Nvidia has announced a new flagship GPU as part of its recently announced RTX 4000 series, following weeks of rumors suggesting that Nvidia would announce a successor to its nearly two-year-old RTX 3000 series.
Using Nvidia’s new Ada Lovelace architecture, which employs 3rd generation DLSS, the first GPU announced as part of the RTX 40 series is the GeForce RTX 4090, which packs 24GB of G6X memory and is claimed to be between two and four times faster than the RTX 3090 Ti, Nvidia’s previous flagship GPU. The RTX 4090 will cost $1,599 (without a price confirmation in our territory) and will go on sale on October 12.
The RTX 4080 was also announced and included two memory configurations: 12GB and 16GB. The RTX 4080 is two to four times faster than the RTX 3080 Ti. Although no release date has been given for the RTX 4080 GPUs, Nvidia states that the cards will go on sale sometime in November. The RTX 4080 with 12GB of G6X memory will retail for $899, while the 16GB will cost $1,199.
During the presentation, Nvidia CEO Jensen Huang went into detail about the new graphics architecture of the RTX 40. Huang notes that the Ada architecture introduces the third generation of RTX 3, which has new features such as real-time trajectory tracing and the third generation of Deep Learning Super Sampling DLSS. Reportedly, DLSS 3 can generate additional frames and increase game performance by up to two times.