Nvidia has officially announced its highly anticipated RTX 50-series graphics cards, headlined by the flagship RTX 5090 and RTX 5080 GPUs. The new cards, based on the Blackwell architecture, promise unprecedented performance gains and AI capabilities.
The RTX 5090, priced at $1,999, features 32GB of next-generation GDDR7 memory and 21,760 CUDA cores. According to Nvidia, it delivers twice the performance of the previous RTX 4090, demonstrated by running Cyberpunk 2077 at 238fps compared to 106fps on its predecessor. However, this powerhouse requires a robust 1000-watt power supply to handle its 575-watt power draw.
At $999, the RTX 5080 comes equipped with 16GB of GDDR7 memory and 10,752 CUDA cores. Nvidia claims it doubles the performance of the RTX 4080, while demanding a 850-watt power supply for its 360-watt power consumption.
Both cards feature a redesigned Founders Edition cooling system with dual flow-through fans and a 3D vapor chamber. They support PCIe Gen 5 and include DisplayPort 2.1b connections capable of driving displays at 8K resolution and 165Hz refresh rates.
A notable improvement is the RTX 5090's two-slot design, making it compatible with small form factor PCs – a welcome change from the bulkier RTX 4090.
The cards introduce DLSS 4, Nvidia's latest AI upscaling technology, which can generate up to three additional frames per traditional frame, potentially multiplying frame rates by up to 8x over traditional rendering. The technology also employs transformer AI models to enhance image quality and reduce visual artifacts.
Both the RTX 5090 and RTX 5080 will be available starting January 30th through Nvidia's website and various retail partners, including ASUS, GIGABYTE, MSI, and others.
These new GPUs mark a major leap in gaming graphics technology, bringing unprecedented power and AI capabilities to high-end gaming PCs and content creation workstations.