The GeForce RTX 40-series graphics cards are finally here, and at last night's GTC 2022 "GeForce Beyond" special event, NVIDIA announced the GeForce RTX 4090 as well as the GeForce RTX 4080 in 12GB and 16GB versions.
Among them, the GeForce RTX 4090 GPU is the flagship product of the new GeForce RTX 40 series, with 76 billion transistors, 16384 CUDA cores and 24GB of high-speed Micron GDDR6X memory, which can continuously run at more than 100FPS in 4K resolution games. In a full ray traced game, the RTX 4090 with DLSS 3 offers up to 4x the performance boost compared to the previous generation RTX 3090 Ti with DLSS 2.
In addition, NVIDIA also announced the launch of the RTX 4080 in two configurations. Among them, the RTX 4080 16GB has 9728 CUDA cores and 16GB of high-speed Micron GDDR6X memory, and the performance in modern games can reach 2 times that of the GeForce RTX 3080 Ti; even at lower power, the performance is more powerful than the GeForce RTX 3090 Ti. The RTX 4080 12GB has 7680 CUDA cores and 12GB of Micron GDDR6X memory, and its performance is more powerful than the previous generation flagship GPU RTX 3090 Ti.
The power consumption of these three graphics cards is 450W, 320W and 285W respectively, and the system power requirements are also 850W, 750W and 700W respectively. Among them, the RTX 4090 will be launched on October 12, with a suggested retail price of $1,599. The RTX 4080 in two configurations will be available in November, with a suggested retail price of $1,199 and ¥899, respectively.
But what is interesting is that before NVIDIA released the GeForce RTX 40 series, AMD executives announced that they will launch the RDNA 3 architecture GPU, which is the long-awaited Radeon RX 7000 series graphics card, on November 3.
According to AMD, the RDNA 3 architecture will use a 5nm process and a small chip design. It is its first consumer-grade GPU in an MCM multi-chip package, and the performance per watt is more than 50% higher than that of the RDNA 2 architecture. Among them, Navi 31 and Navi 32 will use MCM multi-chip packages, equipped with one GCD (graphics computing chip) and four or six MCD (multi-cache I/O chips), and will use two different processes at the same time, the former is TSMC's 5nm, the latter is 6nm. Equipped with the optimized next-generation Infinity Cache, it not only provides higher density and lower power consumption, but also improves effective bandwidth and hit rate.
Previously, it was rumored that the frequency of the Radeon RX 7000 series graphics card is close to 4GHz, which is higher than the frequency of about 3GHz achieved by the current Radeon RX 6000 series graphics card. For the GPU, such a frequency is quite amazing, but the specific content has to wait for the official conference, let us look forward to it.