RDNA 3 features
What is expected alongisde the release the new AMD graphics cards
Share this article...
WePC is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Prices subject to change. Learn more
The 7900 XTX and RX 7900 XT by AMD, respectively, which are based on the latest high-performance, energy-efficient AMD RDNA 3 architecture, have been announced by AMD.
The new graphics cards are the first gaming graphics cards in the world to incorporate an innovative AMD chiplet design, building on the success of the AMD “Zen”-based AMD Ryzen chiplet CPUs.
They support the most demanding games’ high framerate at 4K and more incredible resolution gaming with exceptional performance and superb energy efficiency.
The chiplet design of the AMD RDNA 3 architecture combines 5nm and 6nm process nodes, each of which is tailored for particular tasks. The ground-breaking architecture offers up to 54% greater performance per watt as compared to the AMD RDNA 2 architecture.
It has the fastest interconnect in the world, which can transfer data at up to 5.3 TB/s between the graphics and memory system chiplets. Additionally, it provides up to 24 GB of high-speed GDDR6 memory, 96 new unified compute units and second-generation AMD Infinity Cache technology.
In addition, it has second-generation raytracing technology, which offers up to 1.8X higher raytracing performance than AMD RDNA 2 architecture, and enhanced AI throughput, which enables up to 2.7X higher AI performance.
When compared to AMD RDNA 2 architecture, these features and improvements together increase performance in some games by up to 1.7X at 4K. Additionally, the new graphics cards enable DisplayPort 2.1 displays, which may produce spectacular images and ultra-high framerates at up to 4K 480Hz and 8K 165Hz refresh rates.
AMD RDNA 3 architecture offers up to 54% better performance per watt as compared to the AMD RDNA 2 architecture thanks to an enhanced chiplet design, new compute units, and second-generation AMD Infinity Cache technology.
New compute units share resources across rendering, AI, and raytracing to make the best use of each transistor and perform quicker and more effectively than the previous generation.
