Nvidia unveiled a pair of powerful graphics cards at its virtual GTC 2022 conference, promising dramatic performance and energy efficiency improvements with artificial intelligence-powered visuals.
The Nvidia GeForce RTX 4090 and RTX 4080 are built on the company’s new Ada Lovelace architecture, which Nvidia Chief Executive Jensen Huang touted as a “quantum leap” for gaming.
The two graphics cards have 76 billion transistors, which deliver more performance in a smaller size. The cards also support an enhanced version of the company’s Deep Learning Super Sampling technology, which employs AI to improve the realism and performance of videogames.
During the launch, Huang stated that the RTX 4090 will outperform Nvidia’s current high-end GPU flagship (RTX 3090 Ti) by up to four times.
According to Nvidia, the GeForce RTX 4090 has 24 gigabytes of GDDR6X memory and 16,384 CUDA cores, up from 10,496 in the RTX 3090. It also has a 2.23 gigahertz base clock speed and can overclock to 2.52 gigahertz.
Furthermore, the model features 1,321 Tensor-TFLOPS, 191 Ray Tracing-TFLOPS, and 83 Shader-TFLOPS. This implies that the RTX 4090 can reliably support 4K resolution games at speeds of more than 100 frames per second.
Meanwhile, the RTX 4080 will be offered with two distinct memory configurations, which will most certainly influence the overall performance. The first variant includes 12GB GGR6X while the second variant includes 16GB GDDR6X with significantly more power
Nvidia claims that they will be able to show higher-fidelity visuals with more accurate lighting and more power efficiency. Among the other enhancements are new hardware features such as real-time route tracing and a more advanced version of ray tracing that better replicates the natural path of light.