Nvidia confirmed the existence of the RTX 40-series of graphics cards during its GTC 2022 keynote, putting an end to months of speculation. The RTX 40-series is NVidia’s follow-up to the RTX 30-series, which launched in 2020. It makes use of the company’s new Ada Lovelace graphics architecture.
Less than a month from now, NVIDIA will release the RTX 40-series, so before then, let’s have a look at the specifications, examine the most notable features, and compare the first GPUs in the RTX 40 family to the RTX 3090 Ti and the RTX 3080 Ti.
RTX 40 vs. RTX 3090 Ti and 3080 Ti: Brief Overview
There will be significant improvements from the RTX 30 series to the RTX 40 series, as shown in the above infographic. Even though the RTX 40-series uses GDDR6X memory, the RTX 3090 and 3080 Ti don’t have drastically different specifications.
The RTX 4080 versions, according to Nvidia, will have performance that is two to four times that of the RTX 3080 Ti. It is reported that the RTX 4090 Ti is two to four times as fast as the RTX 3090 Ti.
Aside from sharing a similarity with Nvidia’s supersampling technology, DLSS, they also have the ability to use ray tracing in hardware. While DLSS is supported by the RTX 30-series, there are several distinctions that we’ll go through below.
Interestingly, the RTX 4080 comes in two different memory configurations, with the higher-end model sporting 16 GB of GDDR6X memory for $1,199 and the lower-end model sporting 12 GB of GDDR6X memory retailing for $899. This wouldn’t be the first time Nvidia has done something like this; the company issued two versions of its RTX 3080 graphics cards, the first with 10GB of memory and the second with 12GB of memory, roughly two years apart.
The two RTX 4080 models are distinct from one another in many ways than just memory size (and price). It takes at least 750W to run the 16GB version due to its larger number of RT, Tensor, and CUDA cores. The 12GB variant, on the other hand, has a faster base and boost clock and needs at least 700W of power to operate.
RTX 40 vs. RTX 3090 Ti and 3080 Ti: Architecture
Nvidia’s graphical architecture Ampere, built on Samsung’s 8nm technology, underpins the RTX 30-series. Meanwhile, TSMC’s 4nm N4 technology is used to construct the new Ada Lovelace architecture that powers the RTX 40 series.
The Ampere architecture is RTX’s second generation, and it employs Tensor and RT cores developed in the third generation. Third-generation RTX utilizes fourth-generation Tensor cores and third-generation RT cores and is aptly titled “Ada Lovelace” after the mathematician and author for whom it is named. Every Lovelace RT and Tensor core is more powerful than every Ampere RT and Tensor core, as you would expect from a new generation of microarchitecture.
Additionally, the latest iteration of DLSS is only supported by the Lovelace architecture, making it the only graphical architecture of its kind. According to a blog post by Nvidia, DLSS 3 is “driven by the new fourth-generation Tensor Cores and Optical Flow Accelerator,” both of which are unique to the RTX 40-series.
This paper comparison is useful, but until we can benchmark the RTX 40-series ourselves, we won’t know how the GPUs really compare. Nonetheless, there are encouraging signals that there will be a significant improvement in performance over the previous generation of GeForce graphics cards, thanks to the improved manufacturing process and graphical architecture.
read more:-
