Last week Samsung announced the mass production of advanced chips using its new manufacturing process to 14nm LPP (Low-Power Plus), the 2nd Generation of 14nm FinFET manufacturing process of the company. Now, Samsung announced the start of mass production of a great ally of this node, we are talking about HBM2 memory with a capacity of 4 GB, the ideal companion of AMD Polaris graphics cards without forgetting NVIDIA Pascal, which will also access to ultra-fast memory from Samsung.
Samsung starts mass production of HBM2 4GB for NVIDIA and AMD
“By mass producing next-generation HBM2 DRAM, we can contribute much more to the rapid adoption of next-generation HPC systems by global IT companies,” said Sewon Chun, senior vice president, Memory Marketing, Samsung Electronics. “Also, in using our 3D memory technology here, we can more proactively cope with the multifaceted needs of global IT, while at the same time strengthening the foundation for future growth of the DRAM market.”
This new HBM2 4GB memory is manufactured using a manufacturing process of 20nm promises high performance, energy efficiency, a restrained size and high profitability for the next generation of Graphics cards. These chips reach a bandwidth of 256 Gbps (double compared to HBM1) all being much more efficient, offering twice the bandwidth per watt consumed.
HBM2 4GB of memory is developed by stacking memory chips of 8 gigabits interconnected vertically through holes in the TSV and microbumps CPU die itself. HBM2 one 8Gb chip has more than 5000 TSV holes, 36 times more that of an 8Gb TSV DDR4 die, offering a dramatic improvement in data transmission performance compared to typical wire-bonding based packages.