AMD Radeon RX 7900 XT, Radeon RX 7900 XTX High-End Gaming GPUs Announced
AMD has launched the new Radeon RX 7900 XT and Radeon RX 7900 XTX, its latest high-end GPUs based on the new RDNA 3 architecture and manufactured using many simple ‘chiplets’ designed in the 5nm process. Both will be available globally from December 13. While the flagship Radeon RX 7900 XTX is priced at $999 (approx. Rs. 81,875 before tax), the Radeon RX 7900 XT placed one step below will cost $899 (approx. -Rs. .73,680). Suggested retail prices for India have not yet been announced. AMD claims up to 70 percent better performance or 54 percent higher performance per Watt than the previous generation based on the RDNA 2 design.
Competing against Nvidia’s GeForce RTX 4000 series, the new Radeon RTX 7900 XTX features 96 compute units, clock speeds of up to 2.3GHz, and 24GB of GDDR6 RAM. The total power consumption of the board is limited to 355W. The more affordable Radeon RX 7900 XT has 84 CUs, a clock speed of 2GHz, and 20GB of GDDR6 RAM. The power consumption of this model is 300W. AMD’s promotional images show two standard 8-pin PCIe power connectors for both models, as well as DisplayPort, HDMI, and USB Type-C video outputs. Partner brands will be able to customize their offerings.
Both Radeon RX 7000 series models announced so far support the DisplayPort 2.1 standard for up to 8K output at 165Hz or 4K at 480Hz. There is also hardware-level support for up to 8K 60fps AV1 encoding, and simultaneous encoding/decoding with AVC and HEVC codecs.
A traditional monolithic GPU is divided into a Graphics Compute Die (GCD), which includes the computing units that make up the graphics rendering pipeline, and multiple Memory Cache Dies (MCDs). The Radeon RX 7900 XTX will have six MCDs, each on a 64-bit wide, 384-bit bus, while the Radeon RX 7900 XT has five active MCDs on a 320-bit bus. While the GCD is built on a 5nm platform, the MCDs use a 6nm process, which allows AMD to combine and align components to maximize manufacturing resources and costs.
AMD claims 50 percent faster ray tracing than the previous generation thanks to new AI instructions and improved throughput. Compute units can now issue dual commands, new AI accelerators have been added to each CU, and different sections of the graphics pipeline can now run at different clock speeds. More details on performance, including how it stacks up against Nvidia’s GeForce RTX 4000 series offerings and new Intel Arc GPUs, will likely be announced at that time.
The company also proposes a 2.7X improvement in the bandwidth of using the next-generation Infinity Cache thanks to improved algorithms, compared to the previous generation. The company also claims to stick with GDDR6 RAM for energy efficiency.
