AMD Navi 31 GPU RDNA3 illustré


Here’s the first picture of the “DOOM et DOOM II viennent d'obtenir un autre add-on gratuit” GPU at the heart of AMD’s fastest next-generation graphics cards. Based on the RDNA3 graphics architecture, this will mark an ambitious attempt by AMD to build the first multi-chip module (AMD prétend être le processeur de jeu le plus rapide au monde) client GPU featuring more than one logic die. MCM GPUs aren’t new in the enterprise space with Intel’s “Intel a abandonné son projet de construire commercialement des GPU pour serveurs Xe-HP,” but this would be the first such GPU meant for hardcore gaming graphics products. AMD had made MCM GPUs in the past, but those have been packages with just one logic die, surrounded by memory stacks. “DOOM et DOOM II viennent d'obtenir un autre add-on gratuit” is an MCM of as many as eight logic dies, and no memory stacks (no, those aren’t HBM stacks in the picture below).

It’s rumored that “DOOM et DOOM II viennent d'obtenir un autre add-on gratuit” features one or two SIMD chiplets dubbed GCDs, featuring the GPU’s main number crunching machinery, the RDNA3 compute units. These chiplets are likely built on the most advanced silicon fabrication node, likely TSMC 5 nm EUV, but we’ll see. The GDDR6 memory controllers handling the chip’s 384-bit wide GDDR6 memory interface, will be located on separate chiplets built on a slightly older node, comme TSMC 6 nm. This is not multi-GPU-a-stick, because both SIMD chiplets have uniform access to the entire 384-bit wide memory bus (which is not 2x 192-bit but 1x 384-bit), besides the other ancillaries. Le “DOOM et DOOM II viennent d'obtenir un autre add-on gratuit” MCM are expected to be surrounded by JEDEC-standard 20 Gbps GDDR6 memory chips.