AMD’s Navi 31 based on the RDNA 3 architecture has been taped out
Navi 31 is the top GPU in AMD’s next-generation RDNA 3 architecture. According to Twitter user @greymon55, this chip has been released. This is an important step in the process of GPU marching to the market, but there is still a long way to go before the actual product launch.
This means that the design of Navi 31 has been completed, and the chip will be tested next. If any problems are found, they will be repaired, which still takes a lot of time. Considering that this is AMD’s first consumer GPU in the MCM multi-chip package, the project cycle may be longer to allow sufficient time for related work.
It is understood that the computing module of Navi 31 is manufactured using a 5nm process and has 240 CUs, that is 15,360 stream processors, Infinity Cache is 512MB, and the video memory bit width is still maintained at 256 bits, which is still GDDR6. It is rumored that Infinity Cache will be added to the MCD chiplet, similar to the principle of using 3D V-Cache on the Zen 3 architecture, and manufactured using a 6nm process.
On this generation of GPU, AMD will make changes to the original structure instead of the original CU computing unit and will use WGP as the main computing unit. Each computing module in the Navi 31 core will have 30 WGPs, and each WGP has 256 stream processors, which means that each computing module can provide 7680 stream processors. A Navi 31 core will have a scale of 15,360 stream processors.
Since AMD has not announced the relevant roadmap or made some technical introductions, the specific process and GPU structure have yet to be confirmed. According to news, AMD has a limited production of graphics cards based on the Navi 31 core in order to maintain a stable supply of the entire GPU product line. Navi 31 core will be used for Radeon RX 7900 series graphics cards, which will be AMD’s new flagship graphics cards to replace the existing Radeon RX 6900/6800 series. It is expected to be launched at the end of 2022.