AMD talks RDNA 3 efficiencies just ahead of the RTX 4090 reveal
An AMD blog post went up yesterday to remind potential customers of the RDNA 3 GPUs set to launch November 3(opens in new tab), right as Nvidia gets ready to reveal its anticipated GeForce 40-series GPUs(opens in new tab). In particular, the post talks of «energy prices skyrocketing,» and the need for more energy efficient graphics cards with the same level of graphical prowess.
The post specifically calls out Nvidia's 30-series in that «there is a cost savings benefit as well, as seen when comparing the thermal design power across the current lineup of AMD Radeon RX 6000 Series graphics cards(opens in new tab) versus NVIDIA GeForce RTX 3000 Series products.»
It's true performance per watt has been a major focus for AMD's graphics card lineups, with its 5000-series cards improving 50% over the previous generation of GCN architecture, and the 6000-series upping that by another 65%.
It's worth noting that in our tests, we saw Nvidia's GeForce RTX 3080 ran at 0.32 frames per jule at 4K, and 0.44 frames per jule at 1080p, whereas the RX 6800 XT was 0.22 at 4k and 0.45 at 1080p.
Last month, the company confirmed a chiplet-based RDNA 3 GPU architecture(opens in new tab) for its upcoming RDNA 3 design, which alongside adaptive power management technology that the company has been refining, we can expect the performance per watt numbers to be pretty darn impressive.
It was during the design process for RDNA 2 that AMD began implementing such tactics. «We leveraged the dense CPU L3 memories to implement AMD Infinity Cache,» the post reminds us. That helped them create «a high-density, low-power cache, to make frequently used data in gaming workloads more easily accessible, dramatically increasing bandwidth while reducing the
Read more on pcgamer.com
