Micron collaborated with NVIDIA and brought forward GDDR5X memory last year, having a 2.5 GHz QDR clock i.e. “10Gbps per pin” achieving record bandwidth per pin.
Micron and NVIDIA went 10% up and reached 11Gbps,With GTX 1080 Ti and TITAN Xp, with some lucky owners able to reach 3GHz QDR i.e. “12Gbps per pin”. All that is going to change with SK Hynix launching the GDDR6 memory.
GDDR6 is an upcoming JEDEC standard that runs “twice as fast” as GDDR5, offering a 10% lower operation voltage. Doubling the bandwidth while lowering the power consumption by 10% should keep improving the performance of graphics cards that don’t have the pricing structure to support the use of expensive and rarely available HBM2 memory.
AMD is indeed working on GDDR6 memory technology but they will most likely be sticking with HBM technology in high end graphics cards in 2018. They may or may not be rolling out any GDDR6 based cards in 2018.
New memory like HBM and GDDR6 are designed for such new chip structures, while GDDR5 memory has been designed for older GPUs made using older manufacturing technologies that don’t use stacked chips.
GDDR6 is expected to offer up to 16Gbps of bandwidth per pin, doubling the bandwidth available via GDDR5. Faster memory will help the GPUs process graphics faster, and the graphics can then be sent to memory.