AMD has shared a new slide for its upcoming RDNA 3 architecture GPUs which revealed some new, official, details about it. According to the red team, RDNA3 will offer more than 50% performance per watt over RDNA 2.
AMD also stated that RDNA 3 will be using next-generation AMD Infinity Cache tech, and will be manufactured at 5nm.
AMD RDNA 3 gaming architecture combines a chiplet design, next generation AMD Infinity Cache technology, leading-edge 5nm manufacturing technology, and other enhancements to deliver more than 50% greater performance-per-watt compared to the prior generation. https://t.co/SA2hQl2cio
— AMD Radeon (@amdradeon) June 9, 2022
Do note that this does NOT mean that RDNA 3 will be 50% faster than RDNA 2. Things don’t work that way so temper your expectations. Nevertheless, this is interesting as it’s rare witnessing a GPU manufacturer sharing slides out of the blue. After all, both AMD and NVIDIA usually share details/slides during their special presentations. Yesterday was AMD’s Financial Analyst Day so yeah, this new info did not come out of the blue.
It will be interesting to see whether the new AMD GPUs will be able to compete with NVIDIA’s upcoming graphics cards, especially in Ray Tracing.
Stay tuned for more!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
Pretty dam impressive if true.
Most likely partially true.
AMD cultists can’t stop violently drooling over and gobbling up their lies
And the 4070 is said to be 15% faster than 3090
Next-gen 4050Ti needs to be faster than the 3090Ti for Lovelace to matter for the next-gen in gaming.
CPuzzy2077, etc. right now just destroys our 3090Ti w/RT enabled 4k @~28-ishFPS (we’ve just stopped trying at 8k) so, if the 4000 series 4090Ti is only ~double the performance boost over the 3090Ti, well, it is already DOA for next-gen gaming for a fluid minimum 4k@60p experience w/RT enabled.
I wonder if DLSS becomes more efficient with more tensor cores.
I hope, soon, that NVIDIA construct better use with its ML hardware rather than just for upsampling because other temporal techniques – FSR2.0, TSR, TAAU, etc. are making NVIDIA look stupid right now by not needing the extra hardware to get *close* (sometimes, they best DLSS depending on the application) to its DLSS performance.
Yeah. Those company claims can’t to be taken very seriously.
AMD says RDNA 3 is projected to provide more than a 50% uplift in performance per watt. But as usual, that has to be taken with a large dose of skepticism, as performance per watt is a curve with a lot of possibilities. AMD’s RDNA 2 architecture was also supposedly a 50% improvement in performance per watt, and Nvidia claimed its Ampere architecture delivered 50% more performance per watt than its previous Turing architecture.
While both claims were certainly true in some cases, many comparisons showed far slimmer gains.
The point is, AMD’s claims of 50% higher performance per watt are a best-case scenario.
It could mean RDNA 3 is 50% faster while using the same power as RDNA 2, or it could mean it’s the same performance while using 33% less power. Generally speaking, actual products are likely to land between those two extremes, and it’s a safe bet that not every comparison will show a 50% improvement in performance per watt.
I’m curious to know how AMD will use the chiplet-based approach for these GPUs. Though, AMD also stated that it has re-architected the compute unit with RDNA3 to increase its IPC. The graphics pipeline is bound to get certain major changes, too.
We will find out soon about the performance when these cards land by the end of this year.
RDNA3 features.
This should be these companies slogan: “We love to lie”.
In other news I love how they can’t sell GPUs anymore but still don’t want to lower their prices. After they pit miners against gamers and vice versa for 70% profit margins. It’s sad that the Nvidia people have no idea how much profit Nvidia makes off of them. Well it’s 70%. And then here comes scummy AMD thinking they can achieve the same with nothing near the brand recognition of their even scummier competitor, NVIDIA.
The difference perf per watt between RDNA1 and 2 was huge no matter how you look at it. 5700XT 10TF has 225W, 6600XT 10TF 160W.
nvidia prepping 900w gpus in retaliation;
intel drooling in the corner so far.
amd is great for non rt stuff.