NVIDIA launches GeForce RTX 5090, the best gaming GPU, but you won't like its price
Welcome the best new GPU for gaming
3 min. read
Published on
Read our disclosure page to find out how can you help Windows Report sustain the editorial team. Read more
NVIDIA released their new lineup of GeForce RTX 50 discrete GPUs, which includes RTX 5070, RTX 5070 Ti, RTX 5080 and the new gaming king RTX 5090. The manufacturer also disclosed the prices for the new graphics. The RTX 5070 is the most accessible, $549, followed by RTX 5070 Ti, which will go for $749. RTX 5080’s price is $999, while the top performance RTX 5090 GPU will set you down for $1,999. Here’s a breakdown of the features and specifications of their top graphics card, the GeForce RTX 5090.
NVIDIA integrated neural networks into the rendering process
The top feature of the new GeForce RTX-50 series is what NVIDIA calls neural rendering:
We’ve integrated neural networks inside of programmable shaders to create neural shaders. RTX Neural Shaders will drive the next decade of graphics innovations. They can be used to compress textures by up to 7X, saving massive amounts of graphics memory. And be used to create cinematic-quality textures, and even more advanced lighting effects in games.
RTX Neural Faces offers an innovative, new approach to improve face quality using generative AI. Instead of traditional rendering, Neural Faces takes a simple rasterized face plus 3D pose data as input and uses a real-time generative AI model to infer a more natural face.
As you’ve probably seen in the demonstrative clip above, the new neural shaders bring unbelievable results in terms of graphic detail while sticking to 32 GB of video RAM memory. Here are a few of the technical specifications provided by NVIDIA.
Architecture | GB202 |
GPU Shaders (ALUs) | 21760 |
Tensor Cores | 680 |
Ray Tracing Cores | 170 |
Boost Clock (MHz) | 2407 |
VRAM Speed (Gbps) | 28 |
VRAM (GB) | 32 |
VRAM Bus Width | 512 |
Bandwidth (GB/s) | 1792 |
TBP (watts) | 575 |
Price | $1,999 |
First, NVIDIA says that thanks to the Blackwell architecture’s innovations and DLSS 4, the GeForce RTX 5090 has double performance over GeForce RTX 4090. To demonstrate the new capabilities, NVIDIA ran Cyberpunk 2077: Phantom Liberty reaching a max 244 FPS, while the same instance with RTX 4090 featuring less than half FPS numbers.
Long story, short, GeForce RTX 5090 in the best possible discrete GPU you can get, featuring 32GB of GDDR7 memory, 1792 GB/sec of total memory bandwidth, 21,760 CUDA Cores, 680 5th Generation Tensor, and 170 4th Generation Ray Tracing Cores.
NVIDIA also provided a relative performance chart comparison between RTX 5090 and RTX 4090 in several games. However, in terms of power consumption, RTX 5090 will also put a strain on your PSU with average top 575 W.
As you would expect, you will need 2 slots to fit this 304mm long x 137mm high GPU. NVIDIA also mentioned that the card is cooled using a revolutionary new double flow through design.
According to the press release, we can expect the first laptops with the new GTX 50 series of GPUs:
GeForce RTX 5090, GeForce RTX 5080, and GeForce RTX 5070 Ti laptops will be available starting in March, followed by GeForce RTX 5070 Laptops in April. There will be designs from the world’s top manufacturers, including Acer, ASUS, Dell, GIGABYTE, HP, Lenovo, MECHREVO, MSI, and Razer.
Also, GeForce RTX 5090 will be available starting January 30 for a hefty $1,999. It would be interesting to pair the new GPU with the latest top CPU for gaming from AMD. However, you could also chose the top tier CPU from Intel, also launched at CES 2025.
What do you think about the latest gaming GPU from NVIDIA? Let’s talk about it in the comments below.
User forum
0 messages