Throughout 2016 it is expected that AMD and NVIDIA will not make use of HBM2 memory only, but also use of JESD232 Graphics Double Data Rate (GDDR5X memory) to form much faster but cheaper graphics cards. Today JEDEC announced the official specifications of GDDR5X memory, which revealed what we already knew, the GDDR5X memory is up to twice as fast as the current GDDR5 found in flagship graphics cards from both NVIDIA and AMD side.
GDDR5X memory Official specifications unveiled
While the GDDR5 memory today has access to 32 bits that are saturated to 7 Gbps, the GDDR5X memory comes to solve it with access 64 bits allow the graphics cards to accomplish initially speeds of 10 or 14 Gbps, which translates to double its peak performance, but that’s not everything, since its maximum potential will be in the 16 Gbps, so there will be a range of further improvement too.
“GDDR5X represents a significant leap forward for high end GPU design,” said Mian Quddus, JEDEC Board of Directors Chairman. “Its performance improvements over the prior standard will help enable the next generation of graphics and other high-performance applications.”
The GDDR5X memory being touted as a very profitable memory compared with the next generation of HBM memory (HBM2). So, both AMD and NVIDIA will use this GDDR5 memory in low-mid-rage Graphics cards and book more expensive HBM2 memory for its top-end graphics, while GDDR5X come as a high performance solution at a lower cost, thus maintaining a strong performance at a competitive price.
It is expected that graphics cards featuring GDDR5X will be more popular than HBM2, as the latter will be more focused on computers with Ultra High Resolution or multi-monitor configurations monitors, which can be translated as Graphic solutions of very high cost.