Nvidia has released its next-generation GPU powered by the company’s new Ada Lovelace architecture. Both the RTX 4090 and RTX 4080 are coming later this year and promise big performance boosts.
The RTX 4090 is Nvidia’s new flagship GPU with 24GB of GDDR6X memory, 76 billion transistors, and 16,384 CUDA cores. Nvidia claims it is capable of delivering frame rates higher than 100fps when gaming at 4K, while consuming the same 450W as the RTX 3090 Ti. Using DLSS 3 (the latest version of Nvidia’s deep learning supersampling technology), Nvidia says the RTX 4090 has four times the performance of the RTX 3090 Ti using DLSS 2. The RTX 4090 starts at $1599 and will launch on October 12th.
The RTX 4080 was released in two different configurations. The first has 16GB GDDR6X memory, 76 RT cores, and 9,728 CUDA cores, while the second has 12GB GDDR6X memory, 60 RT, and 7,680 CUDA. The 12GB version also has slower memory with a throughput of 21Gbps on a 192-bit bus, while the 16GB version has a throughput of 22Gbps and a 256-bit bus. Nvidia claims that the RTX 4080 16GB is three times faster than the previous RTX 3080 Ti when using DLSS 3, and the RTX 4080 12GB can outperform the RTX 3090 Ti while using DLSS 3 while using DLSS 3. The RTX 4080 16GB starts at $1199, while the 12GB version starts at $899. Both will launch sometime in November.
In addition to releasing new GPUs, Nvidia has also taken a deep dive into DLSS 3. This latest version of supersampling can take advantage of new hardware in the Ada Lovelace architecture to generate new frames without any information from the game’s rendering pipeline. It will be exclusive to RTX 40 series GPUs.
Nvidia also showed off Portal RTX — a reimagining of the Valve classic that revolutionizes lighting implementations and adds ray tracing support. Nvidia has done the same with Quake RTX before.
The products discussed here have been independently selected by our editors. GameSpot may share revenue if you purchase any product on our site.