Interesting, over the weekend new slides appeared on the web originating from Micron - regarding a new memory technology. As you guys all know, HBM memory that AMD right now uses on their Fiji GPUs is the next big thing. Future HBM2 and higher revisions are however costly and complicated to fab to get large amounts of memory onto that die. It looks like the intermediate solution could be GDDR5X.
GDDR5X is your standard GDDR5 memory however, opposed to delivering 32 byte/access to the memory cells, this is doubled up towards 64 byte/access. And that in theory could double up graphics card memory bandwith. Early indications according to the presentation show numbers with the memory capable of doing up-to 10 to 12 Gbps, and in the future 16 Gbps. So your high-end graphics cardsthese days hover at say 400 GB/s. With GDDR5X that could increase to 800~1000 GB/sec and thus these are very significant improvements, actually they are competitive enough with HBM.
The big advantage of 64 byte/access GDDR5X is that it doesn't need hefty design changes other it would need compatible memory controllers, so this might be a very cost efficient methodology until HBM has matured and will be a more affordable solution. This info is derived from mydrivers.com who also states that both AMD and NVIDIA are working on GDDR5X support. We have no doubt that HBM2 will still make an entry on their enthusiast SKUs, but for the lower stacked SKUs, this memory type could be an excellent solution. See the screenshots for the presentation data.
Anybody else as well thinks this memory type actually makes more sense compared to HBM ? The one uncommon denominator whoever will be the power consumption of this memory type.