With six months to go before we can get our hands on the new NVIDIA and AMD gaming graphics cards, some are already claiming that the NVIDIA RTX 40 won’t be better than the AMD RX 7000. So it would all point to a generation where the Radeons would surpass the GeForces in performance… What is the truth in these assertions and what is the source?
It’s no secret that NVIDIA’s next generation of graphics cards will once again take on AMD’s rivals for consumer interest in the PC gaming market. Today, NVIDIA is in a comfortable position due to its huge market share advantage. Which means Lisa Su’s company has to conduct balloon-inflating marketing even before launch. And let’s not forget that despite their increasingly high-performance graphics architectures, they still have shortcomings in certain aspects compared to NVIDIA solutions.
It is therefore a challenge for AMD to offer a better gaming graphics architecture than its competitors. Although what is particularly important is the so-called flagship or Halo, the top of the range that will remain in people’s minds as the most powerful. And AMD seems very convinced that the RTX 40 is worse than its RX 7000. The advantage they hold? The use of chiplets or much smaller chips for creating the GPU, the same idea as the Ryzen desktop but brought to the world of graphics cards.
Will the AMD RX 7000 be better than the NVIDIA RTX 40?
We have turned into a question an assertion that has been appearing on the net in recent days and whose origin is the YouTube channel MLID, which is very close to AMD and is in charge of filtering the things that happen to it from the different companies, but in general its bias tends towards Lisa Su’s company, especially when the rumors that appear are far removed from the launch of the final product. NVIDIA, on the other hand, is much more tight-lipped about its future products, often saving things for its own presentations.
And what are the claims we are talking about? MLID puts statements in the mouths of AMD engineers such as the following:
Currently, we estimate the best NVIDIA can do (on RTX 40) is 60-80% better than Ampere (RTX 30). This is less than our estimates for the RDNA 3 cap.
“If they get serious, maybe NVIDIA can double the amp performance to some really insane power levels.”
The information we know at the moment is that we will have models of the RTX 4090, or whatever they decide to call it, that will take advantage of the 600W power consumption PCIe Gen 5 slot. We highly doubt AMD will take advantage not about the new connector’s ability to achieve higher clock speeds. In any case, these rumors of the RTX 40 and RX 7000 do not draw the complete scenario. And let’s not forget that AMD has homework to do in two areas: having a real alternative to NVIDIA’s DLSS and catching up on its performance in Ray Tracing. So to say that the RX 7000 will be better than the RTX 40 right now is propaganda.