After months of rumors and teasers, Nvidia has finally announced its RTX 40 series of graphics cards during its GTC 2022 keynote. Featuring the new Ada Lovelace graphics architecture, the RTX 40 series will be the successor to the RTX 30 series, which debuted in 2020.
With the RTX 40 series set to launch in less than a month, let’s take a look at the specs, break down the standout features, and see. how the early RTX 40 family GPUs compare to the RTX 3090 Ti and RTX 3080 Ti.
RTX 40 vs. RTX RTX 3090 Ti and 3080 Ti: overall comparison
As indicated in the infographic above, there are many changes from the RTX 30 series to the RTX 40 series. However, the specifications do not differ at all Ti variants of the RTX 3090 and 3080, and all RTX 40 series use GDDR6X memory.
Nvidia claims that variants RTX 4080 will provide two to four times the performance of the RTX 3080 Ti. For its part, the RTX 4090 Ti is between two and four times faster than the RTX 3090 Ti.
Other similarities worth noting are that the GPUs support hardware ray tracing, in addition to Nvidia’s super-sampling technology, DLSS. Although the RTX 30 series supports DLSS, there are some differences that we will discuss in the next section.
Another interesting note is that there are two variants of the RTX 4080: one with 16GB of GDDR6X memory, which will be sold for $1,199, and another with 12GB of GDDR6X memory that will cost $899 (waiting for price confirmation in our territory). It wouldn’t be the first time Nvidia has done this, as the company released two different versions of its RTX 3080 graphics cards: the launch model included 10GB of memory, while another version released nearly two years later included 12GB of memory.
But the memory (and the price) it is not the only difference between the two variants of the RTX 4080. The 16GB model has more RT, Tensor, and CUDA cores and requires at least 750W of power. Instead, the 12GB model has a higher base and boost clock and requires at least 700W of power.
RTX 40 vs. RTX RTX 3090 Ti and 3080 Ti: Architecture
The RTX 30 series is based on Nvidia’s Ampere graphics architecture, which uses Samsung’s 8nm technology. The RTX 40 series, meanwhile, is based on the new Ada Lovelace architecture, which uses TSMC’s 4nm N4 technology.
The Ampere architecture it is the second generation of RTX, which uses 3rd generation Tensor cores and 2nd generation RT cores. The Ada Lovelace architecture, named after the mathematician and writer, is the third generation of RTX, using fourth generation Tensor cores and third generation RT cores. And as you would expect in a new generation microarchitecture, each Lovelace RT and Tensor core is more powerful than each Ampere RT and Tensor core.
The Lovelace architecture is also the only graphics architecture that supports the new third generation of DLSS. Nvidia noted in a blog post that DLSS 3 is “powered by the new 4th generation Tensor Cores and Optical Flow Accelerator”, found exclusively in the RTX 40 series.
Of course, these comparisons they are only a reality on paper, and we won’t know how the GPUs actually compare until we get a chance to compare the RTX 40 series ourselves. Still, the new manufacturing process and graphics architecture show promising signs that this will probably be a nice performance jump between the two generations of GeForce graphics cards.