AMD Unveils Radeon RX 7800 XT and Radeon RX 7700 XT Graphics Cards
AMD has announced the launch of the Radeon RX 7800 XT and Radeon RX 7700 XT graphics cards at Gamescom. These cards are designed to deliver high-performance gaming at 1440p with ray tracing capabilities, and they are set to compete with NVIDIA's GeForce RTX 4070 series in terms of both performance and pricing. Built on AMD's latest RDNA3 graphics architecture and utilizing the 5 nm foundry process, these cards not only offer superior performance compared to their NVIDIA counterparts, but also provide better future-proofing with increased video memory.
The key component of both cards is the new "Navi 32" GPU, which is AMD's second largest chip in this generation. Similar to the "Navi 31" chip that powers the RX 7900 series, the "Navi 32" is a chiplet GPU, although slightly scaled down. The graphics compute die (GCD), responsible for graphics rendering and compute tasks, is manufactured using the 5 nm EUV foundry node. It is accompanied by four memory cache dies (MCDs), each built on the 6 nm foundry node. These MCDs are the same as those found in the "Navi 31," but the "Navi 32" features four MCDs instead of six, resulting in a 256-bit wide GDDR6 memory interface.
The Radeon RX 7800 XT fully utilizes the "Navi 32" chip, enabling all 60 compute units (CU) for a total of 3,840 stream processors, 120 AI accelerators, 60 Ray accelerators, 64 MB of Infinity Cache, and 16 GB of GDDR6 memory across the chip's full 256-bit memory interface. The GPU operates at game clocks of 2124 MHz and boost clocks of 2430 MHz, while the memory runs at 19.5 Gbps, delivering a memory bandwidth of 624 GB/s. The card has a total board power of 263 W and comes with two 8-pin PCIe power connectors.
The Radeon RX 7700 XT, on the other hand, is based on the same "Navi 32" silicon as the RX 7800 XT and features the same reference board design. However, it has reduced memory size to 12 GB, Infinity Cache to 48 MB, and memory bus width to 192-bit. One of the four MCDs on the "Navi 32" silicon is disabled. The RX 7700 XT is configured with 54 CU, resulting in 3,456 stream processors, 108 AI accelerators, 54 Ray accelerators, and 180 TMUs. The GPU operates at higher clock speeds compared to the RX 7800 XT, with game clocks of 2171 MHz and boost clocks of 2544 MHz. The memory speed is lower at 18 Gbps, providing a bandwidth of 432 GB/s over the 192-bit memory interface. The RX 7700 XT has a total board power of 245 W and features the same dual 8-pin power input setup as the RX 7800 XT.
In terms of performance, AMD claims that the Radeon RX 7800 XT offers between 2% to 23% performance gains over the GeForce RTX 4070 in 13 out of the 19 tested games, all of which were played at 1440p with maximum settings. Additionally, AMD states that the extra 4 GB of memory on the RX 7800 XT provides better future-proofing. The Radeon RX 7700 XT, on the other hand, outperforms the GeForce RTX 4060 Ti 16 GB by 1% to 31% in 16 out of the 19 tested games.
AMD specifically chose the RTX 4070 and RTX 4060 Ti 16 GB for comparison as it intends to competitively price the RX 7800 XT and RX 7700 XT against them. The Radeon RX 7800 XT is priced at $499, while the RX 7700 XT is priced at $449. Both cards will be available for purchase starting from September 6, 2023.