The Nvidia RTX 4090 is an upcoming flagship graphics card that looks set to dominate high-end gaming PCs and benchmark top 10 lists in the not too distant future. Without much news on AMD’s upcoming next-gen cards, however, the conversation comes down to how well the RTX 4090 will fare against the existing kings of the hill: the RTX 3090 and RTX 3090 Ti.
Will these last-gen cards still hold their own, or will the RTX 4090 dominate in unprecedented fashion?
Nvidia released the (almost) full specs for the RTX 4090 in its GTC keynote, debuting it alongside the RTX 4080 for the first time.
|RTX 4090||RTX 3090 Ti||RTX 3090|
|A process node||4nm TSMC||8nm Samsung||8nm Samsung|
|RT cores||144 3rd generation (rumored)||84||82|
|Tensor kernels||576 4th Generation (Rumor)||336||328|
|Basic watch||2235 MHz||1560MHz||1395 MHz|
|Increase the clock||2520 MHz||1860 MHz||1.695MHz|
|memory||24GB GDDR6X||24GB GDD6X||24GB GDDR6X|
|Memory speed||21 Gbps (rumored)||21Gbps||19.5 Gbps|
|Memory bus width||384-bit||384-bit||384-bit|
|Bandwidth||11018GBps (rumored)||1018 Gbps||936GBps|
With the next generation of GPUs, we will again see a large increase in CUDA cores, as well as additional RT and Tensor cores, helping with faster ray tracing and better DLSS. Clock speeds have also taken a big leap, which will help deliver the big performance gains Nvidia claims are possible, as will the next generation of Nvidia RT and tensor cores. We don’t have confirmed numbers for these new cores yet, but a graphic was shown during Nvidia’s GTC demo that suggests they could appear in much larger quantities as well.
The added efficiency of the new architecture of its new 4nm process is what seems to have allowed Nvidia to increase core counts and clock speeds so dramatically, while maintaining the same power requirements as the RTX 3090 Ti. That’s still a big TDP, but it’s certainly more manageable than some of the common rumors that have been floating around in recent weeks.
The memory configurations seem to remain pretty much the same as the RTX 3090 Ti, although they are still a big improvement over the RTX 3090.
Pricing and Release
The RTX 3090 was launched in September 2020, shortly after the flagship RTX 3080. It was initially priced at $1,500, but has fluctuated widely over the past two years, sometimes reaching over $3,000. As of this writing, you can buy it for around $1,100 at most retailers.
The RTX 3090 Ti was released in March 2022 with a debut price of $2000. Scarcity at the time meant it could reach $4,000 in some cases, but prices crashed shortly after and today it can be found for as little as $1,200.
The RTX 4090 first debuted at Nvidia’s GTC keynote on September 20. It will be released on October 12 with a price of $1599. That’s a big drop from the 3090 Ti’s starting price, indicating that Nvidia isn’t dramatically raising costs across the board, but it’s raising them from the price of the original RTX 3090, suggesting we might see a modest jump in card prices over this new generation of GPU.
The Nvidia RTX 4090 will undoubtedly be faster than the RTX 3090 and 3090 Ti, but the question is: by how much?
Nvidia didn’t go into great detail with first-party benchmarks at its GTC showcase, instead claiming that the Ada Lovelace GPU at the heart of the new card is capable of up to two times the overall rasterization performance and up to four times the tracking performance of beams when DLSS is enabled. He cites just a few games in these comparisons, claiming that the two-fold increase in performance can be seen in Microsoft Flight Simulator and up to a threefold increase in performance in the newly announced Portal RTX mod.
Some allegedly leaked RTX 4090 benchmarks suggest that it will easily break all the records, even when paired with a weak processor. That would make sense, but we’ll reserve our judgment for fairness how this card is good until we have had a chance to test it ourselves.
We have some data on the RTX 3090 and 3090 Ti from our earlier tests, but it mostly shows little progression from one card to the next. The RTX 4090 is likely to offer a much more dramatic performance boost.
Like the 3090 Ti and the 3090 before it, the RTX 4090 will support both ray tracing and DLSS. Also, thanks to its higher number of RT and Tensor cores and next-gen hardware, it will see a big jump in performance when either or both are enabled compared to the previous generation. Nvidia talked about its new DLSS 3 algorithm in GTC, demonstrating that it can deliver a huge performance boost when enabled. However, this will be game-dependent and require specific implementation, so like previous versions of DLSS, its impact will be limited at first.
However, the RTX 4090 will be able to use DLSS 1 and DLSS 2 far better than its predecessors, so it’s possible that this card will be able to have much higher frame rates in ray tracing and DLSS games, maybe even allows 8K gaming in some cases. This is something Nvidia teased when the RTX 3000 series first debuted, but it wasn’t viable outside of a few hand-picked games, with DLSS doing most of the heavy lifting. However, the RTX 4090 may have the power and feature set to pull it off.
The jury is out—for now
It’s still too early to tell how well these cards will stack up against each other in the real world, although so far all indications are that the RTX 4090 dominates in terms of performance, but it requires a lot of power in its turn. It’s big, expensive and power-hungry, but it’s expected to be Nvidia’s new flagship card for some time.
Watch this space for in-depth coverage of the card and what it can do once the embargo is lifted in our review.
Now all eyes are on what AMD has planned with its the upcoming RDNA 3 generation GPU.