NVIDIA is currently using various different RAM standards on their graphics cards, with their Tesla P100 graphics card using HBM2 (stop the press! In before AMD’s upcoming Radeon RX Vega), the monster GDDR5X at 11Gbps on the TITAN Xp, Titan X(P), GeForce GTX 1080 Ti, and including the new GTX 1080 11Gbps model, and then the usual GDDR5 for the rest of its graphics cards.
SK Hynix displayed the differences between GDDR5 and GDDR6, with some huge increases in speeds and bandwidth – and up to 16Gb chips, while offering less power consumption. Which is a win-win for those gamers who want to get maximum FPS and graphical settings but without having the need of a hefty power supply, who knows they may even require less power than current generation 10xx cards?
GDDR6, on the other hand, is quite a giant leap considering it offers 16Gbps of bandwidth over the 11Gbps max we have seen on GDDR5X, at around 10% lower voltage.
NVIDIA might end up using a 384-bit memory bus on their next-gen Volta GPU architecture and see somewhere in the region of 768GB/sec of memory bandwidth, potentially quelling the high 512GB/sec that AMD’s upcoming Radeon RX Vega is rumoured to be capable of. Even using a 256-bit memory bus, which is what NVIDIA will most probably use on their mid-range Volta offering, could be capable of and AMD equalling 512GB/sec.