory Config | 32 GB GDDR7 | 16 GB GDDR7 | 16 GB GDDR7 | 12 GB GDDR7 |
| Memory Interface Width | 512-bit | 256-bit | 256-bit | 192-bit |
| Display Support: | ||||
| Maximum Digital Resolution (1) | 4K at 480Hz or 8K at 120Hz with DSC | 4K at 480Hz or 8K at 120Hz with DSC | 4K at 480Hz or 8K at 120Hz with DSC | 4K at 480Hz or 8K at 120Hz with DSC |
Nvidia did provide some general benchmarks comparing a few games running on the 50 Series and the 40 Series. The comparisons aren't exactly apples to apples, with differences between DLSS, ray reconstruction, and frame generation settings differing between the two generations of cards. Nvidia did not share actual framerate figures nor were pure rasterization (no DLSS Super Resolution) figures shown.
Of the comparisons offered, the closest 1 to 1 comparison would be for A Plague Tale: Requiem, which only supports DLSS 3 regardless of using the 40 Series or 50 Series.
For those interested in upgrading it is, as always, best to wait for third-party benchmark and reviews before purchasing.
The 50 Series from Nvidia will also introduce Nvidia DLSS 4, with Nvidia claiming will "increase performance by up to 8x over traditional rendering."
DLSS 4 also introduces the graphics industry’s first real-time application of the transformer model architecture. Transformer-based DLSS Ray Reconstruction and Super Resolution models use 2x more parameters and 4x more compute to provide greater stability, reduced ghosting, higher details and enhanced anti-aliasing in game scenes.
One of the bigger features revealed by Nvidia comes from something called RTX Neural Shaders.
This breakthrough technology is used for three applications: RTX Neural Texture Compression, RTX Neural Materials, and Neural Radiance Cache (NRC).
- RTX Neural Texture Compression uses AI to compress thousands of textures in less than a minute. Their neural representations are stored or accessed in real time or loaded directly into memory without further modification. The neurally compressed textures save up to 7x more VRAM or system memory than traditional block compressed textures at the same visual quality.
- RTX Neural Materials uses AI to compress complex shader code typically reserved for offline materials and built with multiple layers such as porcelain and silk. The material processing is up to 5x faster, making it possible to render film-quality assets at game-ready frame rates.
- RTX Neural Radiance Cache uses AI to learn multi-bounce indirect lighting to infer an infinite amount of bounces after the initial one to two bounces from path traced rays. This offers better path traced indirect lighting and performance versus path traced lighting without a radiance cache. NRC is now available through the RTX Global Illumination SDK, and will be available soon through RTX Remix and Portal with RTX.
The tech company also shared a look at ACE, their "suite of digital human technologies" that utilize generative AI to bring NPCs to life.
Apparently PUBG: Battlegrounds will make use of ACE-powered characters. Other Krafton projects such as InZOI and MIR5 will also utilize Nvidia ACE. Nvidia says that the PUBG companions powered by ACE will "plan and execute strategic actions, dynamically working with human players to ensure survival." It will be interesting to see how these ACE powered NPCs differ from traditional AI companions in games.
