Nvidia Introduces RTX 4000 Gaming GPUs With Up to 400% Faster Performance

 

RTX 4090

Last night, Nvidia unveiled its RTX 4000 graphics cards, the first of which will go on sale as early as October 12. Only the RTX 4080 and 4090 are available right now; later this year, their less expensive siblings will be released.

Nvidia's top-tier discrete GPU for pros and gamers is now the RTX 4090. As implied by the name, it is the RTX 3090's replacement, however, Nvidia claims it is 2X faster than the RTX 3090 Ti and up to 4X faster when using DLSS. It is a GPU that is absolutely enormous, with 16,384 CUDA cores and boost rates that can reach 2.52GHz. A 384-bit memory interface and 24 GB of GDDR6X video memory are included.

Given that your computer uses 450W of power, you will require a backup power source. It uses either one 16-pin PCIe Gen 5 or three 8-pin PCIe cables to operate. Starting on October 12, it will be offered for sale for a starting price of $1599.

The current RTX 3090 will be replaced, however sales of the 3060, 3070, and 3080 graphics cards will continue. Before 2023, lower-end RTX 4000 GPUs are not anticipated to be available.

RTX 4080

For the RTX 4080, Nvidia has released two models, one with 16 GB of GDDR6X RAM and the other with 12 GB. While the latter is a significantly scaled-down counterpart, the former has 9728 CUDA cores, a 2.51GHz boost clock, and a 256-bit memory interface, while the later has 7680 CUDA cores, a 2.61GHz boost clock, and a 192-bit memory interface. According to Nvidia, the RTX 3080 Ti is 2-4 times slower than the 4080.

The power consumption of the 16 GB model is 320W, whereas the 12 GB model uses 285W. The 12 GB RTX 4080 will cost you $899, while the 16 GB model will be available for $1199. In November, these two will be put up for sale.

A scaled-down version of Nvidia's full-size Ada Lovelace GPU developed on TSMC's 4nm technology powers both of the new graphics cards. Third-generation RT cores, fourth-generation Tensor Cores, and support for new features like DLSS 3 and AV1 encoder are all included in this GPU.

A scaled-down version of Nvidia's full-size Ada Lovelace GPU developed on TSMC's 4nm technology powers both of the new graphics cards. Third-generation RT cores, fourth-generation Tensor Cores, and support for new features like DLSS 3 and AV1 encoder are all included in this GPU.

DLSS 3

Nvidia's AI-based DLSS technology, also known as Deep Learning Super Sampling 3, increases the frame rate (FPS) of video games. Recently, the business unveiled DLSS 3, which will only be available on RTX 4000 GPUs. Instead of creating frames within the game, it employs an Optical Flow Accelerator and data from sequential frames. The in-game FPS should increase by up to 4 times as a result.

As the frame rate rises, inherent CPU restrictions on conventional upsampling and image reconstruction techniques are likewise bypassed by DLSS 3. In plain English, this technology may significantly increase frame rates in all types of games, including those that require a lot of CPU power.

The feature is already supported by more than 35 games, including Cyberpunk 2077, Microsoft Flight Simulator, and Dying Light 2.